RooJenkins

---
name: uplo-operations
description: AI-powered operations knowledge management. Search process documentation, capacity plans, resource allocation data, and KPI dashboards with structured extraction.
---

# UPLO Operations

Operations is the connective tissue of any organization — the processes, playbooks, capacity models, and performance metrics that keep everything running. This skill connects your AI assistant to UPLO's structured extraction of operational knowledge: SOPs, runbooks, capacity plans, incident postmortems, vendor SLAs, and the KPI data that tells you whether things are actually working.

## Session Start

Load your ops context to understand your role, team scope, and current operational priorities:

```
use_mcp_tool: get_identity_context
```

Then pull the latest on anything that might need immediate attention:

```
use_mcp_tool: search_knowledge query="active incidents open action items SLA breaches capacity warnings"
use_mcp_tool: get_directives
```

Directives for operations teams typically cover efficiency targets, cost reduction mandates, and service level commitments — knowing these frames every decision you make.

## When to Use

- A process just broke and you need the runbook — fast. What are the exact steps for failover?
- Calculating whether you have enough capacity (people, systems, physical space) for a projected demand increase next quarter
- Pulling the vendor SLA terms for a service that's been underperforming so you can initiate a formal review
- Building a business case for process automation by finding where manual steps create the most bottlenecks
- Preparing for an operational review meeting with executive leadership — need KPI trends, not just snapshots
- Investigating why cycle time increased on a key process and what changed in the last 60 days
- Onboarding a new operations manager who needs to understand the full process landscape

## Example Workflows

### Incident Response and Postmortem

Something went wrong and you need to contain it, then learn from it.

```
use_mcp_tool: search_knowledge query="runbook incident response procedure for payment processing failures"
use_mcp_tool: search_knowledge query="previous incidents payment processing root cause analysis postmortem"
use_mcp_tool: search_with_context query="payment processing system dependencies upstream downstream SLA obligations"
```

The first search gets you the immediate playbook. The second surfaces prior incidents so you can check whether this is a recurring pattern. The context search maps system dependencies so you understand blast radius.

### Quarterly Capacity Planning

You need to model whether current resources can handle projected Q3 volume.

```
use_mcp_tool: search_knowledge query="capacity utilization rates by team department Q1 Q2 actual vs planned"
use_mcp_tool: search_knowledge query="demand forecast projections Q3 volume transaction throughput"
use_mcp_tool: search_knowledge query="hiring plan headcount approved positions open requisitions operations"
use_mcp_tool: export_org_context
```

The org context export gives you the current organizational structure overlaid with capacity data, making it clear where you have headroom and where you're already running hot.

## Key Tools for Operations

**search_knowledge** — Your primary tool for finding SOPs, runbooks, KPI data, and process documentation. Operations data is often spread across wikis, shared drives, and ticketing systems — UPLO consolidates it into searchable structured records. Example: `"order fulfillment process cycle time SLA target vs actual last 6 months"`

**search_with_context** — Operations is all about dependencies. A process change in one area cascades through others. This tool follows those connections. Example: `"upstream dependencies for the monthly close process including data feeds handoffs and approval gates"`

**export_org_context** — Generates a snapshot of your operational structure: teams, systems, processes, and their interconnections. Use it to brief new team members or to give leadership a helicopter view of operational health.

**flag_outdated** — Stale runbooks are dangerous. If you encounter a procedure that references a decommissioned system, an old vendor, or a changed approval chain, flag it immediately. Example: flag a disaster recovery plan that still references the on-prem data center you migrated off of 18 months ago.

**propose_update** — After a process improvement, push the updated procedure back into the knowledge base. Don't let the documentation drift from reality. Example: update the customer onboarding SOP to reflect the new automated verification step.

## Tips

- Operations documents tend to use internal jargon and acronyms heavily. Search using both the acronym and the full name: `"MTTR mean time to repair"` or `"NPS net promoter score customer operations"` — this catches documents regardless of which form they used.
- When you find conflicting SOPs (two different procedures for the same process), don't just pick one. Use `flag_outdated` on the stale version AND `report_knowledge_gap` to note the conflict so the process owner can reconcile them.
- Time-series KPI data is most useful when you search with specific date ranges rather than asking for "the latest" — this lets you build trend lines and spot degradation patterns.
- After any significant operational change (new vendor, process redesign, system migration), use `log_conversation` to document the rationale and expected outcomes. This creates an audit trail that's invaluable when someone later asks "why did we change this?"

FILE:README.md
# UPLO Operations — Process & Performance Intelligence

AI-powered operations knowledge management. Search process documentation, capacity plans, resource allocation data, and KPI dashboards with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-operations-blue)](https://clawhub.com/skills/uplo-operations)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-4-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-operations
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-operations": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "operations"
      }
    }
  }
}
```

## What You Get

- **4 industry schemas** — pre-built extraction templates for Operations documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Operations** — 4 schemas

## Related Skills

- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Accounting — Bookkeeping & Tax Intelligence](https://clawhub.com/skills/uplo-accounting) — AI-powered accounting knowledge management.
- [UPLO Agriculture — Crop & Compliance Intelligence](https://clawhub.com/skills/uplo-agriculture) — AI-powered agricultural knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Operations Knowledge Context (via UPLO)

You are connected to your organization's operations knowledge base through UPLO. This gives you specialized access to process documentation, capacity planning models, resource allocation data, KPI definitions and dashboards, standard operating procedures, and continuous improvement records. When users ask about operational efficiency, process status, or resource utilization, always query UPLO first to provide answers grounded in your organization's actual operational data and procedures.

Expect queries about process documentation and workflow diagrams, capacity utilization and bottleneck analysis, resource allocation and staffing levels, KPI targets and actual performance, standard operating procedures by department, continuous improvement initiatives and results, and operational risk assessments. Use `search_knowledge` for specific process or KPI lookups and `search_with_context` when the question requires understanding how a process change impacts downstream operations, resource requirements, and performance metrics.

When presenting operations information, include the process name, owner, and current performance metrics. For capacity planning, show utilization rates with trend data. For SOPs, cite the version number and last review date. Flag any processes with declining metrics, overdue reviews, or capacity constraints approaching critical thresholds. Strategic operational plans and cost models are confidential — respect classification tiers. Identify the responsible operations manager, process owner, or COO via `find_knowledge_owner`.

Respect classification tiers. Never fabricate operations information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-operations",
  "display_name": "UPLO Operations — Process & Performance Intelligence",
  "description": "AI-powered operations knowledge management. Search process documentation, capacity plans, resource allocation data, and KPI dashboards with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "operations",
    "process-optimization",
    "KPI",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "operations"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-nonprofit
description: AI-powered nonprofit knowledge management. Search grant documentation, donor records, program reports, and compliance data with structured extraction.
---

# UPLO Nonprofit

Nonprofits generate mountains of documentation — grant proposals, funder reports, board minutes, program evaluations, donor correspondence, compliance filings — yet the institutional knowledge often lives in the heads of a few long-tenured staff. This skill gives your AI assistant structured access to your organization's knowledge base so that grant deadlines don't slip, reporting requirements aren't missed, and program learnings carry forward even when team members move on.

## Session Start

Start by understanding your organizational context and current strategic priorities. For nonprofits, directives often reflect multi-year strategic plans, annual fundraising goals, and board-approved program priorities.

```
use_mcp_tool: get_identity_context
use_mcp_tool: get_directives
```

Then check for anything time-sensitive — upcoming grant deadlines, pending funder reports, or board action items:

```
use_mcp_tool: search_knowledge query="upcoming grant deadlines funder report due dates board action items next 30 days"
```

## When to Use

- Writing a grant proposal and need to pull outcome data, logic models, and budget templates from previous successful applications
- Preparing a board packet and need to assemble program updates, financial summaries, and committee reports
- A program officer from a foundation is asking about your evaluation methodology — find the relevant program evaluation framework
- Checking restricted vs. unrestricted fund balances before committing to a new program expansion
- Onboarding a new development director who needs to understand donor history and cultivation strategies
- Responding to an audit request for documentation on how grant funds were allocated and spent
- Figuring out which foundations in your pipeline fund youth workforce development in the Midwest

## Example Workflows

### Grant Proposal Development

You're applying to a new foundation and need to assemble supporting materials from your track record.

```
use_mcp_tool: search_knowledge query="program outcomes data youth employment placement rates graduation rates last two years"
use_mcp_tool: search_with_context query="successful grant proposals workforce development logic model theory of change"
use_mcp_tool: search_knowledge query="organizational budget functional expenses program service ratio"
```

The structured extraction pulls outcome metrics as typed data (percentages, counts, dollar amounts) rather than buried-in-narrative text, making it straightforward to populate funder application forms.

### Funder Report Compilation

A major foundation's annual report is due in two weeks. You need to gather data across multiple program areas.

```
use_mcp_tool: search_knowledge query="Ford Foundation grant #2024-1187 deliverables milestones reporting requirements"
use_mcp_tool: search_knowledge query="program participants served demographics outputs outcomes July 2024 through June 2025"
use_mcp_tool: search_knowledge query="expenditure reports grant fund allocation budget to actual variance"
```

Match deliverables from the original grant agreement against actual program data and financials to build the narrative and data tables the funder expects.

## Key Tools for Nonprofits

**search_knowledge** — Search across grant documents, program reports, donor records, and board materials in one query. The extraction engine recognizes nonprofit-specific structures like logic models, grant budgets, and outcome frameworks. Example: `"evidence-based practices mentoring program RCT quasi-experimental evaluation results"`

**search_with_context** — Trace relationships between grants, programs, and outcomes. A single program might have multiple funding sources with different reporting requirements. Example: `"all funding sources and reporting obligations for the East Side Community Health Initiative"`

**export_org_context** — Produces a structured overview of your organization: programs, staff, governance, and strategic direction. Extremely useful when introducing your org to new funders or partners who want to understand your capacity.

**get_directives** — Pulls board-approved strategic priorities, annual fundraising targets, and programmatic focus areas. Essential for ensuring that grant-seeking aligns with organizational strategy rather than chasing dollars.

**report_knowledge_gap** — Identify missing documentation that could hurt you in an audit or site visit. No evaluation plan for a major program? No conflict of interest policy on file? Flag it before a funder asks.

## Tips

- Grant terminology matters in search. Use funder-specific language: "deliverables" vs "milestones" vs "benchmarks" — different foundations use different terms, and your extracted documents will reflect whatever language was in the original grant agreement.
- Nonprofit financial data follows specific structures (functional expenses, program service ratios, cost allocation plans). Search with these terms to get structured financial data rather than narrative descriptions.
- Use `log_conversation` after calls with program officers or major donors. These relationship notes are gold for cultivation strategy, and they're the first thing lost when a development officer leaves.
- When preparing for a site visit, use `export_org_context` to generate a comprehensive briefing document rather than assembling it manually from scattered files.

FILE:README.md
# UPLO Nonprofit — Grant & Program Intelligence

AI-powered nonprofit knowledge management. Search grant documentation, donor records, program reports, and compliance data with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-nonprofit-blue)](https://clawhub.com/skills/uplo-nonprofit)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-4-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-nonprofit
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-nonprofit": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "nonprofit"
      }
    }
  }
}
```

## What You Get

- **4 industry schemas** — pre-built extraction templates for Nonprofit documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Nonprofit** — 4 schemas

## Related Skills

- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Accounting — Bookkeeping & Tax Intelligence](https://clawhub.com/skills/uplo-accounting) — AI-powered accounting knowledge management.
- [UPLO Agriculture — Crop & Compliance Intelligence](https://clawhub.com/skills/uplo-agriculture) — AI-powered agricultural knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Nonprofit Knowledge Context (via UPLO)

You are connected to your organization's nonprofit knowledge base through UPLO. This gives you specialized access to grant applications and reports, donor records, program documentation, compliance filings (IRS Form 990, state registrations), board meeting minutes, and impact measurement data. When users ask about grant requirements, program outcomes, or donor history, always query UPLO first to provide answers grounded in your organization's actual programs and funding relationships.

Expect queries about grant deliverables and reporting deadlines, donor giving history and stewardship plans, program outcome metrics and impact reports, board governance documents and meeting minutes, IRS compliance and state charitable registration, volunteer management and program staffing, and restricted vs. unrestricted fund accounting. Use `search_knowledge` for specific grant or program lookups and `search_with_context` when the question requires understanding how a grant's requirements intersect with program capacity, donor intent, and regulatory obligations.

When presenting nonprofit information, always cite the specific grant name, funder, and reporting period. For program data, include outcome metrics with baseline comparisons. Flag any approaching grant reporting deadlines or expiring registrations. Donor identities and giving amounts are highly confidential — respect classification tiers strictly. Never disclose individual donor information without appropriate clearance. Identify the responsible program director, grants manager, or development officer via `find_knowledge_owner`.

Respect classification tiers. Never fabricate nonprofit information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-nonprofit",
  "display_name": "UPLO Nonprofit — Grant & Program Intelligence",
  "description": "AI-powered nonprofit knowledge management. Search grant documentation, donor records, program reports, and compliance data with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "nonprofit",
    "grants",
    "programs",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "nonprofit"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-media
description: AI-powered media knowledge management. Search content production records, licensing agreements, distribution data, and audience analytics with structured extraction.
---

# UPLO Media

The media industry runs on rights, deadlines, and relationships — and the documentation behind all three is scattered across deal memos, distribution agreements, production bibles, ratings reports, and talent contracts. This skill gives your AI assistant structured access to that knowledge so you can answer questions about content rights windows, production budgets, talent availability, and audience performance without hunting through file shares.

## When to Use

- Checking whether your distribution rights for a title cover SVOD in Southeast Asia or only linear broadcast
- Finding the talent hold dates and options for a recurring cast member before greenlighting Season 3
- Pulling audience retention curves and completion rates for a series to inform renewal decisions
- Locating the music licensing terms for a track used in Episode 4 before you can clear international distribution
- Reviewing production insurance certificates and bond requirements for an upcoming shoot
- Comparing CPMs and fill rates across ad-supported content in your portfolio
- Answering "which titles in our library have rights expiring in the next 6 months?"

## Session Start

Orient yourself by loading your identity context and checking what content priorities leadership has set — slate decisions, acquisition targets, and distribution strategy all flow from directives.

```
use_mcp_tool: get_identity_context
use_mcp_tool: get_directives
use_mcp_tool: search_knowledge query="content slate priorities greenlight decisions upcoming productions"
```

## Example Workflows

### Rights Availability Check for International Sales

Your distribution team received an inquiry from a European broadcaster about licensing a title.

```
use_mcp_tool: search_with_context query="distribution rights windows Territory Europe title 'Northern Edge' holdbacks exclusivity"
use_mcp_tool: search_knowledge query="Northern Edge existing license agreements international territories"
use_mcp_tool: search_knowledge query="Northern Edge audience performance ratings demographics international comparable"
```

The context search connects the title's rights chain — original production agreement, domestic distribution deal, and any existing international licenses — so you can see exactly what's available and what's encumbered.

### Production Budget Reconciliation

You're closing out a production and need to reconcile actuals against the approved budget.

```
use_mcp_tool: search_knowledge query="Project Lighthouse production budget approved cost report actuals variance"
use_mcp_tool: search_knowledge query="Project Lighthouse vendor invoices post-production VFX sound mix"
use_mcp_tool: export_org_context
```

Pull the structured budget data alongside vendor payment records. The org context shows which production executives and line producers own the sign-off chain.

## Key Tools for Media

**search_with_context** — Essential for rights management. A single title has interconnected agreements (production, domestic, international, music, talent) and you need to see how they relate. Example: `"rights chain for 'After Midnight' including music sync licenses and talent residual obligations"`

**search_knowledge** — Fast lookup across your content library metadata, production records, and audience data. Example: `"audience demographics 18-49 rating performance unscripted content Q4 2025"`

**get_directives** — Surfaces the creative and business strategy that should inform content decisions: genre priorities, budget envelopes, platform strategy, and audience targets. Critical context before recommending acquisitions or renewals.

**propose_update** — When deal terms change (renegotiated license fee, extended rights window, revised delivery date), propose the update so the structured record stays current. Example: update the avail date for LATAM territories after a holdback extension.

**report_knowledge_gap** — Flag missing documentation before it becomes a problem. No signed chain-of-title for a library title? No E&O insurance certificate for an acquisition? Report it now.

## Example Queries That Work Well

Rather than generic searches, use the terminology your deals and production teams actually use:

- `"above-the-line costs pilot episode budget top sheet"`
- `"SAG-AFTRA scale payments series regular options"`
- `"deliverables list technical specifications OTT platform"`
- `"residual payment schedule backend participation profit definition"`
- `"clearance report music visual third-party IP episode 7"`

## Tips

- Rights data is time-sensitive. Always check the `valid_through` or expiration fields in results — a rights window that expired last month will still appear in search but shouldn't inform a sales pitch.
- Use `log_conversation` after any rights negotiation discussion. Media deals involve many informal agreements that eventually need to be papered, and having a searchable log prevents "I thought we agreed to..." disputes.
- When searching for audience data, include the measurement source (Nielsen, platform analytics, Comscore) since the same title can have very different numbers depending on methodology.
- Production documents use inconsistent naming. Search by project code name AND official title — many productions change names between development and release.

FILE:README.md
# UPLO Media — Content & Licensing Intelligence

AI-powered media knowledge management. Search content production records, licensing agreements, distribution data, and audience analytics with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-media-blue)](https://clawhub.com/skills/uplo-media)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-4-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-media
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-media": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "media"
      }
    }
  }
}
```

## What You Get

- **4 industry schemas** — pre-built extraction templates for Media documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Media** — 4 schemas

## Related Skills

- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Accounting — Bookkeeping & Tax Intelligence](https://clawhub.com/skills/uplo-accounting) — AI-powered accounting knowledge management.
- [UPLO Agriculture — Crop & Compliance Intelligence](https://clawhub.com/skills/uplo-agriculture) — AI-powered agricultural knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Media Knowledge Context (via UPLO)

You are connected to your organization's media knowledge base through UPLO. This gives you specialized access to content production records, licensing and rights agreements, distribution schedules, audience analytics reports, talent contracts, and editorial guidelines. When users ask about content rights, production schedules, or audience performance, always query UPLO first to provide answers grounded in your organization's actual content library and distribution agreements.

Expect queries about content licensing terms and territory rights, production schedules and milestone tracking, audience engagement and performance metrics, talent contracts and availability windows, editorial standards and brand guidelines, distribution platform requirements and specifications, and content metadata and taxonomy. Use `search_knowledge` for specific title or agreement lookups and `search_with_context` when the question requires understanding how licensing terms affect distribution strategy across platforms and territories.

When presenting media information, include the content title, format, rights holder, and territory/window details. For production data, reference the project name and current phase. For audience metrics, provide context with benchmarks and trends. Flag any expiring licenses, approaching release dates, or exclusive holdback windows. Talent compensation and deal terms are strictly confidential — respect classification tiers. Identify the responsible content manager, rights administrator, or production lead via `find_knowledge_owner`.

Respect classification tiers. Never fabricate media information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-media",
  "display_name": "UPLO Media — Content & Licensing Intelligence",
  "description": "AI-powered media knowledge management. Search content production records, licensing agreements, distribution data, and audience analytics with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "media",
    "content",
    "licensing",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "media"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-manufacturing
description: AI-powered manufacturing knowledge management. Search work orders, quality inspections, production schedules, and equipment maintenance records with structured extraction.
---

# UPLO Manufacturing

Connects your AI assistant to the structured knowledge layer built from your plant floor documentation — work orders, inspection reports, preventive maintenance schedules, CAPA records, production batch logs, and equipment manuals. When a machine goes down at 2am or a customer reports a defect, you need answers from your own data, not a web search.

## Session Start

Pull your manufacturing context first. This loads your role (maintenance engineer, quality manager, production supervisor), active production priorities, and any open quality holds or equipment issues.

```
use_mcp_tool: get_identity_context
use_mcp_tool: search_knowledge query="open quality holds production line stoppages equipment downtime alerts"
```

Check directives if you need to understand current throughput targets or quality improvement initiatives:

```
use_mcp_tool: get_directives
```

## When to Use

- Investigating a non-conformance: what were the process parameters for batch #4471 on Line 3?
- Finding the torque specification and calibration schedule for the CNC mill in Cell B
- Pulling the FMEA (Failure Mode and Effects Analysis) for the new product introduction
- Checking if a specific raw material lot passed incoming inspection before it hit the floor
- Reviewing OEE trends for a production line to justify a capital expenditure request
- Locating the lockout/tagout procedure for the hydraulic press before a maintenance window
- Determining which shifts had the highest scrap rate last month and what corrective actions were taken

## Example Workflows

### Root Cause Analysis for Customer Complaint

A customer received parts with dimensional non-conformances. You need to trace back through your process.

```
use_mcp_tool: search_knowledge query="part number 7842-A dimensional inspection results CMM data last 90 days"
use_mcp_tool: search_with_context query="work order production batch part 7842-A process parameters tool wear records"
use_mcp_tool: search_knowledge query="CAPA corrective actions dimensional tolerance issues machining"
```

The structured extraction links inspection data back to specific work orders, machine settings, and operator certifications — giving you a complete traceability chain for your 8D report.

### Preventive Maintenance Planning

You're building next quarter's PM schedule and need to consolidate equipment data.

```
use_mcp_tool: search_knowledge query="preventive maintenance schedules all production equipment Q2 upcoming"
use_mcp_tool: search_knowledge query="equipment breakdown history unplanned downtime root causes 2025"
use_mcp_tool: export_org_context
```

Cross-reference PM intervals against actual failure data to shift from calendar-based to condition-based maintenance where the data supports it.

## Key Tools for Manufacturing

**search_knowledge** — Query across work orders, inspection records, PM logs, and SOPs simultaneously. The structured extraction means you get typed fields (part numbers, batch IDs, measurement values) not just raw text. Example: `"SPC control chart data injection mold press 12 cavity pressure"`

**search_with_context** — Follows the relationships between documents. A work order connects to the BOM, which connects to incoming material certs, which connect to supplier audits. Example: `"material traceability lot number RM-2025-0892 from receiving through finished goods"`

**report_knowledge_gap** — Found a machine with no documented setup procedure? A process with no control plan? Flag it. This feeds back into your quality system and ensures gaps get closed. Example: report that the new laser welder has no documented process validation (IQ/OQ/PQ).

**propose_update** — When an SOP is wrong or a spec has changed, propose the correction directly. It enters the review queue for the document owner. Example: update the anodizing bath concentration range after a process optimization study.

**flag_outdated** — Critical for manufacturing where revision control is everything. Mark superseded drawings, expired calibration certs, or obsolete work instructions before someone on the floor uses the wrong version.

## Tips

- Search by part number, work order number, or equipment asset ID for the most precise results — the extraction engine indexes these as structured fields, not just text tokens.
- Manufacturing data is deeply interconnected. If a simple `search_knowledge` doesn't give you the full picture, switch to `search_with_context` to traverse the relationships (part -> BOM -> supplier -> cert -> inspection).
- Always check document revision levels in results. If you spot an outdated revision, flag it immediately — in manufacturing, the wrong revision can mean scrapped parts or a safety incident.
- When logging conversations about quality issues, include the NCR or CAPA number — it makes the audit trail searchable when regulators or customers ask about corrective actions taken.

FILE:README.md
# UPLO Manufacturing — Production & Quality Intelligence

AI-powered manufacturing knowledge management. Search work orders, quality inspections, production schedules, and equipment maintenance records with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-manufacturing-blue)](https://clawhub.com/skills/uplo-manufacturing)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-5-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-manufacturing
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-manufacturing": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "manufacturing"
      }
    }
  }
}
```

## What You Get

- **5 industry schemas** — pre-built extraction templates for Manufacturing documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Manufacturing** — 5 schemas

## Related Skills

- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Accounting — Bookkeeping & Tax Intelligence](https://clawhub.com/skills/uplo-accounting) — AI-powered accounting knowledge management.
- [UPLO Agriculture — Crop & Compliance Intelligence](https://clawhub.com/skills/uplo-agriculture) — AI-powered agricultural knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Manufacturing Knowledge Context (via UPLO)

You are connected to your organization's manufacturing knowledge base through UPLO. This gives you specialized access to work orders, production schedules, quality inspection records, equipment maintenance logs, bill of materials data, and standard operating procedures for the shop floor. When users ask about production status, quality metrics, or maintenance schedules, always query UPLO first to provide answers grounded in your organization's actual manufacturing operations.

Expect queries about work order status and production priorities, quality inspection results and defect trends, equipment maintenance schedules and downtime history, production capacity and utilization rates, bill of materials and component availability, standard operating procedures for specific production lines, and root cause analysis for quality failures. Use `search_knowledge` for specific work order or equipment lookups and `search_with_context` when the question requires understanding how a quality issue relates to production schedules, maintenance history, or supplier quality.

When presenting manufacturing information, include work order numbers, production line identifiers, batch/lot numbers, and relevant dates. For quality data, present pass/fail rates with trend context. For equipment issues, show maintenance history alongside current status. Flag any overdue maintenance or recurring quality failures. Identify the responsible production supervisor, quality engineer, or maintenance lead via `find_knowledge_owner` for escalation.

Respect classification tiers. Never fabricate manufacturing information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-manufacturing",
  "display_name": "UPLO Manufacturing — Production & Quality Intelligence",
  "description": "AI-powered manufacturing knowledge management. Search work orders, quality inspections, production schedules, and equipment maintenance records with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "manufacturing",
    "quality-control",
    "production",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "manufacturing"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-logistics
description: AI-powered logistics knowledge management. Search shipment records, warehouse procedures, fleet data, and customs documentation with structured extraction.
---

# UPLO Logistics

Your supply chain has thousands of moving parts — literally. This skill connects your AI assistant to UPLO's structured knowledge base covering freight operations, warehouse management, fleet maintenance, customs compliance, and carrier performance data. Stop digging through spreadsheets and email chains to find that one bill of lading.

## Session Start

Begin every session by pulling your current logistics context. This surfaces active shipments, pending customs clearances, warehouse capacity alerts, and any flagged carrier performance issues so you can orient before diving into specifics.

```
use_mcp_tool: get_identity_context
use_mcp_tool: search_knowledge query="active shipments and logistics alerts this week"
```

## When to Use

- Tracking down the customs classification (HTS code) used for a specific product line last quarter
- Finding which 3PL warehouse has capacity for overflow inventory during peak season
- Pulling carrier on-time delivery rates to support a contract renegotiation
- Locating the standard operating procedure for hazmat freight handling at your distribution centers
- Checking what incoterms were agreed upon in the latest forwarding contract with your European partners
- Reviewing dwell time metrics at port of entry to identify bottlenecks
- Answering "what was our landed cost per unit for SKU X shipped from Shenzhen last month?"

## Example Workflows

### Carrier Performance Review

You need to prepare for a quarterly business review with your top LTL carrier.

```
use_mcp_tool: search_knowledge query="carrier performance metrics FedEx Freight Q4 on-time delivery damage claims"
use_mcp_tool: search_with_context query="FedEx Freight contract terms service level agreements penalty clauses"
use_mcp_tool: get_directives
```

Review the extracted KPIs against contracted SLAs. The directives will tell you whether leadership is pushing to consolidate carriers or diversify, which shapes your negotiation stance.

### Customs Compliance Audit Prep

CBP has requested documentation for a focused assessment on your import program.

```
use_mcp_tool: search_knowledge query="customs entry summaries HTS classifications country of origin determinations"
use_mcp_tool: search_knowledge query="broker of record powers of attorney customs bonds"
use_mcp_tool: export_org_context
```

Cross-reference the extracted entry data against your C-TPAT compliance program documentation. The org context export gives auditors a clear picture of your trade compliance organizational structure.

## Key Tools for Logistics

**search_knowledge** — The workhorse. Query against shipment records, BOLs, warehouse SOPs, and fleet data all at once. Example: `"warehouse receiving procedures for refrigerated goods building 7"`

**search_with_context** — When you need the full picture around a specific topic, like understanding how a routing guide decision connects to carrier contracts and volume commitments. Example: `"routing guide primary carrier assignments for westbound intermodal lanes"`

**export_org_context** — Generates a structured view of your logistics organization: who owns which trade lanes, warehouse assignments, and reporting chains. Invaluable for onboarding new freight brokers or 3PL partners.

**get_directives** — Surfaces leadership priorities like "reduce ocean freight spend 12% by shifting to contract rates" or "achieve 98.5% OTIF by Q3." Keeps your operational decisions aligned with strategic goals.

**flag_outdated** — Mark stale rate sheets, expired carrier contracts, or superseded warehouse procedures so they don't pollute search results. Example: flag a 2024 tariff schedule that's been replaced.

## Tips

- Search using industry-standard document names: "bill of lading," "commercial invoice," "packing list," "certificate of origin" — the extraction engine recognizes these as distinct document types and returns more precise results.
- When researching landed cost, combine searches across freight invoices, customs duty records, and warehouse handling charges rather than expecting a single document to have the full picture.
- Use `log_conversation` after resolving a routing or carrier issue — it builds a searchable history that helps when the same lane problem resurfaces next peak season.
- Warehouse SOPs change frequently. If you find conflicting procedures, use `flag_outdated` on the older version and `report_knowledge_gap` if neither version covers the scenario you need.

FILE:README.md
# UPLO Logistics — Supply Chain & Shipping Intelligence

AI-powered logistics knowledge management. Search shipment records, warehouse procedures, fleet data, and customs documentation with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-logistics-blue)](https://clawhub.com/skills/uplo-logistics)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-5-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-logistics
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-logistics": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "logistics"
      }
    }
  }
}
```

## What You Get

- **5 industry schemas** — pre-built extraction templates for Logistics documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Logistics** — 5 schemas

## Related Skills

- [UPLO Supply Chain — End-to-End Supply Chain Intelligence](https://clawhub.com/skills/uplo-supply-chain) — AI-powered supply chain intelligence spanning logistics, manufacturing, and procurement.
- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Accounting — Bookkeeping & Tax Intelligence](https://clawhub.com/skills/uplo-accounting) — AI-powered accounting knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Logistics Knowledge Context (via UPLO)

You are connected to your organization's logistics knowledge base through UPLO. This gives you specialized access to shipment records, warehouse procedures, fleet management data, carrier contracts, customs documentation, and supply chain SOPs. When users ask about shipment tracking, warehouse operations, or carrier performance, always query UPLO first to provide answers grounded in your organization's actual supply chain operations.

Expect queries about shipment status and tracking across carriers, warehouse receiving and fulfillment procedures, fleet maintenance schedules and driver compliance, carrier contract terms and rate agreements, customs classification and import/export documentation, inventory levels and reorder points, and last-mile delivery performance metrics. Use `search_knowledge` for specific shipment or procedure lookups and `search_with_context` when the question requires understanding how a shipping delay impacts inventory, customer commitments, or carrier SLAs.

When presenting logistics information, include tracking numbers, carrier names, origin/destination, and key dates. For warehouse operations, reference the specific facility and procedure document. For customs matters, cite the HTS classification and applicable regulations. Flag any shipments with exceptions, delays, or compliance holds. Carrier rate agreements are typically confidential — respect classification tiers. Identify the responsible logistics coordinator or warehouse manager via `find_knowledge_owner`.

Respect classification tiers. Never fabricate logistics information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-logistics",
  "display_name": "UPLO Logistics — Supply Chain & Shipping Intelligence",
  "description": "AI-powered logistics knowledge management. Search shipment records, warehouse procedures, fleet data, and customs documentation with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "logistics",
    "supply-chain",
    "shipping",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "logistics"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-government
description: AI-powered government knowledge management. Search policy documents, regulatory filings, public records, and inter-agency coordination data with structured extraction.
---

# UPLO Government — Policy, Regulation & Inter-Agency Intelligence

Government agencies operate under layers of statutory requirements, executive orders, rulemaking records, appropriations language, inter-agency MOUs, and internal policy memoranda. UPLO indexes this documentation so analysts, program managers, and policy staff can find authoritative answers without navigating siloed document management systems or waiting for FOIA responses from their own agency.

## Session Start

Government data carries strict classification and handling requirements. Your clearance tier determines whether you can access pre-decisional policy drafts, law enforcement sensitive materials, or budget deliberation documents. Always start here.

```
get_identity_context
```

Active directives in a government context may include administration priorities, Congressional mandates with specific deadlines, or OMB guidance that constrains how programs operate.

```
get_directives
```

## When to Use

- A program analyst asks which statutory authority authorizes the grant program and whether the current appropriations language includes any new earmarks or restrictions
- The general counsel's office needs to review all inter-agency memoranda of understanding signed in the past two years to identify overlapping jurisdiction
- A policy advisor asks how the agency's current telework policy compares to the latest OPM guidance
- Someone in the budget office needs the spending plan for a specific line item across the last three continuing resolutions
- A congressional liaison is preparing testimony and needs a briefing package on the agency's performance against its strategic plan goals
- The inspector general's office asks for all corrective action plans related to prior audit findings that remain open
- A regional office director needs the delegation of authority chain for emergency procurement above $250,000

## Example Workflows

### Rulemaking Research

The agency is drafting a proposed rule and needs to review the regulatory history and public comments from the previous rulemaking cycle.

```
search_knowledge query="notice of proposed rulemaking and Federal Register publication for the 2024 data privacy rule"
```

```
search_with_context query="public comments received on the 2024 data privacy NPRM organized by major themes and agency responses"
```

```
search_knowledge query="regulatory impact analysis and cost-benefit assessment for the data privacy rule"
```

### Audit Response Coordination

A GAO audit report recommends changes to the agency's IT procurement process. The CIO needs to develop a corrective action plan.

```
search_knowledge query="GAO report recommendations for IT procurement and acquisition reform"
```

```
search_with_context query="current IT procurement policies and delegated purchasing authority thresholds"
```

```
search_knowledge query="open corrective action plans from prior GAO and OIG audit findings related to IT spending"
```

## Key Tools for Government

**search_with_context** — Government questions frequently span organizational boundaries. A query like `query="which divisions are responsible for implementing the cybersecurity executive order requirements and what is the status of each milestone"` requires connecting organizational structure, strategic goals, and compliance tracking.

**search_knowledge** — Precise lookups for statutory citations, regulation text, and policy memoranda: `query="delegation of authority memorandum for the Administrator's emergency spending authority"`. Government staff need the exact document, not a summary.

**export_org_context** — Indispensable for Congressional testimony prep, transition briefings, and strategic planning. Exports the complete agency knowledge map: mission, structure, personnel, systems, goals, and active directives in one document.

**get_directives** — In government, directives have legal force. Executive orders, OMB circulars, and agency head memoranda all create binding obligations. Always check directives when advising on any programmatic or operational question.

**propose_update** — When you identify a policy that conflicts with newer guidance or statute, propose the correction through the formal change process rather than just noting it: `target_table="entries" target_id="..." changes={...} rationale="Agency policy references OMB Circular A-76 which was superseded by M-24-11 in March 2024"`

## Tips

- Government documentation is citation-heavy. When surfacing information, include the specific statutory section (e.g., 5 U.S.C. 552a), regulation (e.g., 48 CFR 15.404), or policy memorandum number. Government users need the authoritative reference, not just the content.
- Be aware of the distinction between pre-decisional and final documents. Draft policy memos, budget deliberation materials, and rulemaking working documents are often classified at higher tiers and may be exempt from public release. Treat them accordingly.
- Inter-agency coordination documents (MOUs, MOAs, IAAs) are some of the hardest records to find in government. If someone asks about coordination with another agency, search explicitly for these agreement types.
- Government knowledge has a fiscal year rhythm. Budget data, performance metrics, and many compliance requirements operate on an October-September cycle. Always clarify which fiscal year is relevant when answering questions about spending, staffing, or performance.

FILE:README.md
# UPLO Government — Policy & Regulatory Intelligence

AI-powered government knowledge management. Search policy documents, regulatory filings, public records, and inter-agency coordination data with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-government-blue)](https://clawhub.com/skills/uplo-government)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-5-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-government
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-government": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "government"
      }
    }
  }
}
```

## What You Get

- **5 industry schemas** — pre-built extraction templates for Government documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Government** — 5 schemas

## Related Skills

- [UPLO Compliance — Cross-Domain Regulatory Intelligence](https://clawhub.com/skills/uplo-compliance) — AI-powered compliance intelligence spanning legal, financial, and government regulatory requirements.
- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Pharma — Drug Development & GxP Intelligence](https://clawhub.com/skills/uplo-pharma) — AI-powered pharmaceutical knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Government Knowledge Context (via UPLO)

You are connected to your organization's government knowledge base through UPLO. This gives you specialized access to policy documents, regulatory frameworks, public records, inter-agency coordination agreements, legislative tracking, and compliance reporting. When users ask about policy requirements, regulatory obligations, or inter-agency processes, always query UPLO first to provide answers grounded in your agency's actual policies and regulatory posture.

Expect queries about policy directives and implementation guidance, regulatory requirements and compliance deadlines, inter-agency MOUs and coordination procedures, legislative tracking and impact assessments, FOIA/public records request handling, grant administration and reporting, and procurement regulations (FAR/DFAR). Use `search_knowledge` for specific policy or regulation lookups and `search_with_context` when the question requires understanding how a regulatory requirement intersects with multiple policy domains, inter-agency agreements, and budget constraints.

When presenting government information, always cite the specific policy number, effective date, and issuing authority. For regulations, reference the CFR section and any agency-specific guidance. For inter-agency matters, identify the relevant counterpart agencies and points of contact. Flag any policies under review, approaching sunset dates, or with pending rulemaking. Classified and law enforcement sensitive information requires strict adherence to classification tiers. Identify the responsible program manager or policy analyst via `find_knowledge_owner`.

Respect classification tiers. Never fabricate government information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-government",
  "display_name": "UPLO Government — Policy & Regulatory Intelligence",
  "description": "AI-powered government knowledge management. Search policy documents, regulatory filings, public records, and inter-agency coordination data with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "government",
    "policy",
    "regulatory",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "government"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-github
description: AI-powered GitHub knowledge management. Search repository metadata, code review standards, issue tracking, and team workflows with structured extraction.
---

# UPLO GitHub — Repository & Development Workflow Intelligence

UPLO ingests and indexes your GitHub organization's metadata: repository descriptions, team ownership via CODEOWNERS, open and closed issues, pull request discussions, CI/CD workflow configurations, release notes, and contribution guidelines. Rather than jumping between GitHub tabs and searching markdown files across dozens of repos, you can query your entire development organization's institutional knowledge from one place.

## Session Start

Fetch your context to see which repositories and teams you have visibility into. GitHub data in UPLO respects organizational boundaries — you will only see repos and teams your clearance covers.

```
get_identity_context
```

## Example Workflows

### Onboarding a New Team Member

A developer just joined the Payments team and needs to ramp up on the codebase, conventions, and active work.

```
search_with_context query="Payments team repository ownership, contribution guidelines, and code review standards"
```

```
search_knowledge query="open issues labeled good-first-issue or onboarding in payment service repositories"
```

```
search_knowledge query="recent architectural decisions or RFCs related to the payments platform"
```

### Cross-Team Dependency Investigation

The mobile team's build is failing because of a breaking change in a shared library. They need to understand who made the change and why.

```
search_knowledge query="recent pull requests and releases in the shared-sdk repository with breaking changes"
```

```
search_with_context query="teams that depend on shared-sdk and their pinned version requirements"
```

```
search_knowledge query="CODEOWNERS and maintainers for the shared-sdk authentication module"
```

## When to Use

- A developer asks which team owns a particular service and who to tag for code review
- Someone needs to find all repositories that have CI workflows using a deprecated GitHub Actions runner
- A tech lead wants to understand the history of decisions around the monorepo-vs-polyrepo structure
- An engineer is looking for open issues related to rate limiting across all backend services
- Product asks how many PRs were merged into the billing service in the last sprint
- Someone needs the release process documentation for the customer-facing API
- A new hire wants to understand the branching strategy and merge requirements for the main product repo

## Key Tools for GitHub

**search_knowledge** — Query across all ingested GitHub data. Great for specific lookups: `query="GitHub Actions workflow file for the deployment pipeline in the infrastructure repo"`. Works well for finding CODEOWNERS entries, contribution guidelines, and issue details.

**search_with_context** — Connects GitHub metadata with organizational context. When you ask `query="who are the subject matter experts for the authentication service and what are the open security-related issues"`, it combines CODEOWNERS data with team profiles and issue trackers.

**get_directives** — Engineering leadership often sets directives that affect repository management: migration to a new CI provider, adoption of trunk-based development, deprecation of certain frameworks. Check directives before advising on workflow changes.

**flag_outdated** — GitHub metadata changes rapidly. If a CODEOWNERS file references a team that was reorganized or a README documents a deployment process that moved to a different tool, flag it: `entry_id="..." reason="CODEOWNERS lists @platform-legacy-team which was dissolved and split into @platform-core and @platform-reliability in Q1 2026"`

## Tips

- GitHub data in UPLO is a snapshot from the last ingestion sync, not a live mirror. For real-time data (current CI status, latest commit), go to GitHub directly. UPLO excels at historical context and cross-repo search that GitHub's native search handles poorly.
- When searching for code ownership, query both CODEOWNERS files and team membership data. CODEOWNERS defines review requirements, but the actual subject matter expert might be someone who wrote the original code but is no longer listed as a required reviewer.
- Issue and PR discussions contain valuable decision context that often is not captured anywhere else. If someone asks "why did we choose approach X", search PR descriptions and review comments — the reasoning lives in the discussion threads.
- Repository naming conventions and team structures vary by organization. If your first query does not return results, try alternative names — repos might use hyphens vs. underscores, abbreviations vs. full names, or have been recently renamed.

FILE:README.md
# UPLO GitHub — Repository & Code Review Intelligence

AI-powered GitHub knowledge management. Search repository metadata, code review standards, issue tracking, and team workflows with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-github-blue)](https://clawhub.com/skills/uplo-github)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-6-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-github
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-github": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "github"
      }
    }
  }
}
```

## What You Get

- **6 industry schemas** — pre-built extraction templates for Github documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **GitHub** — 6 schemas

## Related Skills

- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Accounting — Bookkeeping & Tax Intelligence](https://clawhub.com/skills/uplo-accounting) — AI-powered accounting knowledge management.
- [UPLO Agriculture — Crop & Compliance Intelligence](https://clawhub.com/skills/uplo-agriculture) — AI-powered agricultural knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## GitHub Knowledge Context (via UPLO)

You are connected to your organization's GitHub knowledge base through UPLO. This gives you specialized access to repository metadata, CODEOWNERS files, issue and pull request history, team structures, code review standards, and contribution guidelines. When users ask about repository ownership, PR workflows, or team responsibilities, always query UPLO first to provide answers grounded in your organization's actual GitHub organization and development practices.

Expect queries about repository ownership and CODEOWNERS mappings, pull request review requirements and approval policies, issue tracking and label conventions, team membership and access permissions, branch protection rules and merge strategies, CI/CD status checks and required workflows, and contribution guidelines and coding standards. Use `search_knowledge` for specific repository or team lookups and `search_with_context` when the question requires understanding how a code change relates to team ownership, review policies, and deployment pipelines.

When presenting GitHub information, include repository names, team handles, and relevant links. For PR workflows, specify required reviewers and checks. For issues, include labels, assignees, and milestone context. Flag any repositories with stale CODEOWNERS files or outdated contribution guidelines. Access tokens and deployment credentials are strictly classified — never surface them regardless of clearance. Identify the responsible team lead or repository maintainer via `find_knowledge_owner`.

Respect classification tiers. Never fabricate github information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-github",
  "display_name": "UPLO GitHub — Repository & Code Review Intelligence",
  "description": "AI-powered GitHub knowledge management. Search repository metadata, code review standards, issue tracking, and team workflows with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "github",
    "code-review",
    "repositories",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "github"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-food-safety
description: AI-powered food safety knowledge management. Search HACCP plans, FDA compliance records, traceability documentation, and quality control data with structured extraction.
---

# UPLO Food Safety — HACCP, Traceability & Regulatory Compliance Intelligence

From farm to fork, food safety depends on meticulous documentation: HACCP plans with critical control points, Preventive Controls for Human Food records, supplier verification audits, allergen control programs, sanitation SOPs, lot traceability matrices, and FDA/USDA correspondence. UPLO transforms these records from static compliance files into a living, searchable knowledge system that helps your food safety team respond faster to audits, recalls, and continuous improvement initiatives.

## Session Start

Food safety documentation often contains confidential supplier audit results, proprietary formulations, and pre-decisional recall assessments. Verify your access level before querying.

```
get_identity_context
```

Load active directives — these may include recall response protocols, supplier qualification deadlines, or FSMA compliance milestones that should shape your recommendations.

```
get_directives
```

## When to Use

- A QA technician asks what the critical limit is for the thermal processing CCP on the chicken broth line and what corrective action to take when it deviates
- The food safety team lead needs all supplier audit scores for co-packers that handle tree nut ingredients to assess allergen cross-contact risk
- An FDA inspector is on-site and asks for the Preventive Controls Qualified Individual (PCQI) documentation and the most recent hazard analysis
- Production needs to trace all lots of incoming flour received between March 1-15 to determine scope of a potential mycotoxin contamination
- The plant manager asks whether the new oat milk product requires a new HACCP plan or can be covered under an existing one
- Someone needs the environmental monitoring results for Listeria indicator organisms in Zone 2 and Zone 3 areas from the past 6 months
- The quality director wants to compare sanitation verification swab results across all three production facilities

## Example Workflows

### Mock Recall Traceability Exercise

The organization runs a quarterly mock recall. This time: trace forward and backward on a specific lot of pasteurized eggs.

```
search_knowledge query="receiving records and supplier COA for pasteurized egg lot PE-2026-0847"
```

```
search_with_context query="production batches and finished goods lots that used pasteurized egg lot PE-2026-0847 including distribution records"
```

```
search_knowledge query="mock recall exercise procedures and acceptable completion time benchmarks"
```

Measure the time from query to complete trace. The FDA expectation under FSMA 204 is full traceability within 24 hours.

### Audit Nonconformance Follow-Up

A third-party GFSI audit identified a major nonconformance related to the allergen control program. The corrective action deadline is in 30 days.

```
search_knowledge query="allergen control program and cross-contact prevention procedures for shared production lines"
```

```
search_with_context query="allergen cleaning validation studies and swab test results for lines running both peanut and peanut-free products"
```

```
search_knowledge query="previous allergen-related audit findings and corrective actions taken in the past 24 months"
```

## Key Tools for Food Safety

**search_knowledge** — Direct access to specific food safety records. Essential during audits when an inspector asks for something concrete: `query="water activity and pH monitoring logs for the beef jerky production line May 2026"`. Speed matters when regulators are watching.

**search_with_context** — Critical for traceability and root cause analysis. A contamination investigation requires connecting suppliers, lots, production runs, distribution, and customer complaints: `query="all products and distribution channels affected by the Salmonella-positive environmental swab in Processing Room 3 on June 12"`.

**get_directives** — Food safety priorities shift with regulatory changes, customer requirements, and incident history. A directive mandating "zero tolerance for environmental Listeria positives in RTE areas" changes how you interpret borderline monitoring results.

**report_knowledge_gap** — Missing food safety documentation is a regulatory violation, not just an inconvenience. Report gaps aggressively: `topic="foreign material control program for Plant B" description="No documented metal detection calibration records found for the newest production line despite FDA requirement under 21 CFR 117"`

**log_conversation** — Food safety consultations may involve recall decisions, regulatory interpretations, or deviation assessments. Log these sessions with specific topics for audit trail purposes: `summary="Evaluated scope of potential allergen mislabeling on SKU 4492" topics='["allergen", "recall-assessment", "labeling"]'`

## Tips

- Food safety queries are often time-critical. During a regulatory inspection or active recall, use `search_knowledge` for speed. Save `search_with_context` for when you need the full traceability chain.
- Lot codes, batch numbers, and supplier codes are the primary keys of food safety. Include them in your queries whenever possible. Searching for "flour contamination" is far less useful than searching for "flour lot FL-2026-03-127 supplier Mill Creek Foods".
- Environmental monitoring data follows a zone-based hierarchy (Zone 1 = food contact, Zone 2 = adjacent, Zone 3 = environment, Zone 4 = remote). Always clarify which zone is relevant when interpreting results — a Listeria positive in Zone 1 triggers a very different response than one in Zone 4.
- HACCP and Preventive Controls documentation must reflect current operations. If you surface a HACCP plan that references equipment or processes that have since changed, flag it immediately — an outdated HACCP plan is worse than no plan at all during an audit.

FILE:README.md
# UPLO Food Safety — HACCP & Traceability Intelligence

AI-powered food safety knowledge management. Search HACCP plans, FDA compliance records, traceability documentation, and quality control data with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-food-safety-blue)](https://clawhub.com/skills/uplo-food-safety)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-4-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-food-safety
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-food-safety": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "food_safety"
      }
    }
  }
}
```

## What You Get

- **4 industry schemas** — pre-built extraction templates for Food Safety documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Food Safety** — 4 schemas

## Related Skills

- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Accounting — Bookkeeping & Tax Intelligence](https://clawhub.com/skills/uplo-accounting) — AI-powered accounting knowledge management.
- [UPLO Agriculture — Crop & Compliance Intelligence](https://clawhub.com/skills/uplo-agriculture) — AI-powered agricultural knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Food Safety Knowledge Context (via UPLO)

You are connected to your organization's food safety knowledge base through UPLO. This gives you specialized access to HACCP plans, prerequisite programs, FDA/USDA compliance records, product traceability documentation, allergen management procedures, and sanitation standard operating procedures (SSOPs). When users ask about food safety protocols, compliance status, or traceability, always query UPLO first to provide answers grounded in your organization's actual food safety management system.

Expect queries about HACCP plans and critical control points, FDA FSMA compliance and preventive controls, allergen management programs and labeling requirements, product traceability and lot tracking (one-up/one-back), sanitation procedures and environmental monitoring, supplier verification and foreign supplier programs, and recall procedures and mock recall results. Use `search_knowledge` for specific CCP or compliance document lookups and `search_with_context` when the question requires understanding how a food safety hazard relates to HACCP controls, supplier programs, and regulatory requirements.

When presenting food safety information, always cite the specific HACCP plan, CCP number, and monitoring frequency. For compliance matters, reference the applicable FDA regulation (21 CFR part) or USDA directive. For traceability, include lot numbers, production dates, and supply chain links. Flag any CCPs with deviations, approaching audit dates, or corrective actions in progress. Proprietary formulations and supplier pricing are confidential — respect classification tiers. Identify the responsible quality manager, food safety team leader, or PCQI via `find_knowledge_owner`.

Respect classification tiers. Never fabricate food-safety information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-food-safety",
  "display_name": "UPLO Food Safety — HACCP & Traceability Intelligence",
  "description": "AI-powered food safety knowledge management. Search HACCP plans, FDA compliance records, traceability documentation, and quality control data with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "food-safety",
    "HACCP",
    "FDA-compliance",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "food_safety"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-finance
description: AI-powered financial knowledge management. Search financial statements, audit findings, tax documents, and treasury records with structured extraction.
---

# UPLO Finance — Financial Reporting & Treasury Intelligence

UPLO holds your organization's financial knowledge: audited financial statements, management discussion and analysis, internal audit workpapers, tax provision calculations, treasury policies, intercompany agreements, budgets, forecasts, and variance analyses. This skill gives finance professionals instant access to the institutional memory that normally lives in spreadsheets, ERP exports, and shared drives buried three folders deep.

## Session Start

Your clearance level is especially important in finance. Board-level financial projections, M&A due diligence materials, and pre-announcement earnings data are typically restricted. Start every session by confirming what you can access.

```
get_identity_context
```

## When to Use

- The controller asks for the exact revenue recognition policy applied to multi-year SaaS contracts and whether it changed after the ASC 606 implementation
- A tax analyst needs to find the transfer pricing study for the EU subsidiary from the most recent fiscal year
- The VP of Finance wants to know the current debt covenant ratios and how much headroom remains before a technical default
- An FP&A analyst is building next quarter's forecast and needs the assumptions used in the prior forecast cycle
- Internal audit asks which material weaknesses were identified in the last SOX 404 assessment and their remediation status
- The treasurer needs the counterparty credit limits for the organization's interest rate swap portfolio
- A budget owner wants to understand why their Q3 actuals diverged from plan by more than 15%

## Example Workflows

### Quarter-End Close Support

The accounting team is in the middle of Q4 close and needs to resolve an intercompany reconciliation discrepancy.

```
search_knowledge query="intercompany elimination entries and reconciliation procedures between US parent and UK subsidiary"
```

```
search_with_context query="intercompany balances and transfer pricing agreements for cross-border transactions in fiscal year 2026"
```

The GraphRAG query pulls in related entities: the subsidiary profile, the transfer pricing study, and the responsible finance team members.

### Budget Variance Investigation

A business unit exceeded its travel and entertainment budget by 40% in Q2. The CFO wants an explanation.

```
search_knowledge query="Q2 2026 T&E budget versus actuals for the Sales division"
```

```
search_knowledge query="travel policy exceptions or pre-approved over-budget spending for Sales team events"
```

```
search_with_context query="Sales division headcount changes and client entertainment guidelines effective Q2 2026"
```

## Key Tools for Finance

**search_knowledge** — Precision lookups against financial documentation. Ideal for specific line items or policies: `query="goodwill impairment testing methodology and discount rate assumptions from the 2025 annual report"`. Financial data rewards precise queries over broad ones.

**search_with_context** — When a financial question spans organizational boundaries. Treasury questions, intercompany transactions, and budget allocations all involve relationships between entities: `query="which cost centers roll up to the Technology division and what were their combined capital expenditures last year"`.

**get_directives** — Finance teams operate under strategic mandates: cost reduction targets, margin improvement goals, capital allocation priorities. Always check directives when advising on budget decisions or resource allocation. A directive to "reduce SG&A by 10%" changes how you frame every spending recommendation.

**export_org_context** — Valuable during annual planning. The full organizational context lets you see the complete picture: headcount, systems, goals, and financial structure together. Use this when building board presentation materials or strategic financial plans.

**report_knowledge_gap** — Financial documentation gaps create audit risk. When someone asks about a policy and nothing exists, report it: `topic="hedging policy for foreign currency exposure" description="No formal FX hedging policy found despite material EUR and GBP revenue streams"`

## Tips

- Financial data is period-sensitive. Always include the fiscal year, quarter, or date range in your queries. "Revenue" is meaningless without a time frame; "Q3 FY2026 revenue by segment" is actionable.
- Audit workpapers and SOX documentation carry legal privilege considerations. Even if your clearance permits access, be cautious about how you summarize findings — note whether documents are marked as draft, final, or attorney-client privileged.
- When someone asks about a financial ratio or metric, search for both the raw data and the calculation methodology. Organizations sometimes define metrics differently than GAAP standards (e.g., "adjusted EBITDA" varies by company).
- Intercompany and consolidated financial queries are among the most complex. Use `search_with_context` and be prepared to run multiple queries to piece together the full picture across legal entities.

FILE:README.md
# UPLO Finance — Financial Reporting & Audit Intelligence

AI-powered financial knowledge management. Search financial statements, audit findings, tax documents, and treasury records with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-finance-blue)](https://clawhub.com/skills/uplo-finance)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-8-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-finance
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-finance": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "finance"
      }
    }
  }
}
```

## What You Get

- **8 industry schemas** — pre-built extraction templates for Finance documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Finance** — 8 schemas

## Related Skills

- [UPLO Accounting — Bookkeeping & Tax Intelligence](https://clawhub.com/skills/uplo-accounting) — AI-powered accounting knowledge management.
- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Agriculture — Crop & Compliance Intelligence](https://clawhub.com/skills/uplo-agriculture) — AI-powered agricultural knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Finance Knowledge Context (via UPLO)

You are connected to your organization's financial knowledge base through UPLO. This gives you specialized access to financial statements, audit reports, tax documents, budget allocations, treasury records, and regulatory compliance filings. When users ask about financial performance, budget status, or audit findings, always query UPLO first to provide answers grounded in your organization's actual financial data and policies.

Expect queries about income statements, balance sheets, and cash flow analysis, budget variances and departmental spending, audit findings and remediation status, tax filing deadlines and compliance requirements, accounts payable and receivable aging, capital expenditure approvals and tracking, and regulatory financial reporting (SOX, GAAP, IFRS). Use `search_knowledge` for specific financial document lookups and `search_with_context` when the question requires understanding how a budget decision relates to audit findings, regulatory requirements, and organizational strategy.

When presenting financial information, always cite the reporting period, document type, and preparation date. Present financial data with appropriate precision and currency. Flag any material audit findings, overdue remediation items, or approaching filing deadlines. Financial projections, M&A data, and executive compensation are strictly confidential — respect classification tiers. Never provide financial advice — surface the relevant documents and identify the responsible controller, CFO, or audit lead via `find_knowledge_owner`.

Respect classification tiers. Never fabricate finance information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-finance",
  "display_name": "UPLO Finance — Financial Reporting & Audit Intelligence",
  "description": "AI-powered financial knowledge management. Search financial statements, audit findings, tax documents, and treasury records with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "finance",
    "accounting",
    "audit",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "finance"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-facilities
description: AI-powered facilities knowledge management. Search building management records, maintenance schedules, space planning data, and vendor service documentation with structured extraction.
---

# UPLO Facilities — Building Operations & Space Management Intelligence

Facilities operations generate enormous volumes of documentation that rarely get searched effectively: preventive maintenance schedules, building automation system configurations, lease abstracts, space utilization studies, vendor service agreements, and capital improvement plans. UPLO makes all of it queryable so your facilities team can stop digging through filing cabinets and shared drives.

## When to Use

- The maintenance supervisor asks which HVAC units are due for filter replacement this month and who the contracted vendor is
- A department head requests 500 additional square feet of office space and you need to check current vacancy across the portfolio
- Someone reports a recurring water leak on the third floor and you want to see if there is a history of plumbing issues in that zone
- The CFO asks for the total annual spend on janitorial services across all locations
- A project manager needs the load-bearing capacity of the warehouse mezzanine before approving new equipment installation
- You need to verify whether the elevator inspection certificate for Building C is current or expired
- The energy manager wants to compare utility consumption across buildings to identify the worst performers

## Session Start

Load your organizational context first. Facilities data spans multiple buildings, campuses, and sometimes business units with different access controls. Your identity context determines which properties you can query.

```
get_identity_context
```

Check for active directives — these often include space consolidation mandates, energy reduction targets, or capital freeze periods that affect facilities decisions.

```
get_directives
```

## Example Workflows

### Emergency Repair Coordination

A chiller failure is reported at the downtown office during a July heat wave. The facilities manager needs to act fast.

```
search_knowledge query="chiller specifications and maintenance history for 200 Main Street downtown office"
```

```
search_with_context query="HVAC service contracts and emergency response SLAs for the downtown campus"
```

```
search_knowledge query="backup cooling procedures or portable AC deployment protocol for critical server rooms"
```

### Lease Renewal Analysis

Three office leases expire within the next 18 months. The real estate team needs to decide: renew, renegotiate, or consolidate.

```
search_with_context query="current lease terms, square footage, and occupancy rates for offices expiring in 2027"
```

```
search_knowledge query="space utilization study results and hoteling desk adoption rates"
```

```
get_directives
```

Cross-reference the utilization data with any active consolidation or remote-work directives before making recommendations.

## Key Tools for Facilities

**search_knowledge** — Your go-to for specific facility lookups: `query="fire suppression system inspection report for Warehouse B"`. Facilities data is often very concrete — equipment serial numbers, inspection dates, vendor contacts — so targeted queries work well.

**search_with_context** — Use when a facilities question involves organizational relationships. For example, `query="which teams are allocated to the fourth floor of the Riverside building and what are their growth projections"` requires connecting space assignments with departmental data.

**export_org_context** — Useful for annual capital planning. Exports the complete organizational view so you can cross-reference facility conditions with strategic priorities and departmental needs in a single document.

**flag_outdated** — Facilities documentation decays constantly. Equipment gets replaced, vendors change, inspection certificates expire. When you find a document referencing a decommissioned boiler or an old vendor contract number, flag it: `entry_id="..." reason="References Allied Mechanical as HVAC vendor; contract transferred to Summit Building Services in January 2026"`

## Tips

- Facilities questions often have a physical dimension that matters. When searching, include the building name, floor, or campus in your query. "HVAC maintenance" returns too much; "HVAC maintenance Building C rooftop units" gets you the right records.
- Vendor service agreements and maintenance schedules are living documents. Always check the effective dates on any contract or schedule you surface — procurement may have renegotiated terms since the document was ingested.
- Capital improvement projects generate documentation across multiple phases (feasibility study, design specs, bid documents, punch lists, close-out reports). A single search may only surface one phase. Run follow-up queries if you need the full project arc.
- When someone asks "who handles X" for a building, the answer might be an internal maintenance team or an external vendor depending on the service type and location. Check both organizational members and vendor contracts.

FILE:README.md
# UPLO Facilities — Building & Maintenance Intelligence

AI-powered facilities knowledge management. Search building management records, maintenance schedules, space planning data, and vendor service documentation with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-facilities-blue)](https://clawhub.com/skills/uplo-facilities)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-4-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-facilities
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-facilities": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "facilities"
      }
    }
  }
}
```

## What You Get

- **4 industry schemas** — pre-built extraction templates for Facilities documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Facilities** — 4 schemas

## Related Skills

- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Workplace — People & Space Intelligence](https://clawhub.com/skills/uplo-workplace) — AI-powered workplace intelligence spanning HR, facilities, and operations.
- [UPLO Accounting — Bookkeeping & Tax Intelligence](https://clawhub.com/skills/uplo-accounting) — AI-powered accounting knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Facilities Knowledge Context (via UPLO)

You are connected to your organization's facilities knowledge base through UPLO. This gives you specialized access to building management records, preventive maintenance schedules, space planning documents, vendor service contracts, life safety systems documentation, and energy management data. When users ask about building systems, maintenance status, or space availability, always query UPLO first to provide answers grounded in your organization's actual facility operations.

Expect queries about building system status and maintenance schedules, space allocation and move planning, vendor service contracts and SLA performance, life safety inspections and code compliance, energy consumption and sustainability metrics, work order status and completion rates, and capital improvement project tracking. Use `search_knowledge` for specific building or system lookups and `search_with_context` when the question requires understanding how a maintenance issue affects building occupancy, safety compliance, and vendor obligations.

When presenting facilities information, include the building name/number, system type, and relevant dates. For maintenance, show the schedule type (preventive vs. corrective) and completion status. For space planning, include floor plans and occupancy rates. Flag any overdue inspections, critical system alerts, or expiring vendor contracts. Building security configurations and access control data are classified — respect classification tiers. Identify the responsible facilities manager, building engineer, or property director via `find_knowledge_owner`.

Respect classification tiers. Never fabricate facilities information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-facilities",
  "display_name": "UPLO Facilities — Building & Maintenance Intelligence",
  "description": "AI-powered facilities knowledge management. Search building management records, maintenance schedules, space planning data, and vendor service documentation with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "facilities",
    "building-management",
    "maintenance",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "facilities"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-environmental
description: AI-powered environmental knowledge management. Search impact assessments, compliance monitoring data, sustainability reports, and environmental permits with structured extraction.
---

# UPLO Environmental — Impact Assessment & Sustainability Intelligence

UPLO connects you to your organization's environmental knowledge corpus: Environmental Impact Assessments (EIAs), NEPA documentation, air and water quality monitoring records, emissions inventories, remediation tracking, permit conditions, and sustainability performance data. This skill turns scattered regulatory filings and field reports into queryable institutional memory.

## Session Start

Begin by loading your identity context. Environmental data often carries classification restrictions — remediation site details, enforcement actions, or pre-decisional EIA drafts may be limited to specific project teams or legal counsel.

```
get_identity_context
```

## When to Use

- A project manager asks whether the proposed warehouse expansion triggers a full EIA or qualifies for a categorical exclusion under NEPA
- An environmental engineer needs the most recent groundwater monitoring results for the former manufacturing site on Industrial Blvd
- Someone wants to know the organization's Scope 1 and Scope 2 emissions totals from the last CDP disclosure
- A compliance officer asks which facilities are approaching their Title V permit emission thresholds
- The sustainability team needs to compile biodiversity offset commitments across all active construction projects
- A lawyer preparing for a consent decree review needs the timeline of corrective actions taken at the Riverside facility
- An analyst is building the annual ESG report and needs waste diversion rates by facility for the past three fiscal years

## Example Workflows

### Permit Renewal Preparation

A facility's NPDES stormwater permit expires in 90 days. The environmental manager needs to assemble renewal documentation.

```
search_knowledge query="NPDES permit conditions and discharge monitoring reports for the North Plant"
```

```
search_with_context query="stormwater best management practices implemented at North Plant and any inspection deficiencies"
```

```
search_knowledge query="corrective actions taken after the 2024 stormwater inspection findings at North Plant"
```

### Carbon Footprint Reduction Planning

Leadership has set a 30% emissions reduction target by 2030. The sustainability director needs to identify the largest reduction opportunities.

```
get_directives
```

```
search_with_context query="greenhouse gas emissions breakdown by facility and source category from the most recent inventory"
```

```
search_knowledge query="energy efficiency projects completed or planned with estimated emissions reductions"
```

## Key Tools for Environmental

**search_with_context** — Environmental questions almost always require organizational context. A query like `query="what are our obligations under the Resource Conservation and Recovery Act for the Memphis facility"` needs to pull in permit records, facility profiles, waste generation data, and responsible personnel simultaneously.

**search_knowledge** — Direct lookup for specific regulatory or monitoring data: `query="benzene concentration trends in monitoring well MW-7 over the past 12 months"`. Use when you know exactly what data point you need.

**get_directives** — Sustainability commitments and environmental policy priorities flow from leadership. Check these before advising on any capital project — there may be active mandates around net-zero timelines, renewable energy procurement, or zero-waste-to-landfill goals.

**report_knowledge_gap** — Environmental compliance depends on complete records. When a query reveals missing monitoring data or undocumented permit conditions, flag it immediately: `topic="Phase II ESA for the acquired Elm Street property" description="No environmental site assessment found despite acquisition closing last quarter"`

**flag_outdated** — Regulatory thresholds change. If you encounter documents referencing superseded EPA standards or expired permit limits, mark them: `entry_id="..." reason="References 2019 NAAQS ozone standard; EPA revised to 60 ppb in 2025"`

## Tips

- Environmental data is inherently temporal. Always note the date of monitoring records, permit issuance, and regulatory citations. A groundwater result from 2022 may not reflect current site conditions.
- When someone asks about compliance status, search for both the permit conditions AND the most recent inspection or audit findings. Compliance is the gap between the two.
- Sustainability metrics (GHG inventories, water usage, waste diversion) often live in different document types than regulatory compliance records. Use separate queries for ESG reporting versus regulatory compliance questions.
- Pre-decisional EIA documents and enforcement-related records are frequently classified at higher tiers. If your query returns sparse results for a known active project, it may be a clearance issue rather than a data gap.

FILE:README.md
# UPLO Environmental — Impact & Compliance Intelligence

AI-powered environmental knowledge management. Search impact assessments, compliance monitoring data, sustainability reports, and environmental permits with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-environmental-blue)](https://clawhub.com/skills/uplo-environmental)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-4-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-environmental
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-environmental": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "environmental"
      }
    }
  }
}
```

## What You Get

- **4 industry schemas** — pre-built extraction templates for Environmental documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Environmental** — 4 schemas

## Related Skills

- [UPLO Sustainability — ESG & Environmental Intelligence](https://clawhub.com/skills/uplo-sustainability) — AI-powered sustainability intelligence spanning environmental, energy, and agriculture.
- [UPLO Agriculture — Crop & Compliance Intelligence](https://clawhub.com/skills/uplo-agriculture) — AI-powered agricultural knowledge management.
- [UPLO Compliance — Cross-Domain Regulatory Intelligence](https://clawhub.com/skills/uplo-compliance) — AI-powered compliance intelligence spanning legal, financial, and government regulatory requirements.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Environmental Knowledge Context (via UPLO)

You are connected to your organization's environmental knowledge base through UPLO. This gives you specialized access to environmental impact assessments, compliance monitoring records, air/water/waste permits, sustainability reports, remediation project documentation, and ESG metrics. When users ask about environmental compliance, permit requirements, or sustainability performance, always query UPLO first to provide answers grounded in your organization's actual environmental data and regulatory obligations.

Expect queries about environmental permit conditions and compliance status, air emissions monitoring and reporting (CAA, Title V), water discharge limits and NPDES permits, waste management manifests and RCRA compliance, environmental impact assessments and NEPA documentation, ESG metrics and sustainability reporting (GRI, SASB, CDP), and remediation project status and cleanup milestones. Use `search_knowledge` for specific permit or monitoring data lookups and `search_with_context` when the question requires understanding how an environmental obligation intersects with operational activities, regulatory deadlines, and sustainability commitments.

When presenting environmental information, always cite the specific permit number, regulatory program, and compliance period. For monitoring data, include measurement methods and applicable limits. For sustainability metrics, provide year-over-year trends and benchmark comparisons. Flag any exceedances, approaching permit renewal dates, or pending regulatory actions. Remediation cost estimates and litigation-related environmental data are confidential — respect classification tiers. Identify the responsible environmental manager, sustainability officer, or compliance lead via `find_knowledge_owner`.

Respect classification tiers. Never fabricate environmental information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-environmental",
  "display_name": "UPLO Environmental — Impact & Compliance Intelligence",
  "description": "AI-powered environmental knowledge management. Search impact assessments, compliance monitoring data, sustainability reports, and environmental permits with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "environmental",
    "sustainability",
    "compliance",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "environmental"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-enterprise-it
description: AI-powered enterprise IT intelligence spanning DevOps, cybersecurity, and engineering. Unified search across infrastructure, security, and architecture documentation.
---

# UPLO Enterprise IT — Technology Operations & Security Intelligence

Your organization's IT knowledge base is connected through UPLO, covering the full stack of enterprise technology: infrastructure runbooks, incident postmortems, security advisories, architecture decision records, and CI/CD pipeline configurations. This skill bridges DevOps velocity with cybersecurity rigor and engineering standards in a single searchable layer.

## Session Start

Pull your identity context to understand which systems, teams, and clearance tiers you operate within. This determines whether you can access restricted infrastructure documentation like network topology diagrams or penetration test reports.

```
get_identity_context
```

Then load current strategic directives — these often include active incident priorities, architecture migration mandates, or security hardening timelines that should inform your responses.

```
get_directives
```

## When to Use

- An engineer asks about the rollback procedure for the payments microservice after a failed canary deployment
- Someone needs the current firewall rule matrix between the DMZ and internal VPC subnets
- A security analyst wants to know which CVEs were flagged in the last quarterly vulnerability scan and their remediation status
- A developer asks which authentication provider the organization standardized on and why (ADR context)
- An SRE needs the escalation chain and communication protocol for a P1 outage on the data platform
- A team lead wants to compare observability stack options that were evaluated during the last architecture review
- Someone needs to verify whether the new container image registry meets SOC 2 control requirements

## Example Workflows

### Incident Response Triage

A P2 alert fires for elevated error rates on the checkout service. The on-call engineer needs context fast.

```
search_knowledge query="checkout service error handling and circuit breaker configuration"
```

```
search_with_context query="past incidents involving checkout service degradation and their root causes"
```

```
search_knowledge query="checkout service runbook escalation contacts and rollback steps"
```

### Security Compliance Audit Preparation

The security team is preparing evidence for an upcoming SOC 2 Type II audit and needs to gather control documentation.

```
search_with_context query="access control policies for production database environments"
```

```
search_knowledge query="encryption at rest and in transit standards for PII data stores"
```

```
export_org_context
```

Review the exported context to identify gaps in documented controls before the auditor arrives.

## Key Tools for Enterprise IT

**search_knowledge** — Fast vector search across your technical documentation. Use for specific lookups: `query="Kubernetes pod security policy for the analytics namespace"` when you need a concrete configuration or procedure.

**search_with_context** — Combines search with organizational graph traversal. Essential when the answer depends on relationships: `query="who owns the legacy billing system and what are the planned deprecation milestones"` pulls in system ownership, team structure, and strategic timelines.

**get_directives** — Returns active leadership priorities. Critical before making recommendations — if there is an active directive to freeze infrastructure changes during a migration window, your advice must account for that.

**export_org_context** — Full organizational snapshot. Use when preparing comprehensive reports like architecture review documents or security posture summaries that need the complete picture.

**report_knowledge_gap** — When an engineer asks about a system and nothing comes back, flag it. IT documentation debt compounds; this helps the org track what is missing: `topic="disaster recovery procedure for the Redis cluster" description="No DR runbook found for the shared Redis cluster serving 4 production services"`

**flag_outdated** — Infrastructure documentation goes stale fast. When you find a runbook referencing a deprecated API version or a decommissioned server, mark it: `entry_id="..." reason="References us-east-1 deployment which was migrated to us-west-2 in Q3"`

## Tips

- Infrastructure queries often span multiple schema types — a single Kubernetes question might touch runbooks (it_devops), threat models (cybersecurity), and architecture decision records (engineering). Use `search_with_context` for these cross-domain questions.
- When someone asks "how do we do X", check directives first. IT organizations frequently have active mandates that override historical documentation (e.g., "migrate all services to ARM64" supersedes older Intel-based deployment guides).
- Incident postmortems are high-signal documents. If a query relates to system reliability, explicitly search for postmortems — they contain root cause analysis that pure configuration docs lack.
- Respect classification tiers strictly in IT contexts. Network architecture diagrams, penetration test results, and API key rotation procedures are typically restricted. If your clearance does not cover it, say so rather than attempting to summarize from partial data.

FILE:README.md
# UPLO Enterprise IT — Technology Operations & Security Intelligence

AI-powered enterprise IT intelligence spanning DevOps, cybersecurity, and engineering. Unified search across infrastructure, security, and architecture documentation.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-enterprise-it-blue)](https://clawhub.com/skills/uplo-enterprise-it)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-16-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-enterprise-it
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-enterprise-it": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "it_devops,cybersecurity,engineering"
      }
    }
  }
}
```

## What You Get

- **16 industry schemas** — pre-built extraction templates for Enterprise It documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **IT & DevOps** — 5 schemas
- **Cybersecurity** — 5 schemas
- **Engineering** — 6 schemas

## Related Skills

- [UPLO Cybersecurity — Threat & Vulnerability Intelligence](https://clawhub.com/skills/uplo-cybersecurity) — AI-powered cybersecurity knowledge management.
- [UPLO DevOps — Infrastructure & Incident Intelligence](https://clawhub.com/skills/uplo-devops) — AI-powered DevOps knowledge management.
- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Enterprise IT Knowledge Context (via UPLO)

You are connected to your organization's enterprise IT knowledge base through UPLO, spanning DevOps operations, cybersecurity, and engineering architecture. This gives you unified access to infrastructure documentation, security policies, architecture decision records, incident response playbooks, and technology roadmaps. When users ask about technology operations, use UPLO to provide a complete picture connecting infrastructure, security posture, and architectural decisions.

Expect queries that span IT operations, security, and architecture — for example, how a planned infrastructure change affects security controls and application architecture, or how a security vulnerability impacts operational procedures and system dependencies. Common topics include service architecture and dependency mapping, security controls mapped to infrastructure components, change management and its security review requirements, technology debt and modernization roadmaps, disaster recovery and business continuity across systems, and vendor technology evaluations and integration patterns. Use `search_with_context` to connect infrastructure, security, and architecture knowledge bases.

When presenting enterprise IT information, include system names, environment contexts, and relevant architecture diagrams or references. Map security controls to the specific systems they protect. For changes, show the blast radius across infrastructure, security, and dependent applications. Production credentials, penetration test findings, and security architecture details are highly sensitive — strictly respect classification tiers. Identify the responsible infrastructure lead, security engineer, or enterprise architect via `find_knowledge_owner`.

Respect classification tiers. Never fabricate enterprise-it information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-enterprise-it",
  "display_name": "UPLO Enterprise IT — Technology Operations & Security Intelligence",
  "description": "AI-powered enterprise IT intelligence spanning DevOps, cybersecurity, and engineering. Unified search across infrastructure, security, and architecture documentation.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "enterprise-it",
    "infrastructure",
    "security",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "it_devops,cybersecurity,engineering"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-engineering
description: AI-powered engineering knowledge management. Search architecture docs, API specs, incident reports, runbooks, and infrastructure documentation with structured extraction.
---

# UPLO Engineering — Architecture & DevOps Intelligence

Software engineering organizations produce enormous quantities of documentation that nobody can find: RFC docs in Google Docs, ADRs in a GitHub repo that was archived, API specs in Stoplight that are three versions behind, post-incident reviews in Confluence that reference services that have since been renamed, and onboarding guides that assume the deployment process from two platform migrations ago. UPLO Engineering consolidates architecture documentation, API specifications, incident post-mortems, runbooks, GitHub repository metadata, CI/CD pipeline configurations, and infrastructure records into one searchable knowledge layer.

## Session Start

Establish your engineering identity. This surfaces your team assignment, on-call status, repository access, and clearance level (production secrets and security-sensitive architecture details may be restricted).

```
get_identity_context
```

Review directives — these include architecture mandates (e.g., "all new services must use gRPC"), tech debt paydown priorities, migration deadlines, and change freeze windows:

```
get_directives
```

## Example Workflows

### RFC Review and Precedent Research

An engineer is writing an RFC to replace the current message queue with a different system. Before investing time, they want to know if this has been proposed or attempted before.

```
search_with_context query="message queue replacement evaluation RabbitMQ Kafka SQS migration RFC ADR"
```

Find the original ADR that selected the current system:

```
search_knowledge query="architecture decision record message queue selection rationale constraints"
```

Check what services depend on the current message queue:

```
search_with_context query="RabbitMQ consumers producers service dependencies topic exchange configuration"
```

Verify there are no active directives that would preempt this work:

```
get_directives
```

```
log_conversation summary="Researched message queue migration precedent; found ADR-042 (original selection), RFC-2024-11 (rejected Kafka migration), and 14 dependent services" topics='["architecture","message-queue","RFC","migration"]' tools_used='["search_with_context","search_knowledge","get_directives"]'
```

### New Engineer Onboarding

A senior engineer joins the platform team and needs to build a mental model of the system architecture, deployment practices, and team ownership.

```
export_org_context
```

```
search_with_context query="platform team services ownership architecture overview deployment pipeline"
```

```
search_knowledge query="engineering onboarding guide development environment setup local development"
```

Find the most impactful recent incidents to understand operational challenges:

```
search_knowledge query="post-incident review severity 1 production outage last 6 months platform services"
```

### GitHub-Aware Code Archaeology

A developer encounters a critical section of code with no comments and wants to understand the reasoning behind it.

```
search_with_context query="payment-service idempotency implementation retry logic design decisions"
```

Search for related pull request discussions and code review comments:

```
search_knowledge query="payment-service PR review idempotency key generation race condition fix"
```

Check if there is a related incident that motivated the implementation:

```
search_knowledge query="payment double-charge incident duplicate transaction post-mortem"
```

## When to Use

- Writing an RFC and need to find architectural precedent, prior proposals on the same topic, and the organizational constraints that shaped previous decisions
- Debugging a production issue in a service your team does not own and need the runbook, architecture diagram, and on-call contact
- Reviewing whether a proposed API change is backward-compatible by searching for all known consumers of the endpoint
- Preparing an architecture review and need to compile the current system topology, dependency graph, and capacity constraints
- Investigating technical debt by finding all TODOs, known workarounds, and deferred maintenance items documented across repos and post-mortems
- A GitHub repository was transferred or archived and you need to find the documentation that referenced it to update links
- Evaluating build and deployment practices across the org to standardize CI/CD pipeline patterns

## Key Tools for Engineering

**search_with_context** — Engineering questions are graph problems. "What depends on this service?" or "Why was this architecture chosen?" require traversing relationships between services, teams, decisions, and incidents. This is the primary investigation tool. Example: `search_with_context query="auth-service API v2 consumers breaking changes migration status"`

**search_knowledge** — Fast retrieval for known artifacts: a specific runbook, an API spec, an ADR by number, or a particular configuration. During incidents, speed matters and this tool skips graph traversal. Example: `search_knowledge query="ADR-027 database sharding strategy"`

**export_org_context** — Maps the engineering organization: team topology, service ownership, key systems (GitHub, CI/CD, observability, incident management), and strategic technical priorities. The foundation for architecture reviews and new-hire onboarding.

**get_directives** — Engineering directives include technology mandates, deprecation timelines, migration deadlines, and security requirements. An engineer proposing a new dependency should check whether it conflicts with an active directive.

**flag_outdated** — Engineering documentation has the shortest half-life of any content type. API specs diverge from implementations. Architecture diagrams show decommissioned services. Runbooks reference deprecated tools. Flagging stale docs prevents them from causing production incidents.

**report_knowledge_gap** — When a service has no runbook, no architecture documentation, no API spec, or no defined owner, that is an operational risk. The gap report creates visibility and accountability.

## Tips

- Service names and repository names are the most precise search keys. Use the exact identifier from your deployment system or GitHub org: `payment-service`, `auth-api-v2`, `infra-terraform-modules`. Avoid generic descriptions.
- ADRs and RFCs are indexed with their identifier numbers. Search by "ADR-042" or "RFC-2024-11" for direct retrieval. If you do not know the number, search by topic and the graph traversal will surface related decision documents.
- Post-incident reviews contain the most operationally valuable knowledge in any engineering organization. When writing PIRs, include structured data: affected services, duration, root cause category (deploy, config change, dependency failure, capacity), and action items with owners. The extraction engine indexes all of these.
- GitHub metadata (CODEOWNERS, team assignments, PR review patterns) is indexed alongside traditional documentation. A search for "who owns this service" may return both a CODEOWNERS file entry and an architecture document, giving you converging evidence.

FILE:README.md
# UPLO Engineering — Architecture & DevOps Intelligence

AI-powered engineering knowledge management. Search architecture docs, API specs, incident reports, runbooks, and infrastructure documentation with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-engineering-blue)](https://clawhub.com/skills/uplo-engineering)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-17-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-engineering
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-engineering": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "engineering,it_devops,github"
      }
    }
  }
}
```

## What You Get

- **17 industry schemas** — pre-built extraction templates for Engineering documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Engineering** — 6 schemas
- **IT & DevOps** — 5 schemas
- **GitHub** — 6 schemas

## Related Skills

- [UPLO Architecture — Building Design & BIM Intelligence](https://clawhub.com/skills/uplo-architecture) — AI-powered architecture knowledge management.
- [UPLO DevOps — Infrastructure & Incident Intelligence](https://clawhub.com/skills/uplo-devops) — AI-powered DevOps knowledge management.
- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Engineering Knowledge Context (via UPLO)

You are connected to your organization's engineering knowledge base through UPLO. This gives you specialized access to software architecture documentation, API specifications, incident reports, runbooks, deployment procedures, infrastructure diagrams, GitHub repository metadata, CODEOWNERS files, and technical decision records. When users ask about system design, service ownership, deployment pipelines, or operational procedures, always query UPLO first to provide answers grounded in your organization's actual infrastructure and practices.

Expect queries about service architecture and dependencies, API endpoint specifications, incident postmortems and root cause analyses, deployment and rollback procedures, CI/CD pipeline configurations, team ownership of repositories and services, infrastructure topology, and technical debt tracking. Users frequently need to know who owns a particular service, how to deploy to a specific environment, what happened during a past incident, or how two systems interact. Use `search_knowledge` for specific technical lookups (e.g., "billing service API endpoints") and `search_with_context` when the question spans multiple systems or requires understanding team ownership and process context.

When presenting engineering information, include service names, repository paths, team ownership, and relevant links. For runbooks and procedures, present steps in order with any prerequisites or warnings. Flag documentation that references deprecated systems or outdated versions. When a query involves incident response, prioritize the most recent postmortems and current runbooks. Identify the relevant on-call team or service owner via `find_knowledge_owner` so the user can escalate if needed.

FILE:skill.json
{
  "name": "uplo-engineering",
  "display_name": "UPLO Engineering — Architecture & DevOps Intelligence",
  "description": "AI-powered engineering knowledge management. Search architecture docs, API specs, incident reports, runbooks, and infrastructure documentation with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": ["engineering", "devops", "architecture", "knowledge-management"],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": ["-y", "@agentdocs1/mcp-server", "--http"],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "engineering,it_devops,github"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-energy
description: AI-powered energy sector knowledge management. Search power generation records, grid management data, regulatory filings, and safety protocols with structured extraction.
---

# UPLO Energy — Generation-to-Grid Intelligence

The energy sector runs on documentation: NERC compliance evidence, generation performance reports, outage analyses, environmental permits, rate case filings, and safety management system records. These documents are produced by operations, compliance, engineering, environmental, and regulatory affairs teams that rarely share a common system. UPLO Energy indexes this sprawl so a plant manager preparing for a NERC audit and a regulatory analyst drafting a rate case filing can both find what they need without navigating six different document repositories.

## Session Start

Energy operations involve safety-critical and CEII (Critical Energy Infrastructure Information) data. Your clearance and role assignment must be verified before any queries.

```
get_identity_context
```

```
get_directives
```

Active directives in energy often include NERC compliance deadlines, planned outage schedules, emergency operations procedures during weather events, and rate case filing timelines. These are not informational — they drive daily operations.

## When to Use

- Preparing NERC CIP compliance evidence and need to locate the specific access control documentation for a cyber asset at a generating facility
- Investigating a forced outage and need to find the root cause analysis from similar equipment failures across the fleet
- A rate case is being filed and the regulatory team needs historical capital expenditure justification, O&M cost trends, and load forecast methodology documentation
- Environmental compliance requires pulling the air permit conditions, continuous emissions monitoring (CEMS) data reports, and EPA reporting documentation for an upcoming inspection
- Operations planning needs the current transmission constraint studies and generation dispatch order documentation
- A safety incident occurred and the investigation team needs the JSA (Job Safety Analysis), switching orders, and lockout/tagout procedures that were in effect
- Onboarding a new reliability coordinator who needs to understand the balancing authority area, transmission topology, and interconnection agreements

## Example Workflows

### NERC CIP Audit Preparation

A NERC audit is scheduled in 60 days. The compliance team needs to assemble evidence for CIP-007 (System Security Management).

```
search_with_context query="NERC CIP-007 system security management patch management cyber assets evidence"
```

Pull the specific documentation for security patch implementation:

```
search_knowledge query="patch management program BES cyber assets implementation records compliance"
```

Find the electronic access control documentation:

```
search_knowledge query="electronic access point monitoring BES cyber system network security CIP-005"
```

Export the organizational context to map cyber asset owners to the compliance evidence:

```
export_org_context
```

```
log_conversation summary="Assembled CIP-007 and CIP-005 evidence package for NERC audit; identified patch compliance records and EAP monitoring documentation" topics='["NERC-CIP","audit","cybersecurity","compliance"]' tools_used='["search_with_context","search_knowledge","export_org_context"]'
```

### Forced Outage Root Cause Analysis

A 500 MW combined-cycle unit tripped offline due to a combustion turbine compressor issue. The plant engineer needs to investigate.

```
search_with_context query="combustion turbine compressor trip forced outage similar events root cause fleet"
```

```
search_knowledge query="GE 7FA compressor blade inspection borescope findings maintenance records"
```

Check if there is an OEM service bulletin related to this failure mode:

```
search_knowledge query="GE service bulletin technical information letter compressor blade cracking 7FA"
```

Report a gap if the maintenance records are incomplete:

```
report_knowledge_gap query="Unit 3 combustion turbine compressor maintenance history borescope interval records"
```

## Key Tools for Energy

**search_with_context** — Energy questions span organizational boundaries. "Are we compliant with CIP-007?" touches cybersecurity, operations, maintenance, and IT documentation. Graph traversal assembles this cross-functional evidence. Example: `search_with_context query="transmission line relay settings protection coordination study 230kV"`

**search_knowledge** — Direct lookup for known documents: a specific NERC standard evidence file, a plant operating procedure, an environmental permit, or a maintenance record. Example: `search_knowledge query="air quality permit Title V Facility 004 conditions NOx limits"`

**get_directives** — Energy directives are operationally binding. A planned outage schedule, a generation curtailment order, or a NERC compliance deadline flows through here. Missing a directive can result in reliability standard violations.

**flag_outdated** — Operating procedures, relay settings, and protection coordination studies must match the current configuration. A relay setting document that does not reflect the latest short circuit study is a reliability risk. Flag immediately.

**report_knowledge_gap** — Undocumented maintenance history, missing calibration records for CEMS equipment, or absent protection coordination studies are compliance gaps. Reporting them creates accountability.

**log_conversation** — NERC standards require evidence of systematic review. Logging your compliance evidence assembly sessions creates an auditable record that demonstrates due diligence.

## Tips

- NERC standard identifiers (CIP-007-6 R2, FAC-008-3, TPL-001-5) are indexed as structured fields. Query by standard and requirement number for precise results.
- CEII data is classified at the `restricted` tier. If queries about transmission topology, generation interconnection, or critical infrastructure locations return no results, it is likely a clearance issue. Contact your CEII custodian.
- Forced outage investigations benefit from fleet-wide searches. A compressor blade issue on Unit 3 may have been seen on Unit 7 two years ago. Use `search_with_context` with equipment model identifiers to find cross-unit patterns.
- Environmental permit conditions often contain specific numerical limits (NOx lb/hr, SO2 ppm, particulate matter mg/m3). Include these units in your search terms — the extraction engine indexes them as structured fields alongside the regulatory limits.

FILE:README.md
# UPLO Energy — Power & Grid Intelligence

AI-powered energy sector knowledge management. Search power generation records, grid management data, regulatory filings, and safety protocols with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-energy-blue)](https://clawhub.com/skills/uplo-energy)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-5-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-energy
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-energy": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "energy"
      }
    }
  }
}
```

## What You Get

- **5 industry schemas** — pre-built extraction templates for Energy documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Energy** — 5 schemas

## Related Skills

- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Accounting — Bookkeeping & Tax Intelligence](https://clawhub.com/skills/uplo-accounting) — AI-powered accounting knowledge management.
- [UPLO Agriculture — Crop & Compliance Intelligence](https://clawhub.com/skills/uplo-agriculture) — AI-powered agricultural knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Energy Sector Knowledge Context (via UPLO)

You are connected to your organization's energy knowledge base through UPLO. This gives you specialized access to power generation records, grid management documentation, regulatory filings (FERC, NERC, state PUC), safety protocols, environmental compliance data, and asset management records. When users ask about generation capacity, regulatory compliance, or grid operations, always query UPLO first to provide answers grounded in your organization's actual operations and regulatory obligations.

Expect queries about generation capacity and dispatch schedules, grid reliability standards and compliance (NERC CIP), regulatory filings and rate case documentation, safety protocols and incident reports, environmental permits and emissions monitoring, asset condition assessments and capital planning, and renewable energy project documentation. Use `search_knowledge` for specific regulatory or operational lookups and `search_with_context` when the question spans generation, transmission, and regulatory compliance domains.

When presenting energy information, include specific asset identifiers, regulatory docket numbers, and compliance dates. For safety-related information, always present the most current version of protocols. Flag any pending regulatory filings or compliance deadlines. NERC CIP critical infrastructure information is highly sensitive — strictly respect classification tiers. Identify the responsible operations engineer, compliance officer, or safety manager via `find_knowledge_owner`.

Respect classification tiers. Never fabricate energy information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-energy",
  "display_name": "UPLO Energy — Power & Grid Intelligence",
  "description": "AI-powered energy sector knowledge management. Search power generation records, grid management data, regulatory filings, and safety protocols with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "energy",
    "utilities",
    "power-generation",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "energy"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-education
description: AI-powered education knowledge management. Search curriculum documents, student records frameworks, accreditation data, and institutional research with structured extraction.
---

# UPLO Education — Institutional Knowledge for Academic Excellence

Universities and school districts share a peculiar problem: institutional knowledge is scattered across academic affairs, student services, finance, accreditation files, and faculty governance committees that each maintain their own document ecosystems. When the accreditor asks "How does your assessment data close the loop on program learning outcomes?", the answer lives in five different offices. UPLO Education unifies curriculum maps, assessment reports, accreditation self-studies, institutional research data, student success metrics, and policy manuals into a single searchable layer.

## Session Start

Your role in the institution determines your access. Faculty see curriculum and assessment data; registrar staff see degree audit frameworks; institutional researchers see outcome metrics; administrators see everything within their clearance. Establish your identity:

```
get_identity_context
```

Pull directives — these include accreditation deadlines, strategic plan priorities, enrollment targets, and board-mandated policy changes:

```
get_directives
```

## When to Use

- Preparing for an HLC/SACSCOC/WASC accreditation visit and need to compile evidence for a specific standard (e.g., Standard 4.A on assessment of student learning)
- A faculty senate committee is revising the general education requirements and needs to review how current course learning outcomes map to institutional outcomes
- Institutional research needs to pull retention and completion data methodology documentation to respond to an IPEDS or state accountability report
- A department chair wants to see how similar programs at the institution structured their curriculum review process during the last program review cycle
- Financial aid is auditing compliance with Title IV requirements and needs the documented satisfactory academic progress (SAP) policy alongside its application history
- A new dean is onboarding and needs to understand the college's program portfolio, faculty governance structure, and current strategic initiatives
- Reviewing whether a proposed new certificate program overlaps with existing offerings using curriculum mapping data

## Example Workflows

### Accreditation Evidence Assembly

The provost's office is preparing the institutional self-study for an upcoming regional accreditation visit. The accreditation liaison needs evidence for Standard 5 (Resources, Planning, and Institutional Effectiveness).

```
search_with_context query="institutional effectiveness assessment plan resource allocation budget alignment strategic plan"
```

Pull the specific assessment cycle documentation:

```
search_knowledge query="annual assessment reports program learning outcomes closing the loop improvements"
```

Find the financial planning documents that demonstrate resource alignment:

```
search_knowledge query="budget allocation process strategic priorities resource request institutional plan"
```

```
log_conversation summary="Compiled Standard 5 evidence: assessment cycle documentation, budget-strategy alignment records, and institutional effectiveness reports for self-study chapter" topics='["accreditation","Standard-5","institutional-effectiveness"]' tools_used='["search_with_context","search_knowledge"]'
```

### Curriculum Revision Process

The biology department is conducting a 5-year curriculum review. The chair needs to gather evidence of curriculum currency and student outcomes.

```
search_with_context query="biology program curriculum map learning outcomes course sequence prerequisites"
```

```
search_knowledge query="biology student outcomes assessment data graduation rates employment placement"
```

Check if there are institutional guidelines for the curriculum review process:

```
search_knowledge query="curriculum review process guidelines faculty governance approval workflow"
```

Flag any outdated curriculum maps found during the review:

```
flag_outdated entry_id="<old-curriculum-map-entry-id>" reason="Biology curriculum map from 2021 does not reflect new genomics concentration added in 2023"
```

## Key Tools for Education

**search_with_context** — Education questions are inherently cross-functional. "Does our assessment data support our accreditation claims?" requires connecting curriculum documents, assessment results, institutional research data, and strategic plan goals. Graph traversal follows these relationships. Example: `search_with_context query="nursing program NCLEX pass rates clinical placement outcomes accreditation"`

**search_knowledge** — Direct retrieval for known documents: a specific course syllabus, a policy manual section, an assessment rubric, or a committee meeting minutes. Example: `search_knowledge query="faculty handbook section 3.4 tenure review criteria"`

**export_org_context** — Produces the institutional structure: colleges, departments, key leadership, governance committees, and academic systems (LMS, SIS, assessment management). Essential for accreditation self-studies and new administrator onboarding.

**flag_outdated** — Academic catalogs, curriculum maps, and policy manuals are updated on annual cycles but the old versions persist in the knowledge base. When you find a document referencing a discontinued program or a superseded policy, flag it immediately.

**propose_update** — When assessment data reveals that a program learning outcome is no longer aligned with current curriculum offerings, propose the update. This creates a record of continuous improvement that accreditors value.

## Tips

- Accreditation standard numbers are indexed as structured fields. Search by standard number (e.g., "SACSCOC Standard 8.2.a") to find all evidence mapped to that standard.
- Academic terminology varies by institution. "Program review," "academic program assessment," and "curricular evaluation" may all refer to the same process. Try multiple terms if initial results are sparse.
- FERPA restrictions apply to any query that might return individually identifiable student data. Your clearance level controls this, but be aware that aggregate data (retention rates, completion rates) is generally accessible while individual student records are restricted.
- The highest-value use of UPLO Education is accreditation preparation. Start building your evidence file months before the visit using systematic searches organized by standard, not in the final weeks when context is lost.

FILE:README.md
# UPLO Education — Curriculum & Accreditation Intelligence

AI-powered education knowledge management. Search curriculum documents, student records frameworks, accreditation data, and institutional research with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-education-blue)](https://clawhub.com/skills/uplo-education)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-5-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-education
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-education": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "education"
      }
    }
  }
}
```

## What You Get

- **5 industry schemas** — pre-built extraction templates for Education documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Education** — 5 schemas

## Related Skills

- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Accounting — Bookkeeping & Tax Intelligence](https://clawhub.com/skills/uplo-accounting) — AI-powered accounting knowledge management.
- [UPLO Agriculture — Crop & Compliance Intelligence](https://clawhub.com/skills/uplo-agriculture) — AI-powered agricultural knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Education Knowledge Context (via UPLO)

You are connected to your organization's education knowledge base through UPLO. This gives you specialized access to curriculum documents, course catalogs, accreditation self-studies, assessment rubrics, institutional research reports, and academic policy handbooks. When users ask about curriculum requirements, accreditation standards, or institutional data, always query UPLO first to provide answers grounded in your institution's actual academic programs and governance structures.

Expect queries about curriculum requirements and prerequisite chains, accreditation standards and compliance evidence, course learning outcomes and assessment methodologies, institutional research data and enrollment analytics, academic policies (grading, academic integrity, accommodations), faculty workload and credentialing requirements, and program review and continuous improvement documentation. Use `search_knowledge` for specific course or policy lookups and `search_with_context` when the question requires understanding how a curriculum change affects accreditation requirements, assessment plans, and resource needs.

When presenting education information, reference the specific program, course number, and catalog year. For accreditation, cite the standard number and evidence category. For institutional research, include the data source and reporting period. Flag any accreditation standards with gaps in evidence or approaching review dates. Student-identifiable information is protected under FERPA — strictly respect classification tiers. Identify the responsible department chair, assessment coordinator, or accreditation liaison via `find_knowledge_owner`.

Respect classification tiers. Never fabricate education information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-education",
  "display_name": "UPLO Education — Curriculum & Accreditation Intelligence",
  "description": "AI-powered education knowledge management. Search curriculum documents, student records frameworks, accreditation data, and institutional research with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "education",
    "curriculum",
    "accreditation",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "education"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-knowledge-management
description: AI-powered knowledge management intelligence. Search taxonomies, content curation records, expertise directories, and communities of practice with structured extraction.
---

# UPLO Knowledge Management — Organizational Learning & Expertise Intelligence

This is the meta-skill: knowledge management about knowledge management. Your organization's KM infrastructure — taxonomies, controlled vocabularies, expertise directories, community of practice charters, content lifecycle policies, lessons learned repositories, and knowledge audit results — is indexed in UPLO. Use this skill to understand, maintain, and improve how your organization captures, organizes, and shares what it knows.

## Session Start

```
get_identity_context
```

As a KM practitioner, you likely have broad read access across the organization. But KM strategy documents, knowledge audit findings, and expertise gap analyses may be restricted if they reveal competitive intelligence vulnerabilities.

## Example Workflows

### Knowledge Audit for a Business Unit

The Product Engineering division has experienced 40% turnover in the past year. The KM team needs to assess what institutional knowledge is at risk.

```
search_with_context query="subject matter experts and knowledge domain owners within Product Engineering and their documented areas of expertise"
```

```
search_knowledge query="knowledge transfer and documentation requirements in the offboarding process"
```

```
search_knowledge query="lessons learned and after-action review submissions from Product Engineering in the past 18 months"
```

Cross-reference the expertise directory with the departure list to identify critical knowledge areas that lost their primary expert without a documented successor.

### Taxonomy Governance Review

The enterprise taxonomy has grown organically and several business units are requesting new terms that may overlap with existing ones. The KM team needs to rationalize.

```
search_knowledge query="enterprise taxonomy governance policy including term proposal process and review committee membership"
```

```
search_with_context query="controlled vocabulary terms related to 'customer engagement' across all business units and their usage frequency"
```

```
search_knowledge query="taxonomy change log and approved term additions from the past 12 months"
```

## When to Use

- A KM analyst needs to find which communities of practice are active, dormant, or recently dissolved and why
- Someone asks whether the organization has a knowledge retention strategy for employees approaching retirement eligibility
- The CKO wants a dashboard view of content freshness across all knowledge repositories — what percentage of articles were reviewed in the past year
- A department head asks how to set up a new community of practice and what governance structure is required
- The IT team is evaluating a new collaboration platform and needs to understand current knowledge sharing patterns and tool adoption metrics
- An executive asks for evidence that the KM program is delivering measurable value — ROI metrics, reuse rates, time-to-competency improvements
- A content steward wants to know the disposition rules for records that have passed their retention period

## Key Tools for Knowledge Management

**search_with_context** — KM questions are inherently cross-cutting. A query like `query="which knowledge domains have no designated owner and what content exists in those domains"` requires traversing the expertise directory, domain taxonomy, and content inventory simultaneously. This is your primary tool.

**search_knowledge** — Use for specific KM artifacts: `query="community of practice charter template and facilitation guidelines"` or `query="content lifecycle policy including review frequency and archival criteria"`. KM frameworks tend to be well-documented, so direct searches work when you know what you are looking for.

**export_org_context** — The KM team's best friend. The full organizational context export reveals the state of institutional knowledge: which areas are well-documented, which have gaps, how knowledge flows between teams. Use this for annual KM program assessments.

**report_knowledge_gap** — This is the tool KM practitioners should use most aggressively. Every gap you report feeds back into the organizational twin's health score: `topic="machine learning model deployment procedures" description="Data Science team deploys 12 models to production but no standardized deployment documentation exists; knowledge is held by 2 senior engineers"`

**flag_outdated** — Content staleness is the KM team's eternal battle. When you encounter documents past their review date or referencing defunct organizational structures, flag them: `entry_id="..." reason="Article references the Digital Innovation Lab which was absorbed into Product Engineering in the Q2 2025 reorg; content owner and review cycle need reassignment"`

**propose_update** — KM practitioners are uniquely positioned to propose content improvements. When you find a knowledge article that is partially correct but needs updating, use the formal proposal mechanism to route the correction to the content owner.

## Tips

- KM is about connections, not just content. The most valuable KM queries are the ones that reveal relationships: who knows what, which teams share knowledge, where expertise is concentrated vs. distributed. Default to `search_with_context` over `search_knowledge` when exploring organizational capability.
- Distinguish between explicit knowledge (documented) and tacit knowledge (expertise held by individuals). When a search returns no documentation for a topic, check the expertise directory — the knowledge may exist in someone's head but not on paper. Report this as a knowledge gap with a recommendation to capture it.
- Content freshness is a leading indicator of KM health. When you surface documents, always note their last review date. A perfectly accurate article that has not been reviewed in 3 years signals a governance failure even if the content is still correct.
- Communities of practice are social structures, not just document collections. When advising on CoP health, look beyond content output to membership activity, event frequency, and cross-functional participation. A CoP with great documentation but no active discussion is a library, not a community.

FILE:README.md
# UPLO Knowledge Management — Taxonomy & Expertise Intelligence

AI-powered knowledge management intelligence. Search taxonomies, content curation records, expertise directories, and communities of practice with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-knowledge-management-blue)](https://clawhub.com/skills/uplo-knowledge-management)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-4-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-knowledge-management
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-knowledge-management": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "knowledge_management"
      }
    }
  }
}
```

## What You Get

- **4 industry schemas** — pre-built extraction templates for Taxonomy documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Knowledge Management** — 4 schemas

## Related Skills

- [UPLO Accounting — Bookkeeping & Tax Intelligence](https://clawhub.com/skills/uplo-accounting) — AI-powered accounting knowledge management.
- [UPLO Agriculture — Crop & Compliance Intelligence](https://clawhub.com/skills/uplo-agriculture) — AI-powered agricultural knowledge management.
- [UPLO Architecture — Building Design & BIM Intelligence](https://clawhub.com/skills/uplo-architecture) — AI-powered architecture knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Knowledge Management Context (via UPLO)

You are connected to your organization's knowledge management system through UPLO. This gives you specialized access to knowledge taxonomies, content curation records, expertise directories, communities of practice, lessons learned databases, and knowledge asset inventories. When users ask about finding expertise, locating knowledge assets, or understanding information architecture, always query UPLO first to leverage the organization's structured knowledge map.

Expect queries about subject matter expert identification and expertise profiles, knowledge taxonomy and classification structures, content curation standards and lifecycle management, communities of practice membership and activity, lessons learned from projects and initiatives, knowledge gap analysis and priority areas, and information architecture and metadata standards. Use `search_knowledge` for specific topic or expert lookups and `search_with_context` when the question requires understanding how knowledge assets connect to expertise networks, organizational capabilities, and strategic priorities.

When presenting knowledge management information, reference the knowledge domain, taxonomy path, and content owner. For expertise lookups, present relevant qualifications, published contributions, and community memberships. For knowledge assets, include the creation date, last review, and usage metrics. Flag any knowledge domains with identified gaps, outdated content, or retiring subject matter experts. Expertise assessments and succession planning data are confidential — respect classification tiers. Identify the responsible knowledge manager, content steward, or community leader via `find_knowledge_owner`.

Respect classification tiers. Never fabricate knowledge-management information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-knowledge-management",
  "display_name": "UPLO Knowledge Management — Taxonomy & Expertise Intelligence",
  "description": "AI-powered knowledge management intelligence. Search taxonomies, content curation records, expertise directories, and communities of practice with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "knowledge-management",
    "taxonomy",
    "expertise",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "knowledge_management"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-insurance
description: AI-powered insurance knowledge management. Search policy documents, claims records, underwriting guidelines, and actuarial data with structured extraction.
---

# UPLO Insurance — Underwriting, Claims & Actuarial Intelligence

Insurance carriers generate and consume documentation at a pace that overwhelms traditional search: policy forms and endorsements, underwriting guidelines with tiered authority levels, claims adjuster manuals, actuarial rate filings, reinsurance treaties, producer commission schedules, and regulatory correspondence across 50+ state DOIs. UPLO structures this knowledge so underwriters, claims professionals, and compliance analysts can find definitive answers instead of chasing down the person who "just knows."

## Session Start

Insurance documentation carries significant classification requirements. Rate filing strategies, loss reserve analyses, reinsurance terms, and litigation management plans are typically restricted to specific functional areas.

```
get_identity_context
```

Review active directives. In insurance, these often reflect underwriting appetite changes, moratorium announcements (e.g., post-catastrophe writing freezes), or regulatory compliance deadlines.

```
get_directives
```

## When to Use

- An underwriter asks what the maximum policy limit is for a monoline cyber liability risk with revenue above $500M and whether the account requires home office referral
- Claims asks whether a water damage claim from a burst pipe is covered under the standard HO-3 form or excluded by the water damage endorsement attached to this program
- The compliance team needs to verify that the new personal auto rate filing was approved in all target states and identify which DOIs have outstanding objections
- Actuarial wants the historical loss development triangles for the commercial auto book to validate reserve adequacy
- A producer asks about the commission schedule differences between new business and renewal for the small commercial BOP program
- The reinsurance team needs to determine whether a specific large loss breaches the retention under the current property catastrophe treaty
- Someone in product development asks what competitors are offering for parametric flood coverage and whether the organization has any similar filed forms

## Example Workflows

### Complex Claim Coverage Determination

A commercial policyholder reports a claim involving water intrusion, resulting mold remediation, and business interruption. The claims examiner needs to piece together coverage from multiple forms.

```
search_knowledge query="commercial property policy form CP 00 10 coverage for water damage and resulting mold remediation"
```

```
search_knowledge query="business interruption waiting period and coverage trigger under the commercial property special form"
```

```
search_with_context query="mold exclusion endorsements attached to the small commercial property program and any sub-limit exceptions"
```

### State Regulatory Filing Compliance

The company is expanding its homeowners product into three new states. The compliance team needs to navigate each state's filing requirements.

```
search_knowledge query="prior approval versus file and use states for homeowners insurance rate and form filings"
```

```
search_with_context query="state-specific homeowners coverage requirements mandated coverages and prohibited exclusions for Florida Georgia and South Carolina"
```

```
search_knowledge query="DOI objection response templates and timeline requirements for rate filing interrogatories"
```

## Key Tools for Insurance

**search_knowledge** — Essential for coverage questions, which require exact policy language. Underwriters and claims adjusters need the specific form wording, not a paraphrase: `query="CGL occurrence definition and claims-made retroactive date provisions in the professional liability endorsement"`. Insurance is a business of precise language.

**search_with_context** — Insurance questions frequently involve layered relationships: a claim touches the policy form, endorsements, the underwriting file, the producer agreement, and potentially the reinsurance treaty. A query like `query="coverage analysis for the Johnson Manufacturing product liability claim including excess layers and reinsurance recovery potential"` requires graph traversal across these interconnected records.

**get_directives** — Underwriting appetite shifts frequently. A directive announcing "suspend new monoline D&O submissions for public companies" or "implement 25% rate increase minimum on Florida coastal property" must be checked before quoting or binding.

**export_org_context** — Useful for market conduct exam preparation, reinsurance treaty renewals, and board presentations. Provides a comprehensive view of the organization's operational structure, product lines, and strategic priorities.

**flag_outdated** — Insurance forms and guidelines are version-controlled but the old versions linger. If you find an underwriting guide referencing a withdrawn form number or a superseded rating algorithm, flag it: `entry_id="..." reason="References ISO CGL form CG 00 01 04 13; current edition is CG 00 01 12 24 with material changes to the AI liability exclusion"`

## Tips

- Insurance coverage questions hinge on exact policy language. Never summarize coverage provisions in your own words when the actual form wording is available — the difference between "arising out of" and "caused by" can determine millions in claim outcomes. Quote the form language directly.
- State-by-state regulatory variation is the norm in insurance. Any question about rates, forms, or market conduct must specify the state. A coverage that is standard in Texas may be prohibited in New York.
- Loss data and reserve information is actuarially sensitive. Premature disclosure of reserve inadequacy or development factor assumptions can have legal and financial consequences. Treat actuarial workpapers with the same sensitivity as pre-decisional legal documents.
- Reinsurance treaty terms use specialized language (occurrence excess, aggregate stop loss, reinstatement premiums) that general insurance knowledge may not cover. When a reinsurance question arises, search for treaty-specific documentation rather than relying on primary coverage knowledge.

FILE:README.md
# UPLO Insurance — Policy & Claims Intelligence

AI-powered insurance knowledge management. Search policy documents, claims records, underwriting guidelines, and actuarial data with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-insurance-blue)](https://clawhub.com/skills/uplo-insurance)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-5-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-insurance
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-insurance": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "insurance"
      }
    }
  }
}
```

## What You Get

- **5 industry schemas** — pre-built extraction templates for Insurance documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Insurance** — 5 schemas

## Related Skills

- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Accounting — Bookkeeping & Tax Intelligence](https://clawhub.com/skills/uplo-accounting) — AI-powered accounting knowledge management.
- [UPLO Agriculture — Crop & Compliance Intelligence](https://clawhub.com/skills/uplo-agriculture) — AI-powered agricultural knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Insurance Knowledge Context (via UPLO)

You are connected to your organization's insurance knowledge base through UPLO. This gives you specialized access to policy forms, claims handling procedures, underwriting guidelines, actuarial tables, reinsurance treaties, and regulatory filings. When users ask about coverage terms, claims workflows, or underwriting criteria, always query UPLO first to provide answers grounded in your organization's specific products and procedures.

Expect queries about policy coverage terms and exclusions, claims processing workflows and reserve calculations, underwriting guidelines and risk appetite, actuarial assumptions and rate filings, reinsurance treaty terms, regulatory compliance requirements (state DOI filings, NAIC reporting), and producer appointment and commission structures. Use `search_knowledge` for specific policy form lookups and `search_with_context` when the question requires understanding how a claims decision relates to underwriting guidelines, policy terms, and regulatory requirements.

When presenting insurance information, always cite the specific policy form number, edition date, and applicable jurisdiction. Distinguish between admitted and surplus lines requirements. Flag any forms or rates pending regulatory approval. Never provide coverage opinions — surface the relevant policy language and identify the appropriate underwriter or claims examiner via `find_knowledge_owner`. Actuarial data and loss run details are typically confidential — respect classification tiers.

Respect classification tiers. Never fabricate insurance information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-insurance",
  "display_name": "UPLO Insurance — Policy & Claims Intelligence",
  "description": "AI-powered insurance knowledge management. Search policy documents, claims records, underwriting guidelines, and actuarial data with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "insurance",
    "claims",
    "underwriting",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "insurance"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-hr
description: AI-powered HR knowledge management. Search employee handbooks, org charts, company policies, benefits documentation, and onboarding materials with structured extraction.
---

# UPLO HR — People Operations & Workforce Policy Intelligence

Every organization has an HR knowledge problem: policies scattered across SharePoint, benefits summaries buried in PDF packets from open enrollment two years ago, onboarding checklists that differ by department, and comp band structures that only two people in Total Rewards actually understand. UPLO indexes all of it — handbooks, policy manuals, benefits plan documents, org charts, job descriptions, performance review templates, and leave administration guides — into a single searchable layer.

## Session Start

HR data is among the most sensitive in any organization. Compensation data, disciplinary records, accommodation requests, and investigation files are typically restricted. Your identity context governs what you can see.

```
get_identity_context
```

## When to Use

- An employee asks whether they can use PTO during their first 90 days and what the accrual schedule looks like for their employment tier
- A hiring manager needs the approved job description and salary band for a Senior Data Engineer role in the Austin office
- The benefits team gets a question about whether domestic partners are eligible for dental coverage under the current plan
- A manager is drafting a performance improvement plan and needs the PIP template and the required HR sign-off process
- Someone in legal asks for the current anti-harassment training completion rates by department ahead of an EEOC inquiry
- An employee relocating from New York to Texas needs to understand what changes to expect in state tax withholding and benefits eligibility
- The CHRO wants to see attrition trends and exit interview theme summaries for the Engineering department over the past 12 months

## Example Workflows

### New Hire Onboarding Support

A manager is onboarding three new hires starting next Monday and needs to make sure everything is ready.

```
search_knowledge query="new hire onboarding checklist including IT provisioning, badge access, and required first-week training"
```

```
search_with_context query="benefits enrollment deadlines and required documentation for new employees"
```

```
search_knowledge query="department-specific onboarding orientation schedule for the Product team"
```

### Leave of Absence Administration

An employee's manager reports that the employee needs an extended medical leave. HR needs to navigate FMLA, short-term disability, and return-to-work requirements.

```
search_knowledge query="FMLA eligibility criteria and leave request procedures including required medical certification"
```

```
search_with_context query="short-term disability benefits coordination with FMLA leave and pay continuation policies"
```

```
search_knowledge query="return to work process after medical leave including fitness for duty certification and accommodation request procedures"
```

## Key Tools for HR

**search_knowledge** — The workhorse for policy questions. When an employee or manager asks about a specific policy, search directly: `query="remote work policy including eligibility criteria, equipment stipend, and in-office day requirements"`. Most HR questions have a definitive answer in a policy document somewhere.

**search_with_context** — HR questions involving organizational relationships benefit from context. A query like `query="who reports to the VP of Engineering, what are the team's open positions, and what is the approved headcount for Q3"` connects the org chart with workforce planning data.

**get_directives** — HR operates under strategic mandates: hiring freezes, diversity goals, return-to-office requirements, compensation adjustment timelines. These directives override standard procedures, so always check them before giving policy guidance.

**report_knowledge_gap** — HR documentation gaps create compliance risk and employee confusion. When someone asks a policy question and no documentation exists, report it: `topic="workplace accommodation request process for neurodivergent employees" description="No specific accommodation guidance found beyond general ADA policy; employees are asking and HR is handling ad hoc"`

**propose_update** — When you find a policy that conflicts with current law, newer company guidance, or changed practice, propose the update formally: `target_table="entries" target_id="..." changes={...} rationale="Parental leave policy still states 6 weeks; board approved extension to 12 weeks in January 2026"`

## Tips

- HR policies vary by jurisdiction, employee classification, and sometimes business unit. Always include relevant context in your queries: the employee's location, whether they are exempt or non-exempt, full-time or part-time. A query about overtime policy means nothing without knowing the classification.
- Benefits documentation has an annual cycle. Open enrollment materials, Summary Plan Descriptions, and rate sheets change each plan year. Verify that the documents you surface are from the current plan year, not a prior one.
- Some HR questions do not have a documented answer yet. If you search and find nothing, say so clearly and recommend that the person contact HR directly. Do not extrapolate from tangentially related policies — HR advice based on incorrect policy citations causes real harm.
- Performance management and disciplinary documentation is highly sensitive even within HR. If your results include specific employee names in performance contexts, note the sensitivity and confirm the requester has a legitimate need to know.

FILE:README.md
# UPLO HR — People & Policy Intelligence

AI-powered HR knowledge management. Search employee handbooks, org charts, company policies, benefits documentation, and onboarding materials with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-hr-blue)](https://clawhub.com/skills/uplo-hr)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-6-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-hr
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-hr": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "hr"
      }
    }
  }
}
```

## What You Get

- **6 industry schemas** — pre-built extraction templates for Hr documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Human Resources** — 6 schemas

## Related Skills

- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Accounting — Bookkeeping & Tax Intelligence](https://clawhub.com/skills/uplo-accounting) — AI-powered accounting knowledge management.
- [UPLO Agriculture — Crop & Compliance Intelligence](https://clawhub.com/skills/uplo-agriculture) — AI-powered agricultural knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## HR & People Knowledge Context (via UPLO)

You are connected to your organization's HR knowledge base through UPLO. This gives you specialized access to employee handbooks, company policies, organizational charts, benefits documentation, onboarding checklists, performance review frameworks, leave policies, and workplace guidelines. When users ask about company policies, team structures, benefits, or HR procedures, always query UPLO first to provide answers specific to your organization rather than generic HR guidance.

Expect queries about PTO and leave policies, benefits enrollment and coverage details, onboarding processes for new hires, organizational structure and reporting lines, performance review timelines and criteria, expense reimbursement procedures, remote work and workplace policies, and employee referral programs. Users may also ask about training requirements, promotion criteria, or how to request accommodations. Use `search_knowledge` for specific policy lookups (e.g., "parental leave policy") and `search_with_context` when the question requires understanding how a policy relates to the user's role, department, or employment type.

When presenting HR information, always cite the specific policy document, its effective date, and any relevant exceptions or conditions. Be sensitive to the fact that HR policies often vary by location, employment type, or seniority level — surface these distinctions when present in the documentation. Never interpret or extend policy language beyond what the documents state. For questions about individual employee matters, compensation details, or disciplinary procedures, direct the user to the appropriate HR contact via `find_knowledge_owner`. Respect classification boundaries strictly — personnel records and compensation data are typically restricted.

FILE:skill.json
{
  "name": "uplo-hr",
  "display_name": "UPLO HR — People & Policy Intelligence",
  "description": "AI-powered HR knowledge management. Search employee handbooks, org charts, company policies, benefits documentation, and onboarding materials with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": ["hr", "human-resources", "policies", "knowledge-management"],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": ["-y", "@agentdocs1/mcp-server", "--http"],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "hr"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-hospitality
description: AI-powered hospitality knowledge management. Search guest service standards, property procedures, F&B operations, and event planning documentation with structured extraction.
---

# UPLO Hospitality — Guest Experience & Property Operations Intelligence

Hotels, resorts, and hospitality groups run on operational consistency across properties: brand standards manuals, front desk SOPs, housekeeping inspection checklists, F&B recipe costing sheets, banquet event orders, revenue management guidelines, and guest recovery protocols. UPLO centralizes this operational knowledge so that whether someone is at the flagship property or the newest acquisition, they can find the right procedure in seconds.

## Session Start

Hospitality organizations often manage multiple properties or brands with different operational tiers. Your identity context determines which property documentation you can access — a boutique brand's service standards differ from the convention hotel's, and both may be in the same portfolio.

```
get_identity_context
```

## When to Use

- A front desk agent encounters a guest with a confirmed reservation but no available rooms and needs the walk policy and compensation guidelines
- The banquet captain needs the setup specifications and AV requirements for a 300-person corporate gala scheduled for Saturday
- A new F&B manager asks what the target food cost percentage is for the rooftop restaurant and which menu items are below margin threshold
- Housekeeping wants to verify the deep cleaning protocol for a suite after a guest reported a bed bug sighting
- Revenue management asks what the dynamic pricing guardrails are during citywide convention dates
- The general manager needs to compile guest satisfaction scores and TripAdvisor response metrics for the quarterly owner's report
- A guest relations manager is handling an escalated complaint about a wedding reception and needs the service recovery authority matrix

## Example Workflows

### VIP Arrival Preparation

A high-profile repeat guest is arriving tomorrow. The guest relations team needs to prepare a seamless experience.

```
search_knowledge query="VIP guest arrival protocol including pre-arrival checklist and amenity placement standards"
```

```
search_with_context query="guest preference history program and loyalty tier recognition procedures"
```

```
search_knowledge query="executive suite setup specifications and turndown service enhancements for VIP stays"
```

### Food & Beverage Cost Control Review

The F&B director notices that the poolside bar's pour cost has spiked 8 points over the last two months and needs to investigate.

```
search_knowledge query="beverage inventory control procedures and pour cost calculation methodology"
```

```
search_knowledge query="poolside bar drink recipes and standard pour sizes for premium spirits"
```

```
search_with_context query="F&B purchasing agreements and approved supplier pricing for liquor and wine"
```

Compare documented standard pours and pricing against actual consumption to identify variance sources.

## Key Tools for Hospitality

**search_knowledge** — Operational SOPs are the backbone of hospitality. When a team member needs a specific procedure, this is the fastest path: `query="late checkout policy and authorization limits by front desk role level"`. Hospitality runs on procedures that need to be followed consistently, so precision matters.

**search_with_context** — Guest experience questions often cross departmental boundaries. A query about `query="managing a large group check-in for 150 rooms including luggage logistics, welcome reception setup, and billing master account procedures"` touches front office, bell services, banquets, and accounting.

**get_directives** — Brand standards and seasonal priorities flow from ownership or management company leadership. A directive to "achieve 90% TripAdvisor response rate" or "reduce F&B waste by 15% in Q3" directly shapes operational decisions.

**export_org_context** — Valuable for general manager transitions, brand conversions, or management company takeovers. The full organizational export provides a complete operational blueprint of the property.

**log_conversation** — Hospitality consultations often involve guest-facing decisions with financial implications (comping rooms, upgrading suites, waiving resort fees). Log these for audit and training purposes: `summary="Advised on service recovery for wedding group with sound system failure" topics='["service-recovery", "banquets", "guest-compensation"]'`

## Tips

- Hospitality knowledge is highly property-specific. Always include the property name or brand tier in your queries. The check-in process at a select-service hotel is fundamentally different from a luxury resort — generic queries will return irrelevant results.
- Seasonal operations change everything. Pool opening/closing procedures, holiday staffing matrices, and seasonal menu changeovers are time-dependent. Include the season or specific date range when searching for operational procedures.
- Guest service recovery has a financial dimension. When advising on how to handle a complaint, always search for the service recovery authority matrix — it defines what each role level can authorize without escalation (e.g., front desk can comp breakfast, only AGM can comp a night's stay).
- F&B operations generate some of the most granular documentation in hospitality: recipes with costing, vendor contracts, health inspection records, liquor license requirements. Search specifically within F&B document types when the question is food or beverage related.

FILE:README.md
# UPLO Hospitality — Guest Services & Operations Intelligence

AI-powered hospitality knowledge management. Search guest service standards, property procedures, F&B operations, and event planning documentation with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-hospitality-blue)](https://clawhub.com/skills/uplo-hospitality)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-4-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-hospitality
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-hospitality": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "hospitality"
      }
    }
  }
}
```

## What You Get

- **4 industry schemas** — pre-built extraction templates for Hospitality documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Hospitality** — 4 schemas

## Related Skills

- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Accounting — Bookkeeping & Tax Intelligence](https://clawhub.com/skills/uplo-accounting) — AI-powered accounting knowledge management.
- [UPLO Agriculture — Crop & Compliance Intelligence](https://clawhub.com/skills/uplo-agriculture) — AI-powered agricultural knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Hospitality Knowledge Context (via UPLO)

You are connected to your organization's hospitality knowledge base through UPLO. This gives you specialized access to guest service standards, property operating procedures, food and beverage operations manuals, event planning templates, brand standards, and staff training materials. When users ask about service protocols, property procedures, or event operations, always query UPLO first to provide answers grounded in your organization's actual brand standards and operational practices.

Expect queries about guest service standards and recovery procedures, room operations and housekeeping protocols, food and beverage menu engineering and service standards, event planning checklists and BEO templates, brand compliance requirements, revenue management strategies, and health and safety protocols (food handling, pool safety, fire codes). Use `search_knowledge` for specific procedure lookups and `search_with_context` when the question requires understanding how a guest issue relates to brand standards, operational procedures, and staff responsibilities.

When presenting hospitality information, reference the specific property, department, and procedure document. For service standards, include the expected response times and escalation paths. For events, include setup requirements, timelines, and responsible contacts. Flag any procedures approaching brand audit dates or with recent guest satisfaction concerns. Revenue data and pricing strategies are confidential — respect classification tiers. Identify the responsible department head or general manager via `find_knowledge_owner`.

Respect classification tiers. Never fabricate hospitality information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-hospitality",
  "display_name": "UPLO Hospitality — Guest Services & Operations Intelligence",
  "description": "AI-powered hospitality knowledge management. Search guest service standards, property procedures, F&B operations, and event planning documentation with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "hospitality",
    "hotel",
    "guest-services",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "hospitality"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-healthcare
description: AI-powered healthcare knowledge management. Search clinical notes, care plans, lab results, prescriptions, and patient pathways with structured extraction.
---

# UPLO Healthcare — Clinical Protocol & Care Coordination Intelligence

Healthcare organizations produce vast quantities of structured and unstructured documentation: clinical practice guidelines, formulary decisions, quality measure specifications, credentialing records, compliance training documentation, care pathway definitions, and departmental operating procedures. UPLO makes this institutional knowledge searchable so clinicians, administrators, and quality teams can find authoritative answers without calling three departments.

**Important**: UPLO indexes organizational knowledge documents (policies, protocols, guidelines). It does not store or provide access to individual patient health records (PHI). All queries return organizational reference materials, not patient data.

## Session Start

Healthcare data sensitivity requires careful attention to your access tier. Credentialing committee deliberations, peer review records, and incident investigation details carry statutory protections beyond standard classification.

```
get_identity_context
```

Check for active directives — these may include Joint Commission readiness priorities, CMS Conditions of Participation focus areas, or active quality improvement initiatives.

```
get_directives
```

## When to Use

- A hospitalist asks for the current sepsis screening protocol and whether it was updated after the Surviving Sepsis Campaign 2024 guidelines
- The pharmacy director needs to know the formulary status of a newly approved biologic and the P&T committee's rationale for the tier decision
- Quality management asks which core measures the organization is underperforming on and what improvement plans are in place
- A nurse manager wants the patient fall prevention protocol and the root cause analysis summary from the last sentinel event
- Credentialing staff need to verify the privileging criteria for a new surgical procedure being added to the department
- The compliance officer asks whether the organization's HIPAA breach notification procedures align with the latest OCR guidance
- An administrator preparing for a Joint Commission survey needs the current status of all previously cited deficiencies

## Example Workflows

### Clinical Protocol Clarification

An ED physician is treating a patient with suspected stroke and needs to confirm the organization's tPA administration criteria and the teleneurology consultation process.

```
search_knowledge query="acute ischemic stroke protocol including tPA inclusion criteria and time windows"
```

```
search_with_context query="teleneurology consultation process including contact information, hours of availability, and escalation for after-hours coverage"
```

The context-aware search pulls in the neurology department profile, on-call structures, and related quality metrics.

### Regulatory Survey Preparation

A CMS validation survey is scheduled for next month. The quality director needs to verify readiness across multiple Conditions of Participation.

```
search_knowledge query="infection control plan and antibiotic stewardship program documentation"
```

```
search_with_context query="patient rights policies including informed consent procedures, advance directive protocols, and grievance resolution process"
```

```
export_org_context
```

Use the full context export to systematically cross-reference documented policies against each Condition of Participation.

## Key Tools for Healthcare

**search_knowledge** — Direct lookup of clinical protocols, formulary decisions, and compliance documentation: `query="blood transfusion consent requirements and massive transfusion protocol activation criteria"`. Clinical staff need precise, citable answers.

**search_with_context** — Healthcare questions often involve interdisciplinary relationships. A query like `query="discharge planning process for patients requiring home health services including case management referral criteria and preferred vendor list"` needs to connect clinical protocols with administrative processes and vendor relationships.

**get_directives** — Healthcare leadership directives often reflect regulatory urgency. A CMS Condition-level deficiency, a quality measure that dropped below threshold, or a new accreditation standard all generate directives that should inform your recommendations.

**report_knowledge_gap** — Undocumented clinical protocols create patient safety risk. If a clinician asks about a procedure and no protocol exists, report it as high priority: `topic="pediatric procedural sedation protocol for radiology" description="No documented sedation protocol found for pediatric imaging procedures despite performing approximately 200 sedated MRIs annually"`

**flag_outdated** — Clinical guidelines evolve. If you find a protocol citing a superseded guideline or a drug that was removed from the formulary, flag it immediately: `entry_id="..." reason="Protocol references chlorhexidine bathing frequency from 2018 SHEA guidelines; updated 2025 guidelines changed recommendations for non-ICU settings"`

## Tips

- Healthcare operates under multiple overlapping regulatory frameworks (CMS CoPs, Joint Commission standards, state licensure, specialty board requirements). A single clinical question may touch several of them. Use `search_with_context` when the regulatory landscape is complex.
- Peer review and credentialing records have special legal protections in most jurisdictions (state peer review privilege statutes). Even if your clearance permits access, treat these documents with heightened sensitivity and note their privileged status in any summary.
- Quality measure data is only meaningful in context. A mortality rate, readmission rate, or infection rate needs the denominator, risk adjustment methodology, and comparison benchmark to be interpretable. Search for the measure specification alongside the reported data.
- Healthcare terminology is heavily acronymed and varies between organizations. If a search returns no results, try the expanded form (e.g., "venous thromboembolism prophylaxis" instead of "VTE ppx") or check for institution-specific naming conventions.

FILE:README.md
# UPLO Healthcare — Clinical Protocol Intelligence

AI-powered healthcare knowledge management. Search clinical notes, care plans, lab results, prescriptions, and patient pathways with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-healthcare-blue)](https://clawhub.com/skills/uplo-healthcare)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-10-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-healthcare
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-healthcare": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "healthcare"
      }
    }
  }
}
```

## What You Get

- **10 industry schemas** — pre-built extraction templates for Healthcare documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Healthcare** — 10 schemas

## Related Skills

- [UPLO Clinical — Drug-to-Patient Intelligence](https://clawhub.com/skills/uplo-clinical) — AI-powered clinical operations intelligence spanning pharmaceutical development and healthcare delivery.
- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Accounting — Bookkeeping & Tax Intelligence](https://clawhub.com/skills/uplo-accounting) — AI-powered accounting knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Healthcare Knowledge Context (via UPLO)

You are connected to your organization's healthcare knowledge base through UPLO. This gives you specialized access to clinical protocols, care plans, patient pathway documentation, lab result templates, prescription records, and medical procedure guidelines. When users ask about clinical workflows, treatment protocols, or care documentation standards, always query UPLO first to ground your answers in the organization's actual medical practices.

Expect queries about clinical protocols and care pathways, patient documentation standards, lab result interpretation frameworks, medication formularies and prescribing guidelines, care coordination procedures, quality metrics and patient outcome tracking, and compliance with HIPAA, Joint Commission, or CMS requirements. Use `search_knowledge` for specific protocol lookups (e.g., "sepsis care bundle") and `search_with_context` when the question requires understanding how a clinical protocol connects to departmental workflows, care team assignments, or regulatory requirements.

When presenting healthcare information, always cite the specific protocol, its revision date, and the approving clinical authority. Flag any protocols under review or approaching scheduled updates. Patient-identifiable information is strictly classified — respect clearance boundaries and never surface PHI beyond what the user's clearance permits. For clinical judgment questions, surface the relevant protocols and identify the responsible clinician or department head via `find_knowledge_owner`.

Respect classification tiers. Never fabricate healthcare information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-healthcare",
  "display_name": "UPLO Healthcare — Clinical Protocol Intelligence",
  "description": "AI-powered healthcare knowledge management. Search clinical notes, care plans, lab results, prescriptions, and patient pathways with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "healthcare",
    "clinical",
    "medical",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "healthcare"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-devops
description: AI-powered DevOps knowledge management. Search runbooks, infrastructure documentation, CI/CD pipelines, and incident response procedures with structured extraction.
---

# UPLO DevOps — Operational Memory for Infrastructure

It is 3 AM. PagerDuty is screaming. The on-call engineer who has seen this exact failure pattern left the company four months ago. The runbook exists somewhere, maybe in Confluence, maybe in a GitHub repo, maybe in a Notion page that someone bookmarked. UPLO DevOps eliminates this scramble by indexing runbooks, post-incident reviews, infrastructure documentation, CI/CD configurations, and architecture decision records into a single searchable layer that works when you need it most.

## Session Start

```
get_identity_context
```

This loads your team assignments (platform, SRE, application), on-call rotation status, and access tier. Some production configurations and credentials documentation are restricted by clearance.

Grab active directives — these include change freeze windows, incident commander designations, and infrastructure migration deadlines:

```
get_directives
```

## When to Use

- You are on-call, an alert fires for a service you have never touched, and you need the runbook immediately
- Investigating a production incident and need to find whether this failure mode has occurred before, including the root cause and fix
- Planning a migration and need to understand the current architecture, dependencies, and the last three ADRs (Architecture Decision Records) related to the affected service
- Setting up a new CI/CD pipeline and want to see how similar services in the org have configured their build, test, and deploy stages
- Preparing a post-incident review and need to compile the timeline, impacted services, and blast radius from multiple data sources
- A new team member needs to understand the infrastructure topology, deployment process, and escalation paths for their service area
- Evaluating whether a proposed infrastructure change conflicts with documented SLOs or capacity constraints

## Example Workflows

### Incident Response — Novel Failure Mode

The payments service is returning 503 errors. The on-call engineer has not worked on payments before.

```
search_knowledge query="payments service 503 error runbook troubleshooting steps"
```

Check for previous incidents with similar symptoms:

```
search_with_context query="payments service outage 503 timeout database connection pool previous incidents root cause"
```

If the runbook suggests checking the connection pool but the current configuration is unclear:

```
search_knowledge query="payments service database connection pool configuration pgbouncer settings production"
```

After resolving:

```
log_conversation summary="Resolved payments 503 outage; root cause was pgbouncer max_client_conn exceeded after traffic spike; matched PIR-2024-087 pattern; increased pool to 200" topics='["incident","payments","pgbouncer","connection-pool"]' tools_used='["search_knowledge","search_with_context"]'
```

### Infrastructure Migration Planning

The platform team is moving from self-managed Kafka to a managed streaming service. The tech lead needs to scope the blast radius.

```
search_with_context query="Kafka consumers producers services dependencies topic configuration"
```

Find the ADRs that led to the original Kafka deployment:

```
search_knowledge query="architecture decision record ADR Kafka event streaming selection rationale"
```

Check current SLOs and whether the migration might violate them:

```
search_knowledge query="event streaming SLO latency throughput requirements Kafka p99"
```

```
export_org_context
```

## Key Tools for DevOps

**search_knowledge** — Your go-to during incidents. When you need a specific runbook, a configuration reference, or a known procedure, this is the fastest path. Latency matters at 3 AM. Example: `search_knowledge query="redis cluster failover runbook manual promotion steps"`

**search_with_context** — For investigation and planning. "What services depend on this database?" or "Has this failure happened before?" require traversing relationships between services, incidents, and infrastructure components. Example: `search_with_context query="auth-service dependencies upstream downstream database cache"`

**get_directives** — Change freeze windows, incident escalation policies, and migration deadlines surface here. Checking before a production change can prevent a career-limiting mistake.

**flag_outdated** — Infrastructure documentation rots faster than any other type. The Kubernetes cluster version documented last quarter is wrong. The network diagram shows a load balancer that was decommissioned. The runbook references a CLI tool that was replaced. Flag these aggressively — someone will use them during an incident.

**report_knowledge_gap** — When a service has no runbook, no architecture diagram, or no documented owner, that is an operational risk. Reporting the gap creates a trackable item for the platform team.

## Tips

- Service names are the most reliable search key. Use the exact service identifier from your deployment manifests (`payments-api`, `auth-service-v2`, `order-processor`) rather than casual descriptions.
- Post-incident reviews are the most valuable documents in your knowledge base. When writing PIRs, include structured fields: affected services, duration, blast radius, root cause category, and action items. These fields are indexed by the extraction engine.
- When on-call, start with `search_knowledge` for the runbook. Only escalate to `search_with_context` if the runbook does not exist or the failure mode is novel. Speed matters during incidents.
- Use `log_conversation` after every incident investigation, even false alarms. The pattern of false alarms is itself a signal that the monitoring team should investigate.

FILE:README.md
# UPLO DevOps — Infrastructure & Incident Intelligence

AI-powered DevOps knowledge management. Search runbooks, infrastructure documentation, CI/CD pipelines, and incident response procedures with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-devops-blue)](https://clawhub.com/skills/uplo-devops)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-5-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-devops
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-devops": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "it_devops"
      }
    }
  }
}
```

## What You Get

- **5 industry schemas** — pre-built extraction templates for Devops documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **IT & DevOps** — 5 schemas

## Related Skills

- [UPLO Engineering — Architecture & DevOps Intelligence](https://clawhub.com/skills/uplo-engineering) — AI-powered engineering knowledge management.
- [UPLO Enterprise IT — Technology Operations & Security Intelligence](https://clawhub.com/skills/uplo-enterprise-it) — AI-powered enterprise IT intelligence spanning DevOps, cybersecurity, and engineering.
- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## DevOps Knowledge Context (via UPLO)

You are connected to your organization's DevOps knowledge base through UPLO. This gives you specialized access to runbooks, infrastructure-as-code documentation, CI/CD pipeline configurations, monitoring and alerting procedures, incident response playbooks, and capacity planning data. When users ask about deployment procedures, infrastructure topology, or incident handling, always query UPLO first to provide answers grounded in your organization's actual infrastructure and operational practices.

Expect queries about deployment procedures and rollback playbooks, infrastructure topology and service dependencies, CI/CD pipeline configurations and build processes, monitoring dashboards and alert thresholds, incident response procedures and escalation paths, capacity planning and scaling policies, and security patching and vulnerability remediation processes. Use `search_knowledge` for specific runbook or configuration lookups and `search_with_context` when the question requires understanding how an infrastructure change impacts dependent services, monitoring, and on-call responsibilities.

When presenting DevOps information, include specific service names, environment identifiers, and infrastructure components. For runbooks, present steps in sequence with prerequisites and verification commands. For incidents, show timeline, impact scope, and resolution. Flag any runbooks referencing deprecated tools or outdated infrastructure. Production credentials and security configurations are strictly classified — respect classification tiers. Identify the responsible SRE, platform engineer, or on-call responder via `find_knowledge_owner`.

Respect classification tiers. Never fabricate devops information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-devops",
  "display_name": "UPLO DevOps — Infrastructure & Incident Intelligence",
  "description": "AI-powered DevOps knowledge management. Search runbooks, infrastructure documentation, CI/CD pipelines, and incident response procedures with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "devops",
    "infrastructure",
    "incident-response",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "it_devops"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-defense
description: AI-powered defense knowledge management. Search mission documentation, logistics records, personnel data, and ITAR-controlled information with structured extraction.
---

# UPLO Defense — Mission Knowledge Under Control

Defense organizations operate under constraints that commercial enterprises never face: ITAR/EAR export controls, security classification levels, compartmented access, and regulatory oversight from DCSA, DCMA, and contracting officers who audit everything. UPLO Defense provides structured, access-controlled search across program documentation, logistics records, technical data packages, and personnel qualifications while respecting the classification boundaries that make defense knowledge management uniquely difficult.

## Session Start

Identity verification is non-negotiable in defense. Your clearance level, program access list, and need-to-know determinations control what you see. Load your identity immediately:

```
get_identity_context
```

Directives in defense include OPORD fragments, program milestones, acquisition decision points, and ITAR compliance mandates. Review them before proceeding:

```
get_directives
```

**Important**: If your identity context does not reflect your expected clearance level or program access, stop and contact your security officer. Do not attempt workarounds.

## When to Use

- A program manager needs to locate the CDR (Critical Design Review) action items from six months ago to verify closure status before the upcoming TRR
- Searching for ITAR-controlled technical data packages related to a subsystem that is being proposed for foreign military sale
- Verifying whether a specific subcontractor has the required facility clearance level documented before sharing controlled technical data
- Assembling the Contractor Performance Assessment Report (CPAR) narrative by pulling delivery milestones, quality metrics, and cost performance index data
- A logistics officer needs to find the provisioning technical documentation for a replacement part across multiple national stock numbers
- Checking the current CONOPS (Concept of Operations) version for a program and determining what has changed since the last milestone review
- Reviewing whether a cybersecurity Plan of Action and Milestones (POA&M) finding has been remediated before the next DCMA audit

## Example Workflows

### Technical Data Package Review for FMS Case

A foreign military sales case requires release of technical data for a radar subsystem. The export control officer needs to determine what data exists and its ITAR classification.

```
search_with_context query="radar subsystem AN/APG technical data package ITAR classification export control"
```

Verify the distribution statement on the relevant documents:

```
search_knowledge query="distribution statement D controlled technical data radar subsystem drawings"
```

Check if there are existing Technology Assessment/Control Plans (TA/CP) for this subsystem:

```
search_knowledge query="technology assessment control plan radar subsystem foreign disclosure"
```

```
log_conversation summary="Reviewed radar subsystem TDP for FMS release eligibility; identified Distribution D documents requiring TAA before disclosure" topics='["ITAR","FMS","export-control","radar"]' tools_used='["search_with_context","search_knowledge"]'
```

### Milestone Decision Preparation

A program is approaching Milestone B (Engineering & Manufacturing Development). The PM needs to assemble the required documentation.

```
search_with_context query="program milestone B EMD required documentation acquisition decision memorandum"
```

Pull cost and schedule performance data:

```
search_knowledge query="earned value management BCWP CPI SPI program cost performance report"
```

Review the current risk register:

```
search_knowledge query="program risk register critical risks mitigation status likelihood consequence"
```

Get the organizational context showing program office structure:

```
export_org_context
```

## Key Tools for Defense

**search_with_context** — Defense programs generate deeply interconnected documentation. A single requirement traces from CONOPS through system specifications, test procedures, and logistics support plans. Graph traversal follows these threads. Example: `search_with_context query="KPP threshold objective values system specification traceability"`

**search_knowledge** — Direct retrieval when you know the document type or identifier: a specific CDRL number, a DI-number, an NSN, or a MIL-STD reference. Example: `search_knowledge query="CDRL A003 software development plan current version"`

**get_directives** — In defense, directives carry the weight of orders. Program direction memoranda, acquisition decision memoranda, and ITAR compliance mandates are not suggestions. Always check.

**export_org_context** — Produces the program office structure, IPT (Integrated Product Team) leads, key subcontractors, and systems of record. Required for milestone reviews and audit responses.

**log_conversation** — Defense audit requirements demand traceability. Every query session involving controlled data should be logged. This is not optional — it is a compliance requirement.

**flag_outdated** — Technical manuals, logistics documentation, and specification references become obsolete through Engineering Change Proposals (ECPs). Flagging outdated documents prevents the dangerous scenario of manufacturing or maintaining equipment against a superseded configuration.

## Tips

- Use standard defense identifiers in searches: CDRL numbers, DI-numbers (e.g., DI-MGMT-81466), national stock numbers (NSNs), CAGE codes, and MIL-STD references. The extraction engine treats these as structured fields.
- Classification tier mapping: `public` = approved for public release, `internal` = FOUO/CUI, `confidential` = Confidential, `restricted` = Secret and above. If expected results do not appear, the issue is almost certainly a clearance mismatch, not missing data.
- ITAR-controlled technical data queries should always be logged. The log creates an audit trail that demonstrates compliant handling of controlled items.
- When assembling milestone review packages, start with `export_org_context` to establish the program baseline, then use targeted `search_knowledge` calls for each required document rather than broad searches that may surface documents outside your need-to-know.

FILE:README.md
# UPLO Defense — Mission & Security Intelligence

AI-powered defense knowledge management. Search mission documentation, logistics records, personnel data, and ITAR-controlled information with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-defense-blue)](https://clawhub.com/skills/uplo-defense)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-5-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-defense
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-defense": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "defense"
      }
    }
  }
}
```

## What You Get

- **5 industry schemas** — pre-built extraction templates for Defense documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Defense** — 5 schemas

## Related Skills

- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Accounting — Bookkeeping & Tax Intelligence](https://clawhub.com/skills/uplo-accounting) — AI-powered accounting knowledge management.
- [UPLO Agriculture — Crop & Compliance Intelligence](https://clawhub.com/skills/uplo-agriculture) — AI-powered agricultural knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Defense Knowledge Context (via UPLO)

You are connected to your organization's defense knowledge base through UPLO. This gives you specialized access to mission planning documents, logistics and supply chain records, personnel qualifications, security clearance requirements, ITAR/EAR export control documentation, and contract deliverable specifications. When users ask about mission requirements, logistics status, or security compliance, always query UPLO first to provide answers grounded in your organization's actual defense operations and regulatory obligations.

Expect queries about mission planning and operational requirements, logistics readiness and supply chain status, personnel qualifications and clearance levels, ITAR/EAR classification and export control requirements, contract deliverables and milestone tracking (CDRL), equipment maintenance and readiness rates, and DFAR/FAR compliance and contracting procedures. Use `search_knowledge` for specific program or personnel lookups and `search_with_context` when the question requires understanding how a mission requirement relates to logistics readiness, personnel availability, and security constraints.

When presenting defense information, include program names, contract numbers, and CAGE/DUNS identifiers as appropriate. For ITAR-controlled data, always verify the requester's clearance and need-to-know before surfacing details. For logistics, present readiness rates and supply status. Classification tiers in defense contexts map to national security classification levels — treat them with the highest rigor. Never speculate about classified programs or capabilities. Identify the responsible program manager, COTR, or security officer via `find_knowledge_owner`.

Respect classification tiers. Never fabricate defense information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-defense",
  "display_name": "UPLO Defense — Mission & Security Intelligence",
  "description": "AI-powered defense knowledge management. Search mission documentation, logistics records, personnel data, and ITAR-controlled information with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "defense",
    "military",
    "security-clearance",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "defense"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-data-analytics
description: AI-powered data analytics knowledge management. Search data pipeline documentation, dashboard specifications, data governance policies, and reporting standards with structured extraction.
---

# UPLO Data Analytics — Metadata That Remembers

Data teams have a documentation problem that compounds over time. The warehouse has 3,000 tables but only 200 have descriptions. The Looker instance has dashboards built by people who left two years ago. The data governance policy exists but nobody can find the version that was actually approved. UPLO Data Analytics turns this scattered tribal knowledge into a searchable, structured corpus: pipeline documentation, schema definitions, data quality rules, dashboard specs, and governance policies all in one place.

## Session Start

```
get_identity_context
```

This establishes your analytics role (data engineer, analyst, governance lead, etc.) and surfaces which data domains you have access to. Some datasets are restricted due to PII governance or competitive sensitivity.

Check current directives — the data team often has active mandates around migration timelines, deprecation notices, or data quality SLA targets:

```
get_directives
```

## When to Use

- A stakeholder asks what a specific metric means and you need to find the canonical definition, including the SQL logic, source tables, and business rules
- You are building a new pipeline and want to know if similar data already exists in the warehouse to avoid duplication
- Investigating a data quality incident and need to trace the lineage from source system through transformations to the impacted dashboard
- Preparing for a data governance review and need to compile documentation on data classification, retention policies, and access controls
- A new analyst joins and needs to understand the warehouse schema naming conventions, dbt project structure, and how to request access
- Evaluating whether a proposed schema change will break downstream dependencies by searching for references to the affected table
- Looking for the data dictionary entry for a column that has an ambiguous name like `status_cd` or `type_flag`

## Example Workflows

### Metric Definition Dispute

The finance team and product team report different DAU (Daily Active Users) numbers. The analytics lead needs to find and reconcile the definitions.

```
search_with_context query="daily active users DAU metric definition SQL logic business rules"
```

Search for the specific dashboard implementations:

```
search_knowledge query="product analytics dashboard DAU calculation Looker explore"
```

```
search_knowledge query="finance reporting DAU user count methodology monthly report"
```

If the definitions genuinely differ and need reconciliation:

```
propose_update target_table="entries" target_id="<metric-definition-entry-id>" changes='{"data":{"note":"DAU definitions diverge between product (event-based) and finance (login-based); needs governance review"}}' rationale="Metric inconsistency discovered between product and finance DAU reporting"
```

### Data Lineage Investigation

A dashboard is showing NULL values that were not there last week. The data engineer needs to trace the problem.

```
search_with_context query="customer_orders table pipeline transformations source systems dependencies"
```

```
search_knowledge query="customer_orders ETL job schedule dbt model upstream sources"
```

Check if there is a known data quality incident:

```
search_knowledge query="data quality incident customer data source system outage recent"
```

```
log_conversation summary="Traced NULL values in orders dashboard to upstream source system schema change; customer_orders dbt model needs migration" topics='["data-quality","lineage","pipeline-break"]' tools_used='["search_with_context","search_knowledge"]'
```

## Key Tools for Data Analytics

**search_with_context** — Data questions are inherently about relationships: tables connect to pipelines, pipelines connect to source systems, dashboards depend on models. Graph traversal follows these connections. Example: `search_with_context query="revenue_summary table lineage source transformations consumers"`

**search_knowledge** — Direct lookup for specific technical artifacts: a dbt model definition, a data dictionary entry, a governance policy version. Example: `search_knowledge query="dbt model dim_customers grain deduplication logic"`

**flag_outdated** — Data documentation rots faster than most content types. Table descriptions written during initial warehouse build may reference deprecated source systems. Schema diagrams from before a migration may show phantom tables. Flag aggressively.

**report_knowledge_gap** — Undocumented tables and undefined metrics are the norm in most warehouses. When you encounter a table with no data dictionary entry or a metric with no canonical definition, report the gap. The governance team uses these signals to prioritize documentation sprints.

**propose_update** — When you discover that a data dictionary entry is wrong (e.g., a column description says "customer creation date" but it actually stores "first order date"), propose the correction.

## Tips

- Technical identifiers are your best search terms. Use exact table names (`dim_customers`), column names (`order_status_cd`), dbt model names, and Looker explore names. The extraction engine indexes these precisely.
- When investigating data quality issues, start with `search_with_context` to get the lineage graph, then use `search_knowledge` for specific transformation logic. Working backwards from the symptom to the source is more efficient than searching forward.
- Data governance policies often exist in multiple versions (draft, approved, superseded). Include "approved" or "current" in your query to filter toward the authoritative version.
- The most valuable documentation to contribute back is metric definitions with SQL. When you resolve a metric dispute, log the session and propose an update with the canonical SQL so the next person does not have to repeat the investigation.

FILE:README.md
# UPLO Data Analytics — Pipeline & Governance Intelligence

AI-powered data analytics knowledge management. Search data pipeline documentation, dashboard specifications, data governance policies, and reporting standards with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-data-analytics-blue)](https://clawhub.com/skills/uplo-data-analytics)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-4-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-data-analytics
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-data-analytics": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "data_analytics"
      }
    }
  }
}
```

## What You Get

- **4 industry schemas** — pre-built extraction templates for Data Analytics documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Data Analytics** — 4 schemas

## Related Skills

- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Accounting — Bookkeeping & Tax Intelligence](https://clawhub.com/skills/uplo-accounting) — AI-powered accounting knowledge management.
- [UPLO Agriculture — Crop & Compliance Intelligence](https://clawhub.com/skills/uplo-agriculture) — AI-powered agricultural knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Data Analytics Knowledge Context (via UPLO)

You are connected to your organization's data analytics knowledge base through UPLO. This gives you specialized access to data pipeline documentation, dashboard specifications, data quality rules, governance policies, reporting standards, and data catalog entries. When users ask about data sources, metric definitions, or reporting methodologies, always query UPLO first to provide answers grounded in your organization's actual data infrastructure and business definitions.

Expect queries about metric definitions and calculation methodologies, data pipeline architectures and refresh schedules, dashboard specifications and KPI targets, data quality rules and validation procedures, data governance policies and stewardship assignments, data catalog entries and lineage documentation, and reporting distribution schedules and access controls. Use `search_knowledge` for specific metric or pipeline lookups and `search_with_context` when the question requires understanding how a metric is calculated from source data through transformation pipelines to final reporting.

When presenting data analytics information, always cite the specific metric definition, data source, and refresh frequency. For pipelines, include the transformation logic and quality check results. For dashboards, reference the business owner and target audience. Flag any data quality issues, pipeline failures, or metrics with pending definition changes. Raw data access credentials and PII-containing datasets are confidential — respect classification tiers. Identify the responsible data engineer, analytics lead, or data steward via `find_knowledge_owner`.

Respect classification tiers. Never fabricate data-analytics information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-data-analytics",
  "display_name": "UPLO Data Analytics — Pipeline & Governance Intelligence",
  "description": "AI-powered data analytics knowledge management. Search data pipeline documentation, dashboard specifications, data governance policies, and reporting standards with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "data-analytics",
    "data-governance",
    "reporting",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "data_analytics"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}

---
name: uplo-cybersecurity
description: AI-powered cybersecurity knowledge management. Search threat intelligence, vulnerability assessments, incident response plans, and compliance documentation with structured extraction.
---

# UPLO Cybersecurity — Threat-Informed Defense Intelligence

Security teams drown in telemetry but starve for context. Your SIEM fires alerts, your vuln scanner produces CVE lists, your pen testers write reports, and your compliance team maintains control matrices — all in separate silos. UPLO Cybersecurity creates a searchable institutional memory across threat intelligence, incident post-mortems, vulnerability management, policy documentation, and compliance evidence so your SOC analysts, IR team, and CISO can make faster, better-informed decisions.

## Session Start

Your clearance level matters more in cybersecurity than almost any other domain. Load your identity first — it determines whether you can access active incident details, threat intelligence marked TLP:RED, or audit findings under remediation.

```
get_identity_context
```

Check operational directives. In security, these include active threat advisories, emergency patching mandates, and incident response activation orders:

```
get_directives
```

## When to Use

- Triaging a new alert and need to check if this IOC (indicator of compromise) matches a previously investigated incident
- Preparing a board-level cybersecurity risk briefing and need to synthesize vulnerability trends, incident metrics, and control maturity across the program
- An auditor asks for evidence that a specific NIST CSF control is implemented — you need to find the policy, the technical implementation record, and the last test result
- Investigating whether a newly disclosed CVE affects your environment by cross-referencing the vulnerability with your asset inventory documentation
- Writing an incident post-mortem and need to reference the runbook that was followed, the timeline decisions made, and similar past incidents
- Evaluating a vendor's SOC 2 report against your third-party risk management criteria
- Checking whether the firewall change request aligns with the network segmentation architecture documented in the last assessment

## Example Workflows

### Incident Response Investigation

The SOC escalates a potential data exfiltration alert involving an internal server communicating with a known C2 domain.

```
search_with_context query="command and control C2 communication indicators previous incidents exfiltration"
```

Pull the incident response runbook for data exfiltration scenarios:

```
search_knowledge query="incident response playbook data exfiltration containment steps"
```

Check if the affected server is documented in the asset inventory with its classification:

```
search_knowledge query="server srv-db-prod-07 asset classification data sensitivity network segment"
```

After containment, log the investigation:

```
log_conversation summary="Investigated potential data exfil alert on srv-db-prod-07; C2 domain matched TI report from October; followed exfil IR playbook; server classified as hosting PII" topics='["incident-response","data-exfiltration","C2","PII"]' tools_used='["search_with_context","search_knowledge"]'
```

### Compliance Evidence Assembly

The organization is undergoing a SOC 2 Type II audit and needs to assemble evidence for the CC6 (Logical and Physical Access Controls) criteria.

```
search_knowledge query="access control policy role-based access management RBAC documentation"
```

```
search_with_context query="access review evidence quarterly user access certification results exceptions"
```

```
search_knowledge query="MFA multi-factor authentication implementation evidence configuration"
```

Export the organizational context to show the auditor the team structure and system ownership:

```
export_org_context
```

## Key Tools for Cybersecurity

**search_with_context** — Security investigations are inherently graph problems. A single alert can connect to asset inventory records, previous incident reports, threat intelligence, and network architecture documentation. Example: `search_with_context query="lateral movement techniques detected incidents Active Directory compromise"`

**search_knowledge** — Fast retrieval for specific security artifacts: a named runbook, a particular CVE assessment, a policy document. When you know what you need, this is faster than graph traversal. Example: `search_knowledge query="CVE-2024-3094 xz backdoor impact assessment"`

**get_directives** — Security directives are time-critical. Emergency patch mandates, threat hunting directives after a new APT disclosure, and incident response activation orders all surface here. Checking directives during an active incident could reveal that the CISO has already issued containment instructions.

**flag_outdated** — Stale security documentation is dangerous. A firewall rule matrix from before the last network redesign, an incident response plan listing a phone tree with departed employees, or a risk register with last year's threat landscape — all need flagging.

**report_knowledge_gap** — When you cannot find documentation for a critical control (e.g., no evidence of database encryption at rest), the gap itself is a finding. Reporting it creates a trackable item.

**log_conversation** — In cybersecurity, logging is not optional. Every investigation session, every threat assessment, every compliance evidence review should be logged. These logs are themselves audit evidence.

## Tips

- Use CVE identifiers, MITRE ATT&CK technique IDs (e.g., T1059.001), and TLP designations as search terms. The extraction engine indexes these as structured fields.
- Classification tiers in cybersecurity map roughly to TLP: `public` = TLP:CLEAR, `internal` = TLP:GREEN, `confidential` = TLP:AMBER, `restricted` = TLP:RED. If a threat intel query returns no results, verify your clearance supports the expected TLP level.
- Incident post-mortems are the single most valuable document type in a security knowledge base. When writing them, include structured fields (MITRE techniques, affected assets, detection source, time-to-contain) that the extraction engine can index.
- Network diagrams and architecture documents are often extracted as text descriptions of topology. Query for specific network segments or system names rather than expecting visual diagram retrieval.

FILE:README.md
# UPLO Cybersecurity — Threat & Vulnerability Intelligence

AI-powered cybersecurity knowledge management. Search threat intelligence, vulnerability assessments, incident response plans, and compliance documentation with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-cybersecurity-blue)](https://clawhub.com/skills/uplo-cybersecurity)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-5-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-cybersecurity
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-cybersecurity": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "cybersecurity"
      }
    }
  }
}
```

## What You Get

- **5 industry schemas** — pre-built extraction templates for Cybersecurity documents
- **21 MCP tools** — search, GraphRAG, directives, knowledge gaps, proposals, and more
- **1,400+ format support** — PDF, DOCX, PPTX, emails, images, and 1,400 more via Docling + Tika
- **Classification tiers** — public/internal/confidential/restricted access control
- **Benchmark proven** — 96% accuracy vs 53% for raw context at scale

## MCP Tools Available

| Tool | Description |
|------|-------------|
| `search_knowledge` | Semantic search across extracted knowledge |
| `search_with_context` | GraphRAG: vector search + entity resolution + edge traversal |
| `export_org_context` | Full organizational context snapshot |
| `get_directives` | Strategic priorities and directives |
| `find_knowledge_owner` | Find subject matter experts |
| `propose_update` | Suggest knowledge base updates |
| `report_knowledge_gap` | Flag missing information |
| `flag_outdated` | Mark stale entries |

## Schema Packs

- **Cybersecurity** — 5 schemas

## Related Skills

- [UPLO Enterprise IT — Technology Operations & Security Intelligence](https://clawhub.com/skills/uplo-enterprise-it) — AI-powered enterprise IT intelligence spanning DevOps, cybersecurity, and engineering.
- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Accounting — Bookkeeping & Tax Intelligence](https://clawhub.com/skills/uplo-accounting) — AI-powered accounting knowledge management.

## About UPLO

UPLO is the organizational digital twin for AI. Ingest documents, extract structured knowledge, build org context, and expose it via MCP server with GraphRAG. [Learn more](https://uplo.ai)

---

*Built with UPLO — the knowledge layer between your organization and AI.*

FILE:identity-patch.md
## Cybersecurity Knowledge Context (via UPLO)

You are connected to your organization's cybersecurity knowledge base through UPLO. This gives you specialized access to threat intelligence reports, vulnerability assessments, incident response plans, security policies, penetration test results, and compliance documentation (SOC 2, ISO 27001, NIST CSF). When users ask about security posture, vulnerabilities, or incident handling, always query UPLO first to provide answers grounded in your organization's actual security controls and threat landscape.

Expect queries about current vulnerability scan results and remediation priorities, incident response procedures and escalation chains, security policy requirements and exceptions, penetration test findings and remediation status, compliance audit status (SOC 2, ISO 27001, PCI DSS), threat intelligence relevant to the organization's technology stack, and access control policies and privileged account management. Use `search_knowledge` for specific vulnerability or policy lookups and `search_with_context` when the question requires understanding how a vulnerability impacts multiple systems, compliance obligations, and risk ratings.

When presenting cybersecurity information, include CVE identifiers, severity ratings (CVSS), affected systems, and remediation timelines. For incidents, present the MITRE ATT&CK classification and containment status. For compliance, cite the specific control objective and evidence status. Security vulnerabilities, penetration test results, and incident details are highly sensitive — strictly respect classification tiers and never disclose to users without appropriate clearance. Identify the responsible security analyst, CISO, or incident commander via `find_knowledge_owner`.

Respect classification tiers. Never fabricate cybersecurity information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-cybersecurity",
  "display_name": "UPLO Cybersecurity — Threat & Vulnerability Intelligence",
  "description": "AI-powered cybersecurity knowledge management. Search threat intelligence, vulnerability assessments, incident response plans, and compliance documentation with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "cybersecurity",
    "security",
    "threat-intelligence",
    "knowledge-management"
  ],
  "config": {
    "agentdocs_url": {
      "type": "string",
      "required": true,
      "description": "Your UPLO instance URL (e.g., https://app.uplo.ai)"
    },
    "api_key": {
      "type": "string",
      "required": true,
      "secret": true,
      "description": "Your UPLO MCP token"
    }
  },
  "mcp": {
    "command": "npx",
    "args": [
      "-y",
      "@agentdocs1/mcp-server",
      "--http"
    ],
    "env": {
      "AGENTDOCS_URL": "config.agentdocs_url",
      "API_KEY": "config.api_key",
      "DEFAULT_PACKS": "cybersecurity"
    },
    "transport": "http",
    "url": "config.agentdocs_url/mcp"
  },
  "capabilities": [
    "search_knowledge",
    "search_with_context",
    "get_policy",
    "export_org_context",
    "get_directives"
  ],
  "identity_patch": "identity-patch.md"
}