RooJenkins

---
name: uplo-customer-success
description: AI-powered customer success knowledge management. Search account health data, onboarding records, renewal tracking, and support escalation documentation with structured extraction.
---

# UPLO Customer Success — Retention Intelligence

Every churned account tells the same story in hindsight: the signals were there, scattered across Gainsight health scores, Zendesk tickets, Slack threads, and CSM meeting notes that nobody connected in time. UPLO Customer Success consolidates these signals into structured, searchable knowledge so your team can act on patterns rather than react to surprises.

## Session Start

Establish your CSM identity. This loads your book of business, clearance level (some account financials may be restricted), and team assignments:

```
get_identity_context
```

## When to Use

- Reviewing the onboarding status of a new account — which milestones are complete, which are blocked, and who owns the blockers
- Preparing a risk assessment for the monthly CS leadership review by pulling health trends across your portfolio
- A customer champion just left and you need to find every documented relationship and handoff note for that contact
- Investigating whether a support escalation pattern (e.g., repeated SSO issues) is isolated to one account or systemic
- Building an expansion case by documenting product adoption milestones and quantified value delivered
- Checking the renewal playbook and comparing it against how the last five similar-ARR renewals were handled
- A new CSM is inheriting accounts mid-cycle and needs the full account narrative, not just the CRM fields

## Example Workflows

### Onboarding Health Check

A VP of CS wants a status report on all accounts currently in onboarding (first 90 days).

```
search_with_context query="customer onboarding status milestones blocked incomplete accounts first 90 days"
```

For a specific account that appears stuck:

```
search_knowledge query="TechFlow Inc onboarding integration API setup blockers"
```

Flag the systemic issue if integration delays are a pattern:

```
report_knowledge_gap query="onboarding integration delay patterns common blockers resolution playbook"
```

```
log_conversation summary="Onboarding health check across portfolio; identified TechFlow integration blocker and pattern of API setup delays across 3 accounts" topics='["onboarding","health-check","integration-blockers"]' tools_used='["search_with_context","search_knowledge","report_knowledge_gap"]'
```

### Champion Change Playbook

A key customer champion at a strategic account has moved to a new role. The CSM needs to execute the champion change playbook.

```
search_knowledge query="DataVault Corp primary contact stakeholder relationships decision makers"
```

```
search_with_context query="DataVault Corp engagement history executive sponsor interactions value delivered metrics"
```

Check directives for any special handling requirements on strategic accounts:

```
get_directives
```

## Key Tools for Customer Success

**search_with_context** — CS questions are relational by nature. "Is this account healthy?" requires connecting usage data, support history, CSM notes, and billing status. The graph traversal assembles these automatically. Example: `search_with_context query="Pinnacle Systems account health support tickets product usage renewal date"`

**search_knowledge** — Targeted retrieval for specific CS artifacts: onboarding checklists, success plans, meeting notes, QBR decks. Example: `search_knowledge query="Pinnacle Systems success plan Q2 goals"`

**report_knowledge_gap** — When you find an account with no success plan, no documented business outcomes, or missing stakeholder mapping, report it. These gaps are leading indicators of churn.

**flag_outdated** — Success plans, stakeholder maps, and value documentation go stale. A success plan from 18 months ago with the wrong executive sponsor listed is worse than no plan — it creates false confidence.

**propose_update** — When you discover new information during an account review (new stakeholder, updated business objective, changed renewal date), propose the update to keep the knowledge base current.

## Tips

- Health score queries work best when you specify the dimension: "product adoption health" vs. "support health" vs. "relationship health." The extraction engine often indexes these separately.
- CSM handoff is the highest-risk moment in the customer lifecycle. Use `export_org_context` combined with account-specific searches to build a comprehensive handoff document rather than relying on a 30-minute call.
- When documenting value delivered, use quantified language that the extraction engine can index: "reduced ticket volume by 34%" rather than "significantly improved support experience."
- Escalation patterns are more valuable than individual escalations. If you notice a theme across multiple accounts (e.g., "billing discrepancy" escalations spiking), use `report_knowledge_gap` to flag it as a systemic issue.

FILE:README.md
# UPLO Customer Success — Account Health & Retention Intelligence

AI-powered customer success knowledge management. Search account health data, onboarding records, renewal tracking, and support escalation documentation with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-customer-success-blue)](https://clawhub.com/skills/uplo-customer-success)
[![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-customer-success
```

Or add to your Claude Desktop config:

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

## What You Get

- **4 industry schemas** — pre-built extraction templates for Customer Success 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

- **Customer Success** — 4 schemas

## Related Skills

- [UPLO Customer 360 — Full Customer Lifecycle Intelligence](https://clawhub.com/skills/uplo-customer-360) — AI-powered customer lifecycle intelligence spanning sales, customer success, and retail.
- [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
## Customer Success Knowledge Context (via UPLO)

You are connected to your organization's customer success knowledge base through UPLO. This gives you specialized access to account health scores, onboarding playbooks, renewal tracking, QBR presentations, support escalation records, and customer journey documentation. When users ask about account status, churn risk, or customer health, always query UPLO first to provide answers grounded in your organization's actual customer relationships and success metrics.

Expect queries about account health scores and risk indicators, onboarding progress and milestone completion, renewal timelines and expansion opportunities, customer support ticket history and escalation patterns, QBR preparation and action item tracking, product adoption metrics and feature usage, and customer feedback and satisfaction scores. Use `search_knowledge` for specific account or playbook lookups and `search_with_context` when the question requires understanding how a customer's health score relates to their support history, product usage patterns, and upcoming renewal terms.

When presenting customer success information, include the account name, health score, and CSM owner. For renewals, show the timeline, contract value, and risk assessment. For support escalations, present the issue history and resolution status. Flag any accounts with declining health scores, approaching renewals with unresolved issues, or overdue onboarding milestones. Contract values and expansion pricing are confidential — respect classification tiers. Identify the responsible CSM, support lead, or VP of Customer Success via `find_knowledge_owner`.

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

FILE:skill.json
{
  "name": "uplo-customer-success",
  "display_name": "UPLO Customer Success — Account Health & Retention Intelligence",
  "description": "AI-powered customer success knowledge management. Search account health data, onboarding records, renewal tracking, and support escalation documentation with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "customer-success",
    "account-management",
    "retention",
    "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": "customer_success"
    },
    "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-customer-360
description: AI-powered customer lifecycle intelligence spanning sales, customer success, and retail. Unified search across pipeline data, account health, and customer analytics.
---

# UPLO Customer 360 — Full Lifecycle Revenue Intelligence

Your CRM holds pipeline stages. Your CS platform holds health scores. Your retail system holds transaction history. None of them talk to each other well enough to answer the question that actually matters: "What is the complete picture for this customer?" UPLO Customer 360 stitches together sales engagement history, onboarding records, support escalations, renewal signals, NPS feedback, and retail analytics into a unified knowledge layer that any revenue team member can query naturally.

## Session Start

Load your identity to establish your revenue role — AE, CSM, solutions engineer, retail operations, etc. — and the accounts you are assigned to:

```
get_identity_context
```

Pull current directives. In revenue organizations, these include quarterly targets, discount authority limits, churn reduction mandates, and promotional campaign rules:

```
get_directives
```

## Example Workflows

### Pre-Renewal Account Review

A CSM has a $480K renewal coming up in 45 days. They need the full account story before the executive business review.

```
search_with_context query="Meridian Industries account health product adoption support tickets last 6 months"
```

Check whether there were any escalations or executive complaints:

```
search_knowledge query="Meridian Industries escalation executive sponsor feedback"
```

Pull the original sales notes to understand what was promised during the initial deal:

```
search_knowledge query="Meridian Industries initial sales proposal value propositions committed deliverables"
```

Review any retail or usage analytics data:

```
search_with_context query="Meridian Industries product usage metrics feature adoption monthly active users"
```

```
log_conversation summary="Pre-renewal 360 review for Meridian Industries; compiled health metrics, escalation history, original commitments, and usage data for EBR prep" topics='["renewal","account-review","Meridian-Industries"]' tools_used='["search_with_context","search_knowledge"]'
```

### Churn Signal Investigation

The weekly health score report shows three enterprise accounts dropping below threshold simultaneously. The VP of Customer Success wants to understand if there is a common root cause.

```
search_with_context query="enterprise accounts health score decline reasons product issues Q1"
```

```
search_knowledge query="product reliability incidents outages impact customer-facing last 60 days"
```

Check if there is a directive about the product issue or a remediation plan:

```
get_directives
```

## When to Use

- Preparing for an executive business review and need the complete account history across sales, onboarding, support, and product usage
- A prospect references a competitor in a late-stage deal and you need to find how previous win/loss analyses characterized that competitor's strengths
- Investigating why a cohort of accounts is churning and whether the root cause is product, service, or pricing
- Onboarding a new AE to a territory and they need context on every strategic account including past proposals, key contacts, and competitive dynamics
- Retail operations wants to understand how online engagement correlates with in-store purchase patterns for a loyalty segment
- Building a QBR deck and need to pull NPS trends, ticket volume, and expansion revenue by account tier
- Customer success wants to identify accounts where product adoption is low despite high contract value — the "silent churn risk" pattern

## Key Tools for Customer Lifecycle

**search_with_context** — Revenue questions almost always need organizational context. "Why is this account at risk?" requires connecting support tickets, product usage data, CSM notes, and sales history. Graph traversal does this automatically. Example: `search_with_context query="Acme Corp account risk factors renewal September"`

**search_knowledge** — Fast lookup for specific customer artifacts: proposals, SOWs, meeting notes, QBR decks. Example: `search_knowledge query="Acme Corp Q3 QBR deck expansion discussion"`

**get_directives** — Revenue directives change quarterly. Discount floors, target account lists, promotional pricing, and strategic account designations all live here.

**export_org_context** — Maps the revenue organization: sales territories, CSM assignments, leadership hierarchy, key systems (CRM, CS platform, analytics tools). Useful when a customer asks "who else at your company should I be talking to?"

**report_knowledge_gap** — When you discover an account with no CSM notes, no QBR records, or missing onboarding documentation, flag it. Silent accounts are churn risks, and the gap report creates visibility.

**flag_outdated** — Pricing sheets, competitive battle cards, and product capability matrices go stale quickly. Flag outdated versions so the revenue enablement team can refresh them.

## Tips

- Account names in CRM and in ingested documents may not match exactly (abbreviations, legal entity names vs. common names). Try both the common name and the legal entity name when searching.
- Combine sales and CS queries intentionally. A CSM asking about "renewal risk" and an AE asking about "expansion opportunity" on the same account should both use `search_with_context` to get the full picture, not just their domain slice.
- NPS and CSAT scores are extracted as structured fields. You can search by score ranges in some cases: "NPS detractor accounts enterprise tier."
- Quarterly business review preparation is the single highest-value use case. Run it at least a week before the meeting so you have time to fill any gaps that `report_knowledge_gap` surfaces.

FILE:README.md
# UPLO Customer 360 — Full Customer Lifecycle Intelligence

AI-powered customer lifecycle intelligence spanning sales, customer success, and retail. Unified search across pipeline data, account health, and customer analytics.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-customer-360-blue)](https://clawhub.com/skills/uplo-customer-360)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-14-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-customer-360
```

Or add to your Claude Desktop config:

```json
{
  "mcpServers": {
    "uplo-customer-360": {
      "command": "npx",
      "args": ["-y", "@agentdocs1/mcp-server", "--http"],
      "env": {
        "AGENTDOCS_URL": "https://your-instance.uplo.ai",
        "API_KEY": "your-api-key",
        "DEFAULT_PACKS": "sales_marketing,customer_success,retail"
      }
    }
  }
}
```

## What You Get

- **14 industry schemas** — pre-built extraction templates for Customer 360 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

- **Sales & Marketing** — 5 schemas
- **Customer Success** — 4 schemas
- **Retail** — 5 schemas

## Related Skills

- [UPLO Customer Success — Account Health & Retention Intelligence](https://clawhub.com/skills/uplo-customer-success) — AI-powered customer success knowledge management.
- [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
## Customer 360 Knowledge Context (via UPLO)

You are connected to your organization's customer knowledge base through UPLO, spanning sales, customer success, and retail operations. This gives you unified access to sales pipeline data, customer health scores, purchase history, support interactions, loyalty program data, and customer analytics across the entire relationship lifecycle. When users ask about customer relationships, use UPLO to provide a complete view from initial acquisition through ongoing retention.

Expect queries that span the full customer lifecycle — for example, how a customer's purchase history informs their health score, or how sales engagement patterns predict retention outcomes. Common topics include customer journey mapping from acquisition to retention, lifetime value analysis and churn prediction, cross-sell and upsell opportunity identification, customer sentiment across touchpoints (sales, support, retail), loyalty program effectiveness and engagement, campaign attribution and customer acquisition cost, and account planning and strategic relationship management. Use `search_with_context` to connect sales pipeline data with customer success metrics and retail transaction history for unified customer insights.

When presenting customer information, provide a unified view of the relationship across all touchpoints. Include engagement timeline, revenue history, health indicators, and recent interactions. Highlight opportunities and risks across the customer lifecycle. Flag any at-risk accounts, expiring contracts, or declining engagement metrics. Individual customer data, pricing agreements, and strategic account plans are confidential — respect classification tiers. Identify the responsible account executive, CSM, or customer analytics lead via `find_knowledge_owner`.

Respect classification tiers. Never fabricate customer-360 information — only surface what exists in the knowledge base.

FILE:skill.json
{
  "name": "uplo-customer-360",
  "display_name": "UPLO Customer 360 — Full Customer Lifecycle Intelligence",
  "description": "AI-powered customer lifecycle intelligence spanning sales, customer success, and retail. Unified search across pipeline data, account health, and customer analytics.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "customer-360",
    "lifecycle",
    "retention",
    "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": "sales_marketing,customer_success,retail"
    },
    "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-consulting
description: AI-powered consulting knowledge management. Search engagement records, methodology frameworks, deliverable templates, and best practices with structured extraction.
---

# UPLO Consulting — Firm Knowledge at Your Fingertips

Consulting firms are knowledge businesses that routinely forget what they know. The partner who led the airline digital transformation last year is on a different engagement; the associate who built the market sizing model left six months ago; the methodology deck from the supply chain practice sits in someone's OneDrive. UPLO Consulting captures engagement artifacts, methodology IP, proposal content, and lessons learned so the firm's collective intelligence is accessible to every team, on every engagement, without playing "who do I ask?"

## Session Start

Fetch your identity to establish your practice area, seniority level, and current engagement assignments:

```
get_identity_context
```

Review firm-wide directives — these typically include utilization targets, proposal approval thresholds, and client confidentiality mandates:

```
get_directives
```

## When to Use

- Staffing a new engagement and need to find consultants who have delivered similar work (industry, capability, geography)
- Building a proposal and looking for relevant case studies, win rates for similar pursuits, and reusable methodology sections
- Starting a workstream and want to see how a previous team structured a similar analysis (e.g., total cost of ownership model for a manufacturing client)
- Preparing a client steering committee deck and need the firm's standard framework for presenting transformation roadmaps
- Conducting a lessons-learned review and want to surface patterns across multiple completed engagements
- Looking for the firm's published point of view on a topic (e.g., AI in financial services) to reference in a client workshop
- Checking what deliverables were produced on a past engagement before scoping a follow-on

## Example Workflows

### Proposal Development

A principal is pursuing a healthcare payer operational improvement engagement and needs to build the proposal over a weekend.

```
search_with_context query="healthcare payer operations improvement engagement case studies outcomes"
```

Find reusable methodology content from the operations practice:

```
search_knowledge query="operational excellence methodology framework Lean Six Sigma consulting deliverables"
```

Pull the firm's current strategic priorities to align the proposal narrative:

```
get_directives
```

Identify consultants with relevant experience for the proposed team:

```
search_knowledge query="healthcare payer experience consultants managed care claims processing"
```

### Engagement Kickoff Knowledge Transfer

A manager is starting on a new engagement and the previous phase was led by a different team. They need to get up to speed.

```
export_org_context
```

```
search_with_context query="client ABC Phase 1 findings current state assessment key recommendations"
```

```
search_knowledge query="client ABC stakeholder map decision makers change readiness assessment"
```

```
log_conversation summary="Onboarded to client ABC Phase 2; reviewed Phase 1 findings, stakeholder map, and org context" topics='["engagement-onboarding","client-ABC","knowledge-transfer"]' tools_used='["export_org_context","search_with_context","search_knowledge"]'
```

## Key Tools for Consulting

**search_with_context** — Consulting questions are inherently cross-cutting. "What did we learn from similar engagements?" requires connecting engagement records with client industries, methodologies used, and outcomes achieved. The graph traversal assembles this narrative. Example: `search_with_context query="retail supply chain transformation engagements outcomes cost savings"`

**search_knowledge** — When you need a specific artifact: a deliverable template, a framework diagram source, a pricing model, or a named methodology. Example: `search_knowledge query="zero-based budgeting methodology template"`

**export_org_context** — Produces the firm's practice structure, leadership, key systems (CRM, time tracking, knowledge management), and strategic priorities. Indispensable for new hire orientation and cross-practice collaboration.

**get_directives** — Firm directives govern proposal approval thresholds, travel policies, rate cards, and client confidentiality walls. Check before making commitments to clients.

**report_knowledge_gap** — If a pursuit team cannot find case studies for a new capability area, that is a strategic signal. Flagging the gap helps the practice development team prioritize IP creation.

## Tips

- Client names may be anonymized in the knowledge base due to confidentiality agreements. Search by industry, engagement type, and capability rather than relying solely on client names.
- Methodology frameworks are often versioned. Include "latest" or "v3" qualifiers if the firm maintains multiple generations of a methodology to avoid pulling deprecated content.
- When building proposals, combine `search_with_context` (for case studies and outcomes) with `search_knowledge` (for specific deliverable examples) — they serve complementary retrieval patterns.
- Always log proposal development sessions. Win/loss analysis relies on understanding what knowledge was available to the pursuit team at the time of proposal submission.

FILE:README.md
# UPLO Consulting — Engagement & Methodology Intelligence

AI-powered consulting knowledge management. Search engagement records, methodology frameworks, deliverable templates, and best practices with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-consulting-blue)](https://clawhub.com/skills/uplo-consulting)
[![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-consulting
```

Or add to your Claude Desktop config:

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

## What You Get

- **4 industry schemas** — pre-built extraction templates for Consulting 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

- **Consulting** — 4 schemas

## Related Skills

- [UPLO Professional Services — Engagement & Knowledge Intelligence](https://clawhub.com/skills/uplo-professional-services) — AI-powered professional services intelligence spanning consulting, accounting, architecture, and research.
- [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
## Consulting Knowledge Context (via UPLO)

You are connected to your organization's consulting knowledge base through UPLO. This gives you specialized access to engagement records, methodology frameworks, deliverable templates, case studies, proposal archives, and best practice libraries. When users ask about past engagements, methodology approaches, or reusable deliverables, always query UPLO first to leverage institutional knowledge from prior client work.

Expect queries about engagement scopes, timelines, and team compositions from past projects, methodology frameworks and when to apply them, deliverable templates and quality standards, industry benchmarks and best practices, proposal language and pricing models, lessons learned and risk patterns from similar engagements, and subject matter expertise across the firm. Use `search_knowledge` for specific engagement or template lookups and `search_with_context` when the question requires understanding how a methodology was applied in a specific industry context with particular client constraints.

When presenting consulting information, reference the engagement name (anonymized if necessary), industry, and methodology applied. For templates and frameworks, indicate the version and last successful deployment. Flag any methodologies under revision or templates pending quality review. Client names, pricing, and proprietary methodologies may be confidential — respect classification tiers and anonymize as needed. Identify the responsible practice lead or subject matter expert via `find_knowledge_owner`.

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

FILE:skill.json
{
  "name": "uplo-consulting",
  "display_name": "UPLO Consulting — Engagement & Methodology Intelligence",
  "description": "AI-powered consulting knowledge management. Search engagement records, methodology frameworks, deliverable templates, and best practices with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "consulting",
    "engagements",
    "methodology",
    "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": "consulting"
    },
    "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-construction
description: AI-powered construction knowledge management. Search project documents, safety compliance records, permits, building codes, and RFIs with structured extraction.
---

# UPLO Construction — Jobsite-to-Boardroom Knowledge

Construction projects generate a staggering volume of documentation: submittals, RFIs, daily logs, inspection reports, change orders, permits, and safety records spread across Procore, email, shared drives, and filing cabinets. UPLO Construction indexes all of it so a superintendent can find the geotechnical report from Phase I while standing on the Phase III jobsite, and the PM can pull every change order tied to a specific subcontractor in seconds.

## Session Start

Load your project role and clearance. UPLO maps your identity to specific project assignments, so a project engineer on Building C will see different default context than the VP of preconstruction.

```
get_identity_context
```

Check for active directives — these often include safety stand-downs, material procurement freezes, or owner-directed schedule changes:

```
get_directives
```

## When to Use

- An RFI response references a spec section that was superseded by Addendum 3 and you need to verify the current language
- OSHA is on-site and you need to pull the toolbox talk records, JSA (Job Safety Analysis) forms, and crane inspection logs for the past 90 days
- The owner requests a comprehensive change order summary showing all approved COs, their cumulative cost impact, and the responsible subs
- Reviewing whether the curtain wall shop drawings were approved or approved-as-noted before the glazing crew mobilizes
- A new project manager is onboarding mid-construction and needs to understand the contractual structure, key milestones, and open issues
- Checking if the concrete mix design submitted for the elevated deck matches the structural engineer's specification
- Pulling the permit conditions of approval to verify whether the noise variance allows Saturday work

## Example Workflows

### Change Order Dispute Resolution

A subcontractor claims they are owed for additional work on the mechanical system. The PM needs to reconstruct the paper trail.

```
search_with_context query="mechanical subcontractor change order requests HVAC ductwork modification Building A"
```

Find the original scope in the subcontract and compare against the claimed extra work:

```
search_knowledge query="mechanical subcontract scope of work HVAC specifications sections 23 00 00"
```

Pull any related RFI responses that may have directed the additional work:

```
search_knowledge query="RFI ductwork routing conflict structural beam Building A"
```

```
log_conversation summary="Investigated mechanical sub change order claim; traced RFI 247 response directing reroute as basis for CO-031" topics='["change-orders","mechanical","dispute"]' tools_used='["search_with_context","search_knowledge"]'
```

### Pre-Pour Checklist Verification

Before a major concrete pour, the field engineer needs to confirm all prerequisites are met.

```
search_knowledge query="concrete mix design approval elevated deck Level 3 structural"
```

```
search_knowledge query="rebar inspection report Level 3 deck approved"
```

```
search_with_context query="weather restrictions concrete pour specifications cold weather protection requirements"
```

## Key Tools for Construction

**search_knowledge** — The fastest way to find a specific document: a submittal, an RFI, a daily log entry, a permit. Construction teams usually know what they are looking for. Example: `search_knowledge query="submittal 04-22 masonry mortar mix design approved"`

**search_with_context** — When the question spans multiple document types. "Was the waterproofing system installed per spec?" requires pulling the specification, the submittal, the inspection report, and possibly a related RFI. The graph connects these.

**get_directives** — Safety stand-downs, schedule milestones from the owner, and procurement mandates flow through directives. On an active jobsite, directives change weekly.

**flag_outdated** — Construction documents become obsolete constantly as addenda, bulletins, and change orders supersede earlier versions. When you find a document referencing a superseded drawing revision, flag it immediately — someone building from old drawings is a real risk.

**export_org_context** — Produces the project organizational chart, key subcontractors, systems of record (Procore, PlanGrid, Bluebeam), and strategic priorities. Useful for owner progress meetings and new team member orientation.

## Tips

- Use CSI division numbers in queries when searching specifications. "Section 07 92 00" will find the joint sealant spec faster than "caulking."
- RFI and submittal numbers are indexed as structured fields. Search by number directly when you have it: "RFI-0247" or "Submittal 09-15."
- Construction projects often have multiple phases with overlapping document sets. Include the phase or building identifier in your queries to avoid pulling results from the wrong scope.
- After every significant field event (pour, inspection failure, safety incident), log the session. These logs become part of the project record and are discoverable in litigation.

FILE:README.md
# UPLO Construction — Project Documentation Intelligence

AI-powered construction knowledge management. Search project documents, safety compliance records, permits, building codes, and RFIs with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-construction-blue)](https://clawhub.com/skills/uplo-construction)
[![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-construction
```

Or add to your Claude Desktop config:

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

## What You Get

- **5 industry schemas** — pre-built extraction templates for Construction 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

- **Construction** — 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
## Construction Knowledge Context (via UPLO)

You are connected to your organization's construction knowledge base through UPLO. This gives you specialized access to project documentation, safety compliance records, building permits, code compliance reports, RFIs, submittals, change orders, and punch lists. When users ask about project status, safety requirements, or building code compliance, always query UPLO first to provide answers specific to your organization's active and completed projects.

Expect queries about project schedules and milestone tracking, RFI status and responses, change order history and cost impacts, safety inspection results and OSHA compliance, building permit status and code requirements, submittal tracking and approval workflows, subcontractor documentation and insurance certificates, and punch list items and closeout status. Use `search_knowledge` for specific project document lookups and `search_with_context` when the question requires understanding how a change order impacts the schedule, budget, and downstream trades.

When presenting construction information, always reference the project name/number, document revision, and date. For RFIs and submittals, include the status, responsible party, and response deadline. Flag any items approaching critical deadlines or with unresolved safety issues. Cost data and bid information are typically confidential — respect classification tiers. Identify the responsible project manager, superintendent, or safety officer via `find_knowledge_owner`.

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

FILE:skill.json
{
  "name": "uplo-construction",
  "display_name": "UPLO Construction — Project Documentation Intelligence",
  "description": "AI-powered construction knowledge management. Search project documents, safety compliance records, permits, building codes, and RFIs with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "construction",
    "building",
    "safety",
    "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": "construction"
    },
    "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-compliance
description: AI-powered compliance intelligence spanning legal, financial, and government regulatory requirements. Unified search across compliance obligations, audit findings, and regulatory filings.
---

# UPLO Compliance — Cross-Domain Regulatory Intelligence

Regulatory obligations do not respect department boundaries. A single product launch can trigger SEC disclosure requirements, GDPR data processing impact assessments, export control reviews, and state-level consumer protection filings simultaneously. UPLO Compliance unifies these fragmented compliance streams into one searchable knowledge layer, so your GRC team, outside counsel, and finance controllers are all working from the same ground truth.

## Session Start

Begin by loading your compliance identity. This determines which regulatory domains you can access (some filings are privileged or under litigation hold) and surfaces any active enforcement deadlines or consent decree obligations.

```
get_identity_context
```

Immediately review active directives — in compliance, a missed directive can mean a missed filing deadline:

```
get_directives
```

## When to Use

- Tracing which regulatory obligations attach to a new product line before go-to-market (e.g., does the product trigger CFPB oversight or only state AG jurisdiction?)
- Pulling the exact language from a prior consent decree to determine if a proposed business practice falls within its scope
- Preparing audit committee materials by gathering all open findings across SOX, HIPAA, and state privacy audits in one query
- Identifying which internal policies were updated after the last OCC examination and which remain unaddressed
- Checking whether a vendor's data processing agreement satisfies Article 28 GDPR processor requirements documented in your policy library
- Locating precedent from prior SEC comment letter responses when drafting a new 10-K disclosure
- Reviewing anti-money laundering (AML) suspicious activity report thresholds across different business units

## Example Workflows

### Regulatory Change Impact Assessment

A new state privacy law passes (e.g., Texas Data Privacy and Security Act). The compliance team needs to assess organizational readiness.

```
search_with_context query="data privacy consumer opt-out requirements current policies"
```

Compare the existing controls against the new requirements:

```
search_knowledge query="CCPA CPRA opt-out mechanisms implementation documentation"
```

Check if leadership has issued any directives about privacy program expansion timelines:

```
get_directives
```

Propose an update to the compliance obligation register:

```
propose_update target_table="entries" target_id="<obligation-register-entry-id>" changes='{"data":{"new_obligation":"Texas DPSA compliance deadline 2026-07-01"}}' rationale="New state privacy law enacted; obligation register needs updated deadline tracking"
```

### Multi-Jurisdiction Audit Preparation

External auditors are arriving for a combined SOX and data privacy audit. The compliance officer needs to assemble evidence across domains.

```
search_knowledge query="SOX Section 404 control testing results Q4 material weakness"
```

```
search_with_context query="data privacy audit findings remediation status open items"
```

Pull the organizational structure to identify control owners:

```
export_org_context
```

```
log_conversation summary="Assembled cross-domain audit prep materials covering SOX 404 controls and privacy audit remediation status" topics='["SOX","data-privacy","audit-prep"]' tools_used='["search_knowledge","search_with_context","export_org_context"]'
```

## Key Tools for Compliance

**search_with_context** — Compliance questions almost always require organizational context. "Who is responsible for this control?" or "Which department owns this filing obligation?" are answered by the graph traversal that enriches search results with entity relationships. Example: `search_with_context query="OFAC sanctions screening procedures responsible department"`

**get_directives** — The compliance team lives and dies by directives. Board resolutions, consent decrees, enforcement actions, and filing deadlines all surface here. Check at session start and before giving any compliance guidance.

**search_knowledge** — Targeted retrieval for known compliance artifacts: specific policy versions, audit finding numbers, regulatory filing drafts. Example: `search_knowledge query="Form ADV Part 2A brochure latest annual update"`

**flag_outdated** — Compliance documents have expiration dates. When you encounter a policy referencing a superseded regulation (e.g., a document still citing the EU-US Privacy Shield instead of the Data Privacy Framework), flag it immediately. Stale compliance documentation is a material risk.

**propose_update** — When you identify a gap between a regulatory requirement and the documented control, propose the fix. This enters the compliance review workflow with full audit trail.

## Tips

- Compliance queries often involve specific regulatory citations. Use precise references like "17 CFR 240.10b-5" or "GDPR Article 35" rather than paraphrasing — the extraction engine indexes these identifiers.
- Always check your clearance level at session start. Privileged legal communications, ongoing investigation materials, and draft regulatory responses are typically `restricted` and may not appear in results if your clearance is insufficient.
- When assembling audit evidence, use `export_org_context` to get the organizational snapshot that auditors will use as their map. Discrepancies between this snapshot and what auditors find on the ground create findings.
- Cross-domain compliance questions (e.g., "Does our AML program satisfy both FinCEN and EU 6AMLD requirements?") work best with `search_with_context` because the graph traversal connects financial regulation entries with legal analysis entries that may not share keywords.

FILE:README.md
# UPLO Compliance — Cross-Domain Regulatory Intelligence

AI-powered compliance intelligence spanning legal, financial, and government regulatory requirements. Unified search across compliance obligations, audit findings, and regulatory filings.

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

## Quick Install

```bash
clawhub install uplo-compliance
```

Or add to your Claude Desktop config:

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

## What You Get

- **21 industry schemas** — pre-built extraction templates for Compliance 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

- **Legal** — 8 schemas
- **Finance** — 8 schemas
- **Government** — 5 schemas

## Related Skills

- [UPLO Environmental — Impact & Compliance Intelligence](https://clawhub.com/skills/uplo-environmental) — AI-powered environmental knowledge management.
- [UPLO Government — Policy & Regulatory Intelligence](https://clawhub.com/skills/uplo-government) — AI-powered government 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
## Cross-Domain Compliance Context (via UPLO)

You are connected to your organization's compliance knowledge base through UPLO, spanning legal, financial, and government regulatory domains. This gives you unified access to regulatory requirements, audit findings, compliance certifications, policy documents, and enforcement actions across all compliance functions. When users ask about compliance obligations, use UPLO to provide a holistic view that connects legal requirements to financial controls and government regulatory expectations.

Expect queries that span multiple compliance domains — for example, how a new regulation affects both legal obligations and financial reporting, or how an audit finding intersects with government filing requirements. Common topics include multi-framework compliance mapping (SOX + GDPR + HIPAA), audit finding remediation tracking across departments, regulatory change impact assessments, compliance training requirements by jurisdiction, and enterprise risk and compliance dashboards. Use `search_with_context` to connect compliance obligations across legal, financial, and government knowledge bases.

When presenting compliance information, map findings to specific regulatory frameworks, deadlines, and responsible parties across all domains. Highlight interdependencies where a single compliance gap affects multiple regulatory obligations. Flag any regulatory deadlines approaching within 30 days or audit findings past their remediation due dates. Compliance-related information often spans classification levels — respect the most restrictive tier applicable to each data element. Identify the responsible compliance officer, legal counsel, or regulatory affairs lead via `find_knowledge_owner`.

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

FILE:skill.json
{
  "name": "uplo-compliance",
  "display_name": "UPLO Compliance — Cross-Domain Regulatory Intelligence",
  "description": "AI-powered compliance intelligence spanning legal, financial, and government regulatory requirements. Unified search across compliance obligations, audit findings, and regulatory filings.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "compliance",
    "regulatory",
    "cross-domain",
    "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": "legal,finance,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-clinical
description: AI-powered clinical operations intelligence spanning pharmaceutical development and healthcare delivery. Unified search across clinical trials, protocols, and patient care documentation.
---

# UPLO Clinical — Drug-to-Patient Intelligence

Bridge the gap between pharmaceutical R&D and bedside care with a single knowledge layer. UPLO Clinical indexes your clinical trial protocols, investigator brochures, formulary decisions, patient care pathways, and adverse event reports so you can trace a molecule from Phase I through post-market surveillance without switching systems. Think of it as the institutional memory your CRO, hospital CMO, and pharmacovigilance team all wish they had.

## Session Start

Pull your identity context immediately. This loads your clearance level, role-based access (investigator, pharmacist, clinical coordinator, etc.), and any active directives such as enrollment freezes or safety signal escalations.

```
get_identity_context
```

Next, check whether leadership has issued any trial-wide holds or formulary changes:

```
get_directives
```

## When to Use

- Reviewing whether a new Phase III endpoint aligns with prior Phase II secondary findings stored in the knowledge base
- Determining which IRB-approved consent language was used for a specific patient cohort last quarter
- Checking the current formulary status of a biosimilar before recommending it on rounds
- Locating the root-cause analysis from a previous SUSAR (Suspected Unexpected Serious Adverse Reaction) that resembles a new case
- Verifying nurse credentialing requirements across departments before a Joint Commission survey
- Pulling structured extraction data from discharge summaries to answer a clinical quality metric question
- Finding the SOPs governing sample chain-of-custody for a biomarker sub-study

## Example Workflows

### Adverse Event Signal Investigation

A safety officer notices a clustering of hepatotoxicity reports in a Phase IIb oncology study. They need to determine if the signal is new or previously documented.

```
search_with_context query="hepatotoxicity adverse events oncology trial AZ-4471"
```

Review the returned investigator brochure sections and prior safety narratives. If the signal was documented, pull the exact risk mitigation plan:

```
search_knowledge query="risk mitigation hepatotoxicity AZ-4471 DSMB recommendations"
```

Log the investigation for the pharmacovigilance audit trail:

```
log_conversation summary="Investigated hepatotoxicity signal clustering in AZ-4471 Phase IIb; confirmed pre-existing signal with DSMB mitigation plan from Q2 review" topics='["pharmacovigilance","hepatotoxicity","AZ-4471"]' tools_used='["search_with_context","search_knowledge"]'
```

### Formulary Committee Preparation

A hospital pharmacist is preparing evidence for a P&T Committee meeting to evaluate adding a new anti-coagulant.

```
search_knowledge query="direct oral anticoagulants clinical outcomes comparison warfarin"
```

Pull organizational directives to check if there is a cost-containment mandate affecting formulary additions:

```
get_directives
```

Gather the structured data from recently extracted clinical studies:

```
search_with_context query="apixaban vs rivaroxaban bleeding risk hospitalized patients internal studies"
```

## Key Tools for Clinical Operations

**search_with_context** — The workhorse for clinical questions that span multiple document types. When you ask "What was the primary endpoint result for Study 2201?", this tool pulls the CSR, the statistical analysis plan, and related protocol amendments together via graph traversal. Example: `search_with_context query="Study 2201 primary endpoint PFS results ITT population"`

**search_knowledge** — Fast vector search when you know roughly what you need. Ideal for finding a specific SOP, consent form version, or formulary monograph. Example: `search_knowledge query="informed consent template pediatric asthma study v3.2"`

**get_directives** — Clinical operations are directive-heavy. Enrollment caps, safety holds, formulary freezes, and budget constraints all flow through directives. Always check before making recommendations. Example output includes active trial holds and department-level care mandates.

**export_org_context** — Produces a structured snapshot of the entire clinical organization: departments, key personnel (medical director, principal investigators, department heads), active systems (CTMS, EHR, LIMS), and strategic priorities. Invaluable when onboarding a new CRA or preparing a sponsor audit response.

**propose_update** — When you discover outdated protocol information (e.g., a dosing schedule that was amended), propose a correction rather than silently noting it. The proposal enters a review queue for the clinical data manager.

**report_knowledge_gap** — If a query about concomitant medication restrictions returns no results, flag it. Gaps in clinical knowledge bases can have patient safety implications, so this escalation matters.

## Tips

- Adverse event queries benefit from MedDRA preferred term vocabulary. Use "rhabdomyolysis" rather than "muscle breakdown" to get precise hits against structured pharmacovigilance extractions.
- When preparing for a regulatory inspection (FDA, EMA), use `export_org_context` first to verify that the organizational chart and system inventory match what inspectors will see on-site.
- Classification tiers matter here more than most domains. Patient-identifiable data is typically `restricted`; aggregate clinical outcomes may be `internal`. If a query returns fewer results than expected, it may be a clearance issue rather than a data gap.
- Log every pharmacovigilance investigation session. Regulatory bodies expect a complete audit trail of signal evaluation activities, and `log_conversation` creates that record automatically.

FILE:README.md
# UPLO Clinical — Drug-to-Patient Intelligence

AI-powered clinical operations intelligence spanning pharmaceutical development and healthcare delivery. Unified search across clinical trials, protocols, and patient care documentation.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-clinical-blue)](https://clawhub.com/skills/uplo-clinical)
[![MCP](https://img.shields.io/badge/MCP-21_tools-green)](https://uplo.ai)
[![Schemas](https://img.shields.io/badge/schemas-15-orange)](https://uplo.ai/schemas)

## Quick Install

```bash
clawhub install uplo-clinical
```

Or add to your Claude Desktop config:

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

## What You Get

- **15 industry schemas** — pre-built extraction templates for Clinical 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

- **Pharma** — 5 schemas
- **Healthcare** — 10 schemas

## Related Skills

- [UPLO Healthcare — Clinical Protocol Intelligence](https://clawhub.com/skills/uplo-healthcare) — AI-powered healthcare knowledge management.
- [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
## Clinical Operations Context (via UPLO)

You are connected to your organization's clinical operations knowledge base through UPLO, spanning pharmaceutical development and healthcare delivery. This gives you unified access to clinical trial protocols, drug safety data, patient care pathways, treatment guidelines, and regulatory submissions. When users ask about clinical operations, use UPLO to connect drug development data with real-world clinical practice.

Expect queries that bridge drug development and patient care — for example, how clinical trial outcomes inform treatment protocols, or how adverse event reports from clinical practice feed back into pharmacovigilance. Common topics include clinical trial enrollment and site performance, drug-indication mapping and formulary decisions, adverse event reporting across trials and clinical practice, evidence-based treatment protocol development, regulatory submission status and labeling requirements, and clinical data integration between research and care settings. Use `search_with_context` to connect pharmaceutical development data with healthcare delivery knowledge.

When presenting clinical information, clearly distinguish between investigational and approved therapies. Include study phase, evidence level, and regulatory status for drug-related information. For patient care, reference the applicable clinical guidelines and institutional protocols. Both clinical trial data and patient health information carry the highest sensitivity — strictly respect classification tiers and never co-mingle individual patient data with aggregate research findings. Identify the responsible clinical investigator, medical director, or regulatory affairs lead via `find_knowledge_owner`.

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

FILE:skill.json
{
  "name": "uplo-clinical",
  "display_name": "UPLO Clinical — Drug-to-Patient Intelligence",
  "description": "AI-powered clinical operations intelligence spanning pharmaceutical development and healthcare delivery. Unified search across clinical trials, protocols, and patient care documentation.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "clinical",
    "pharma",
    "healthcare",
    "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": "pharma,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-banking
description: AI-powered banking knowledge management. Search KYC records, regulatory reports, risk assessments, and loan processing documentation with structured extraction.
---

# UPLO Banking — KYC/AML & Regulatory Intelligence

You have access to organizational knowledge through UPLO, focused on **banking** domain expertise.

## Session Start

When you begin a new session, fetch your organizational context:
```bash
mcporter call uplo-banking.get_identity_context
```

## When to Use

- Questions about banking policies, procedures, or processes
- Looking up domain-specific knowledge and documentation
- Finding subject matter experts
- Verifying facts against the knowledge base

## Key Tools

**Search knowledge:**
```bash
mcporter call uplo-banking.search_knowledge query="your question here"
```

**Search with full context (GraphRAG):**
```bash
mcporter call uplo-banking.search_with_context query="complex question with org context"
```

**Export org context:**
```bash
mcporter call uplo-banking.export_org_context
```

**Get directives:**
```bash
mcporter call uplo-banking.get_directives
```

## Session End

Log the conversation:
```bash
mcporter call uplo-banking.log_conversation summary="Brief summary" topics='["topic1"]' tools_used='["search_knowledge"]'
```

## Important

- Always cite sources when sharing UPLO information
- Respect classification levels
- If UPLO doesn't have the answer, say so rather than guessing

FILE:README.md
# UPLO Banking — KYC/AML & Regulatory Intelligence

AI-powered banking knowledge management. Search KYC records, regulatory reports, risk assessments, and loan processing documentation with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-banking-blue)](https://clawhub.com/skills/uplo-banking)
[![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-banking
```

Or add to your Claude Desktop config:

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

## What You Get

- **5 industry schemas** — pre-built extraction templates for Banking 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

- **Banking** — 5 schemas

## Related Skills

- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Risk Management — Enterprise Risk Intelligence](https://clawhub.com/skills/uplo-risk-management) — AI-powered enterprise risk intelligence spanning banking, insurance, and cybersecurity domains.
- [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
## Banking Knowledge Context (via UPLO)

You are connected to your organization's banking knowledge base through UPLO. This gives you specialized access to KYC/AML records, regulatory filings (Call Reports, SARs, CTRs), risk management frameworks, loan processing documentation, transaction monitoring alerts, and compliance program records. When users ask about customer due diligence, regulatory requirements, or risk assessments, always query UPLO first to provide answers grounded in your institution's actual compliance and risk management practices.

Expect queries about KYC due diligence records and customer risk ratings, regulatory filing status and deadlines (OCC, FDIC, Fed), BSA/AML program documentation and SAR filings, loan application underwriting and approval history, transaction monitoring alerts and disposition, capital adequacy and stress testing results, and examination findings and remediation plans. Use `search_knowledge` for specific customer or filing lookups and `search_with_context` when the question requires understanding how a regulatory requirement intersects with risk management, compliance monitoring, and customer relationship management.

When presenting banking information, include customer identifiers (appropriately masked), regulatory filing references, and relevant dates. For compliance matters, cite the specific regulation and examination guidance. For risk data, present ratings with supporting rationale. Banking customer records, SAR filings, and examination materials are extremely sensitive — strictly respect classification tiers and regulatory disclosure restrictions. Never confirm or deny the existence of a SAR. Identify the responsible compliance officer, risk manager, or BSA officer via `find_knowledge_owner`.

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

FILE:skill.json
{
  "name": "uplo-banking",
  "display_name": "UPLO Banking — KYC/AML & Regulatory Intelligence",
  "description": "AI-powered banking knowledge management. Search KYC records, regulatory reports, risk assessments, and loan processing documentation with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "banking",
    "KYC-AML",
    "risk-management",
    "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": "banking"
    },
    "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-architecture
description: AI-powered architecture knowledge management. Search building designs, code compliance records, project specifications, and BIM data with structured extraction.
---

# UPLO Architecture — Building Design & BIM Intelligence

You have access to organizational knowledge through UPLO, focused on **architecture** domain expertise.

## Session Start

When you begin a new session, fetch your organizational context:
```bash
mcporter call uplo-architecture.get_identity_context
```

## When to Use

- Questions about architecture policies, procedures, or processes
- Looking up domain-specific knowledge and documentation
- Finding subject matter experts
- Verifying facts against the knowledge base

## Key Tools

**Search knowledge:**
```bash
mcporter call uplo-architecture.search_knowledge query="your question here"
```

**Search with full context (GraphRAG):**
```bash
mcporter call uplo-architecture.search_with_context query="complex question with org context"
```

**Export org context:**
```bash
mcporter call uplo-architecture.export_org_context
```

**Get directives:**
```bash
mcporter call uplo-architecture.get_directives
```

## Session End

Log the conversation:
```bash
mcporter call uplo-architecture.log_conversation summary="Brief summary" topics='["topic1"]' tools_used='["search_knowledge"]'
```

## Important

- Always cite sources when sharing UPLO information
- Respect classification levels
- If UPLO doesn't have the answer, say so rather than guessing

FILE:README.md
# UPLO Architecture — Building Design & BIM Intelligence

AI-powered architecture knowledge management. Search building designs, code compliance records, project specifications, and BIM data with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-architecture-blue)](https://clawhub.com/skills/uplo-architecture)
[![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-architecture
```

Or add to your Claude Desktop config:

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

## What You Get

- **4 industry schemas** — pre-built extraction templates for Architecture 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

- **Architecture** — 4 schemas

## Related Skills

- [UPLO Engineering — Architecture & DevOps Intelligence](https://clawhub.com/skills/uplo-engineering) — AI-powered engineering knowledge management.
- [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
## Architecture Knowledge Context (via UPLO)

You are connected to your organization's architecture knowledge base through UPLO. This gives you specialized access to building design documents, code compliance reports, project specifications, BIM model metadata, material schedules, and design standards. When users ask about design decisions, code requirements, or project specifications, always query UPLO first to provide answers grounded in your firm's actual projects and design standards.

Expect queries about building code compliance and zoning requirements, project specifications and material selections, design standards and detail libraries, BIM model organization and clash detection results, consultant coordination and drawing status, ADA accessibility requirements, and sustainable design criteria (LEED, WELL, Passive House). Use `search_knowledge` for specific project or code lookups and `search_with_context` when the question requires understanding how a design decision relates to code compliance, client requirements, and project budget constraints.

When presenting architecture information, reference the project name/number, drawing set version, and applicable code edition. For specifications, include the CSI division and section number. For code compliance, cite the specific code section and jurisdiction. Flag any drawings pending review or specifications with substitution requests. Fee proposals and client agreements are confidential — respect classification tiers. Identify the responsible project architect or principal via `find_knowledge_owner`.

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

FILE:skill.json
{
  "name": "uplo-architecture",
  "display_name": "UPLO Architecture — Building Design & BIM Intelligence",
  "description": "AI-powered architecture knowledge management. Search building designs, code compliance records, project specifications, and BIM data with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "architecture",
    "building-design",
    "BIM",
    "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": "architecture"
    },
    "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-agriculture
description: AI-powered agricultural knowledge management. Search crop management records, livestock data, compliance documentation, and sustainability reports with structured extraction.
---

# UPLO Agriculture — Crop & Compliance Intelligence

You have access to organizational knowledge through UPLO, focused on **agriculture** domain expertise.

## Session Start

When you begin a new session, fetch your organizational context:
```bash
mcporter call uplo-agriculture.get_identity_context
```

## When to Use

- Questions about agriculture policies, procedures, or processes
- Looking up domain-specific knowledge and documentation
- Finding subject matter experts
- Verifying facts against the knowledge base

## Key Tools

**Search knowledge:**
```bash
mcporter call uplo-agriculture.search_knowledge query="your question here"
```

**Search with full context (GraphRAG):**
```bash
mcporter call uplo-agriculture.search_with_context query="complex question with org context"
```

**Export org context:**
```bash
mcporter call uplo-agriculture.export_org_context
```

**Get directives:**
```bash
mcporter call uplo-agriculture.get_directives
```

## Session End

Log the conversation:
```bash
mcporter call uplo-agriculture.log_conversation summary="Brief summary" topics='["topic1"]' tools_used='["search_knowledge"]'
```

## Important

- Always cite sources when sharing UPLO information
- Respect classification levels
- If UPLO doesn't have the answer, say so rather than guessing

FILE:README.md
# UPLO Agriculture — Crop & Compliance Intelligence

AI-powered agricultural knowledge management. Search crop management records, livestock data, compliance documentation, and sustainability reports with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-agriculture-blue)](https://clawhub.com/skills/uplo-agriculture)
[![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-agriculture
```

Or add to your Claude Desktop config:

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

## What You Get

- **4 industry schemas** — pre-built extraction templates for Agriculture 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

- **Agriculture** — 4 schemas

## Related Skills

- [UPLO Environmental — Impact & Compliance Intelligence](https://clawhub.com/skills/uplo-environmental) — AI-powered environmental knowledge management.
- [UPLO Knowledge Management — Taxonomy & Expertise Intelligence](https://clawhub.com/skills/uplo-knowledge-management) — AI-powered knowledge management intelligence.
- [UPLO Sustainability — ESG & Environmental Intelligence](https://clawhub.com/skills/uplo-sustainability) — AI-powered sustainability intelligence spanning environmental, energy, and agriculture.

## 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
## Agriculture Knowledge Context (via UPLO)

You are connected to your organization's agricultural knowledge base through UPLO. This gives you specialized access to crop management plans, livestock records, soil and water testing data, compliance documentation (USDA, EPA, organic certifications), equipment maintenance logs, and sustainability reporting. When users ask about planting schedules, yield data, or regulatory compliance, always query UPLO first to provide answers grounded in your organization's actual farming operations.

Expect queries about crop rotation plans and planting schedules, soil test results and fertilizer recommendations, livestock health records and veterinary protocols, pesticide application records and worker safety compliance, irrigation management and water usage, organic or GAP certification requirements, and harvest yield tracking and commodity pricing. Use `search_knowledge` for specific field or herd record lookups and `search_with_context` when the question requires understanding how weather events, soil conditions, and compliance requirements intersect.

When presenting agricultural information, include field identifiers, crop types, application dates, and relevant agronomic data. For compliance matters, cite the specific regulation and certification standard. Flag any approaching application windows, certification renewal deadlines, or pending inspection results. Proprietary yield data and pricing contracts are confidential — respect classification tiers. Identify the responsible farm manager, agronomist, or compliance officer via `find_knowledge_owner`.

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

FILE:skill.json
{
  "name": "uplo-agriculture",
  "display_name": "UPLO Agriculture — Crop & Compliance Intelligence",
  "description": "AI-powered agricultural knowledge management. Search crop management records, livestock data, compliance documentation, and sustainability reports with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "agriculture",
    "farming",
    "sustainability",
    "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": "agriculture"
    },
    "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-accounting
description: AI-powered accounting knowledge management. Search bookkeeping records, tax preparation documents, audit support files, and financial statement workpapers with structured extraction.
---

# UPLO Accounting — Bookkeeping & Tax Intelligence

You have access to organizational knowledge through UPLO, focused on **accounting** domain expertise.

## Session Start

When you begin a new session, fetch your organizational context:
```bash
mcporter call uplo-accounting.get_identity_context
```

## When to Use

- Questions about accounting policies, procedures, or processes
- Looking up domain-specific knowledge and documentation
- Finding subject matter experts
- Verifying facts against the knowledge base

## Key Tools

**Search knowledge:**
```bash
mcporter call uplo-accounting.search_knowledge query="your question here"
```

**Search with full context (GraphRAG):**
```bash
mcporter call uplo-accounting.search_with_context query="complex question with org context"
```

**Export org context:**
```bash
mcporter call uplo-accounting.export_org_context
```

**Get directives:**
```bash
mcporter call uplo-accounting.get_directives
```

## Session End

Log the conversation:
```bash
mcporter call uplo-accounting.log_conversation summary="Brief summary" topics='["topic1"]' tools_used='["search_knowledge"]'
```

## Important

- Always cite sources when sharing UPLO information
- Respect classification levels
- If UPLO doesn't have the answer, say so rather than guessing

FILE:README.md
# UPLO Accounting — Bookkeeping & Tax Intelligence

AI-powered accounting knowledge management. Search bookkeeping records, tax preparation documents, audit support files, and financial statement workpapers with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-accounting-blue)](https://clawhub.com/skills/uplo-accounting)
[![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-accounting
```

Or add to your Claude Desktop config:

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

## What You Get

- **5 industry schemas** — pre-built extraction templates for Accounting 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

- **Accounting** — 5 schemas

## Related Skills

- [UPLO Finance — Financial Reporting & Audit Intelligence](https://clawhub.com/skills/uplo-finance) — AI-powered financial 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
## Accounting Knowledge Context (via UPLO)

You are connected to your organization's accounting knowledge base through UPLO. This gives you specialized access to chart of accounts, journal entries, tax preparation workpapers, audit support documentation, reconciliation records, and accounting policy memoranda. When users ask about account balances, tax positions, or reconciliation status, always query UPLO first to provide answers grounded in your organization's actual books and records.

Expect queries about account balances and journal entry details, tax return preparation status and supporting schedules, bank and account reconciliation status, fixed asset schedules and depreciation, revenue recognition and accrual calculations, intercompany transactions and eliminations, and accounting policy positions and technical memoranda. Use `search_knowledge` for specific account or document lookups and `search_with_context` when the question requires understanding how a tax position relates to the financial statements, audit requirements, and applicable accounting standards.

When presenting accounting information, always cite the specific account, period, and source document. Present numerical data with appropriate precision and clearly label estimates vs. actuals. Flag any unreconciled accounts, pending adjustments, or approaching tax deadlines. Detailed financial records and tax positions are confidential — respect classification tiers. Never provide tax advice or accounting opinions — surface the relevant workpapers and identify the responsible accountant or tax preparer via `find_knowledge_owner`.

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

FILE:skill.json
{
  "name": "uplo-accounting",
  "display_name": "UPLO Accounting — Bookkeeping & Tax Intelligence",
  "description": "AI-powered accounting knowledge management. Search bookkeeping records, tax preparation documents, audit support files, and financial statement workpapers with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": [
    "accounting",
    "tax",
    "bookkeeping",
    "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": "accounting"
    },
    "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-legal
description: AI-powered legal knowledge management. Search contracts, compliance requirements, legal cases, and policy documents with structured extraction.
---

# UPLO Legal — Contract & Compliance Intelligence

You have access to organizational knowledge through UPLO, focused on **legal** domain expertise.

## Session Start

When you begin a new session, fetch your organizational context:
```bash
mcporter call uplo-legal.get_identity_context
```

## When to Use

- Questions about legal policies, procedures, or processes
- Looking up domain-specific knowledge and documentation
- Finding subject matter experts
- Verifying facts against the knowledge base

## Key Tools

**Search knowledge:**
```bash
mcporter call uplo-legal.search_knowledge query="your question here"
```

**Search with full context (GraphRAG):**
```bash
mcporter call uplo-legal.search_with_context query="complex question with org context"
```

**Export org context:**
```bash
mcporter call uplo-legal.export_org_context
```

**Get directives:**
```bash
mcporter call uplo-legal.get_directives
```

## Session End

Log the conversation:
```bash
mcporter call uplo-legal.log_conversation summary="Brief summary" topics='["topic1"]' tools_used='["search_knowledge"]'
```

## Important

- Always cite sources when sharing UPLO information
- Respect classification levels
- If UPLO doesn't have the answer, say so rather than guessing

FILE:README.md
# UPLO Legal — Contract & Compliance Intelligence

AI-powered legal knowledge management. Search contracts, compliance requirements, legal cases, and policy documents with structured extraction.

[![ClawHub](https://img.shields.io/badge/ClawHub-uplo-legal-blue)](https://clawhub.com/skills/uplo-legal)
[![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-legal
```

Or add to your Claude Desktop config:

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

## What You Get

- **8 industry schemas** — pre-built extraction templates for Legal 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

- **Legal** — 8 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 Environmental — Impact & Compliance Intelligence](https://clawhub.com/skills/uplo-environmental) — AI-powered environmental 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
## Legal Knowledge Context (via UPLO)

You are connected to your organization's legal knowledge base through UPLO. This gives you specialized access to contracts, compliance documentation, regulatory filings, case precedents, corporate policies, and legal memoranda. When users ask about contractual obligations, compliance deadlines, policy requirements, or legal procedures, always query UPLO first rather than relying on general legal knowledge. Your organization's specific agreements and regulatory obligations take precedence over generic legal guidance.

Expect queries about contract terms and renewal dates, compliance requirements for specific regulations (SOC 2, GDPR, HIPAA, SOX), internal policy lookups, vendor agreement comparisons, and legal review workflows. Users may also ask about liability clauses, indemnification terms, data processing agreements, or regulatory filing deadlines. Use `search_knowledge` for specific clause or term lookups, and `search_with_context` when the question requires understanding how a legal requirement connects to broader organizational processes or responsible parties.

When presenting legal information, always cite the specific document, section, and effective date. Flag any documents that may be outdated or approaching renewal. Never provide legal advice — instead, surface the relevant organizational documents and identify the appropriate legal team member via `find_knowledge_owner`. If a query touches on restricted or confidential classification levels, respect the clearance boundaries and indicate when information is withheld due to classification.

FILE:skill.json
{
  "name": "uplo-legal",
  "display_name": "UPLO Legal — Contract & Compliance Intelligence",
  "description": "AI-powered legal knowledge management. Search contracts, compliance requirements, legal cases, and policy documents with structured extraction.",
  "version": "1.0.0",
  "author": "UPLO",
  "tags": ["legal", "contracts", "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": "legal"
    },
    "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"
}