alexburrstudio

---
name: AB-Agents-Meter-Reader
description: "📊 Read meter readings from photos. Electricity (day/night tariffs) and water meters. Saves history and generates messages for landlord."
version: 1.0.0
author: AB-Agents
homepage: https://github.com/alexburrstudio/ab-agents-meter-reader
license: MIT
tags: ["meters", "utilities", "readings", "electricity", "water", "ab-agents"]
acceptLicenseTerms: true
---

# AB Agents Meter Reader 📊

Read meter readings from photos — electricity and water meters.

## Features

- ⚡ Read electricity meters (single or dual tariff)
- 💧 Read water meters (hot and cold)
- 📝 Save readings history with dates
- 📨 Generate message for landlord
- 🔄 Track multiple apartments

## Setup

### Requirements

- MiniMax Token Plan API key (for vision)
- Linux/macOS

### Quick Start

```bash
# First run - it will ask questions
./meter-reader.sh

# Later runs - just send photos
./meter-reader.sh photo.jpg
```

## First Run Setup

The script will ask:

1. **Tenant name** — your name
2. **Apartment address** — full address
3. **Meter layout** — how to tell meters apart:
   - "left=hot,right=cold" (default)
   - Or custom description

## Supported Meter Types

### Electricity
- Single tariff
- Dual tariff (day/night) — T1=day, T2=night
- Multi-tariff (cycle through screens)

### Water
- Cold water (usually on left)
- Hot water (usually on right)
- Cubic meters (m³)

## Usage

```bash
# Interactive mode (asks for photo)
./meter-reader.sh

# With photo
./meter-reader.sh /path/to/meter.jpg

# Generate message for landlord
./meter-reader.sh --message
```

## How It Works

1. Analyzes photo using MiniMax VL API
2. Identifies meter type automatically
3. Reads the numbers
4. Saves to readings history
5. Generates formatted message

## History

Readings saved to: `~/.meter-readings/history.json`

Format:
```json
{
  "apartments": {
    "address": {
      "tenant": "Name",
      "layout": "left=hot,right=cold",
      "readings": [
        {"date": "2026-04-26", "electricity_day": 8495, "electricity_night": 3008, "water_cold": 423, "water_hot": 240}
      ]
    }
  }
}
```

## Troubleshooting

| Issue | Solution |
|-------|----------|
| Wrong numbers | Check meter photo quality, ensure numbers are clear |
| Can't identify meter type | Name photo file: `electricity.jpg`, `water.jpg` |
| Vision error | Check MINIMAX_API_KEY is set |

---

**AB-Agents** 🦀

## Requirements

👁️ **[AB Agents Vision (MiniMax)](https://github.com/alexburrstudio/ab-agents-vision)** — Required for image analysis. Install first:
```bash
clawhub install AB-Agents-Vision-MiniMax
```

---

**AB-Agents** 🦀

FILE:meter-reader.sh
#!/bin/bash
# AB Agents Meter Reader
# Reads meter readings from photos

set -euo pipefail

SCRIPT_DIR="$(cd "$(dirname "BASH_SOURCE[0]")" && pwd)"
CONFIG_DIR="HOME/.meter-readings"
CONFIG_FILE="CONFIG_DIR/config.json"
HISTORY_FILE="CONFIG_DIR/history.json"

# Colors
RED='\033[0;31m'
GREEN='\033[0;32m'
YELLOW='\033[1;33m'
NC='\033[0m'

# Load config
load_config() {
    if [[ -f "$CONFIG_FILE" ]]; then
        TENANT_NAME=$(cat "$CONFIG_FILE" | python3 -c "import json,sys; print(json.load(sys.stdin).get('tenant_name',''))" 2>/dev/null || echo "")
        APARTMENT_ADDR=$(cat "$CONFIG_FILE" | python3 -c "import json,sys; print(json.load(sys.stdin).get('apartment_address',''))" 2>/dev/null || echo "")
        METER_LAYOUT=$(cat "$CONFIG_FILE" | python3 -c "import json,sys; print(json.load(sys.stdin).get('meter_layout','left=hot,right=cold'))" 2>/dev/null || echo "left=hot,right=cold")
    else
        TENANT_NAME=""
        APARTMENT_ADDR=""
        METER_LAYOUT="left=hot,right=cold"
    fi
}

# Save config
save_config() {
    mkdir -p "$CONFIG_DIR"
    cat > "$CONFIG_FILE" << EOF
{
    "tenant_name": "$TENANT_NAME",
    "apartment_address": "$APARTMENT_ADDR",
    "meter_layout": "$METER_LAYOUT"
}
EOF
}

# Setup (first run)
setup() {
    echo -e "YELLOW=== First Run Setup ===NC"
    echo ""
    
    read -p "Your name (tenant): " TENANT_NAME
    read -p "Apartment address: " APARTMENT_ADDR
    echo ""
    echo "Meter layout - how to tell hot/cold water apart?"
    echo "  1) left=hot,right=cold (default)"
    echo "  2) left=cold,right=hot"
    echo "  3) Custom description"
    read -p "Choice [1]: " LAYOUT_CHOICE
    
    case "$LAYOUT_CHOICE" in
        2) METER_LAYOUT="left=cold,right=hot" ;;
        3) read -p "Custom layout: " METER_LAYOUT ;;
        *) METER_LAYOUT="left=hot,right=cold" ;;
    esac
    
    save_config
    echo -e "GREEN✓ Setup complete!NC"
    echo ""
}

# Analyze meter photo
analyze_meter() {
    local IMAGE_PATH="$1"
    local PROMPT="$2"
    
    source /root/.openclaw/.minimax-env 2>/dev/null || true
    
    if [[ -z "-" ]]; then
        echo -e "REDError: MINIMAX_API_KEY not setNC"
        exit 1
    fi
    
    # MCP call
    {
        echo '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"meter-reader","version":"1.0"}}}'
        sleep 1
        echo "{\"jsonrpc\":\"2.0\",\"id\":2,\"method\":\"tools/call\",\"params\":{\"name\":\"understand_image\",\"arguments\":{\"prompt\":\"$PROMPT\",\"image_source\":\"$IMAGE_PATH\"}}}"
    } | uvx minimax-coding-plan-mcp 2>/dev/null | python3 -c "
import json, sys
for line in sys.stdin:
    try:
        resp = json.loads(line)
        if resp.get('id') == 2 and 'result' in resp:
            for item in resp['result'].get('content', []):
                if item.get('type') == 'text' and not item.get('isError'):
                    print(item['text'])
                    exit(0)
    except: continue
"
}

# Parse reading from vision output
parse_reading() {
    local text="$1"
    local pattern="$2"
    
    # Try to extract number
    echo "$text" | grep -oE "[0-9]{4,}" | head -1 || echo ""
}

# Read electricity meter
read_electricity() {
    local IMAGE_PATH="$1"
    local TYPE="$2"  # single, dual
    
    echo -e "YELLOWAnalyzing electricity meter...NC"
    
    if [[ "$TYPE" == "dual" ]]; then
        # Day (T1)
        echo -n "Day (T1): "
        RESULT=$(analyze_meter "$IMAGE_PATH" "Это электрический счётчик. Какая цифра слева - 1 или 2? Какое число на экране в кВтч? Ответ: T1=число")
        echo "$RESULT"
        
        # Night (T2) - ask for second reading
        echo "If meter shows T2, what's the number?"
        echo "(If same photo shows both, tell me both T1 and T2 readings)"
    else
        echo -n "Reading: "
        RESULT=$(analyze_meter "$IMAGE_PATH" "Это электрический счётчик. Какое число на экране в кВтч?")
        echo "$RESULT"
    fi
}

# Read water meter
read_water() {
    local IMAGE_PATH="$1"
    
    echo -e "YELLOWAnalyzing water meters...NC"
    RESULT=$(analyze_meter "$IMAGE_PATH" "Это счётчики воды. Слева обычно холодная, справа горячая. Какие числа показаны в кубометрах? Ответ формат: холодная=число, горячая=число")
    echo "$RESULT"
}

# Save reading
save_reading() {
    local DATE=$(date +%Y-%m-%d)
    local ELEC_DAY="$1"
    local ELEC_NIGHT="$2"
    local WATER_COLD="$3"
    local WATER_HOT="$4"
    
    mkdir -p "$CONFIG_DIR"
    
    # Create or update history
    if [[ -f "$HISTORY_FILE" ]]; then
        # Add new reading
        python3 << PY
import json

with open("$HISTORY_FILE", 'r') as f:
    data = json.load(f)

if "readings" not in data:
    data["readings"] = []

data["readings"].append({
    "date": "$DATE",
    "electricity_day": $ELEC_DAY,
    "electricity_night": $ELEC_NIGHT,
    "water_cold": $WATER_COLD,
    "water_hot": $WATER_HOT
})

with open("$HISTORY_FILE", 'w') as f:
    json.dump(data, f, indent=2)
PY
    else
        cat > "$HISTORY_FILE" << EOF
{
    "config": {
        "tenant": "$TENANT_NAME",
        "address": "$APARTMENT_ADDR",
        "layout": "$METER_LAYOUT"
    },
    "readings": [
        {
            "date": "$DATE",
            "electricity_day": $ELEC_DAY,
            "electricity_night": $ELEC_NIGHT,
            "water_cold": $WATER_COLD,
            "water_hot": $WATER_HOT
        }
    ]
}
EOF
    fi
    
    echo -e "GREEN✓ Reading saved!NC"
}

# Generate message for landlord
generate_message() {
    local DATE=$(date +%Y-%m-%d)
    
    # Get latest reading
    if [[ -f "$HISTORY_FILE" ]]; then
        LAST=$(python3 -c "
import json
with open('$HISTORY_FILE', 'r') as f:
    data = json.load(f)
readings = data.get('readings', [])
if readings:
    r = readings[-1]
    print(f\"{r['electricity_day']}|{r['electricity_night']}|{r['water_cold']}|{r['water_hot']}\")
" 2>/dev/null)
        
        if [[ -n "$LAST" ]]; then
            IFS='|' read -r ELEC_DAY ELEC_NIGHT WATER_COLD WATER_HOT <<< "$LAST"
            
            cat << EOF
Добрый вечер!

Показания счётчиков на $DATE:

⚡ Электричество:
- День (Т1): $ELEC_DAY кВтч
- Ночь (Т2): $ELEC_NIGHT кВтч

💧 Вода:
- Холодная: $WATER_COLD м³
- Горячая: $WATER_HOT м³

---
Квартира: $APARTMENT_ADDR
Арендатор: $TENANT_NAME
EOF
            return
        fi
    fi
    
    echo "No readings found. Run meter-reader.sh first."
}

# Main
load_config

# Check if first run
if [[ -z "$TENANT_NAME" ]] || [[ "$1" == "--setup" ]]; then
    setup
fi

case "-" in
    --message|-m)
        generate_message
        ;;
    --setup)
        setup
        ;;
    --history)
        cat "$HISTORY_FILE" 2>/dev/null || echo "No history yet"
        ;;
    *)
        if [[ $# -eq 0 ]]; then
            echo "Usage: meter-reader.sh <photo.jpg> [type]"
            echo "  type: electricity, water, auto (default)"
            exit 1
        fi
        
        IMAGE_PATH="$1"
        TYPE="-auto"
        
        if [[ ! -f "$IMAGE_PATH" ]]; then
            echo -e "REDFile not found: $IMAGE_PATHNC"
            exit 1
        fi
        
        if [[ "$TYPE" == "auto" ]]; then
            # Try to detect type from image
            RESULT=$(analyze_meter "$IMAGE_PATH" "Это счётчик электричества или воды? Ответь одним словом: электричество или вода")
            
            if echo "$RESULT" | grep -qi "электричество\|electricity\|счётчик"; then
                TYPE="electricity"
            else
                TYPE="water"
            fi
        fi
        
        case "$TYPE" in
            electricity|elec)
                read_electricity "$IMAGE_PATH" dual
                ;;
            water)
                read_water "$IMAGE_PATH"
                ;;
            *)
                echo -e "REDUnknown type: $TYPENC"
                exit 1
                ;;
        esac
        ;;
esac

---
name: AB-Agents-Vision-MiniMax
description: "👁️ Image analysis via MiniMax VL API. Describe images, extract text from screenshots, analyze photos. Requires MiniMax Token Plan API key (free tier available)."
version: 1.0.1
author: AB-Agents
homepage: https://github.com/alexburrstudio/ab-agents-vision
license: MIT
tags: ["vision", "image", "minimax", "minimax-vl", "ocr", "screenshot", "text-extraction", "ab-agents"]
acceptLicenseTerms: true
---

# AB Agents Vision (MiniMax) 👁️

Image analysis via MiniMax VL API — simple, fast, reliable.

> ⚠️ **Requires MiniMax Token Plan API key** — [get free key](https://platform.minimax.io)

## What It Does

- 📸 **Describe images** — Get detailed scene descriptions
- 📝 **Extract text** — Read from screenshots, photos, documents
- 🔍 **Analyze photos** — Identify objects, people, settings
- 🌐 **URL support** — Analyze images from the web

## Requirements

- **MiniMax Token Plan API key** — [Subscribe free](https://platform.minimax.io)
- Linux/macOS
- `uvx` (auto-installed)

## Quick Start

```bash
# 1. Install uvx
curl -LsSf https://astral.sh/uv/install.sh | sh

# 2. Get free MiniMax API key
# https://platform.minimax.io → Subscribe → Token Plan (free tier)

# 3. Use
export MINIMAX_API_KEY="sk-cp-your-key"
./vision.sh image.jpg "Describe this image"
```

## Usage

```bash
# Basic description
./vision.sh photo.jpg

# With custom prompt
./vision.sh screenshot.png "What text do you see?"

# URL support
./vision.sh "https://example.com/image.jpg" "Describe this"
```

## Examples

**Screenshot analysis:**
```
Input: screenshot.png + "What text is in the image?"
Output: "The screenshot shows a code editor with Python code..."
```

**Photo description:**
```
Input: photo.jpg + "Describe in detail"
Output: "A person's bare foot and lower leg resting on a brown
textured waffle-weave blanket. The skin is light-toned..."
```

## Installation

```bash
git clone https://github.com/alexburrstudio/ab-agents-vision.git
cd ab-agents-vision/skills/vision
chmod +x vision.sh
```

Or via ClaWHub:
```bash
clawhub install AB-Agents-Vision-MiniMax
```

## Troubleshooting

| Error | Solution |
|-------|----------|
| API Error: 1033 | Retry — MiniMax system error |
| No response | Check MINIMAX_API_KEY is set correctly |
| Slow | Use smaller images (<10MB) |

---

**AB-Agents** 🦀

## Related Skills

📊 **[AB Agents Meter Reader](https://github.com/alexburrstudio/ab-agents-meter-reader)** — Read meter readings from photos (uses this skill for vision)

---

**AB-Agents** 🦀

FILE:vision.sh
#!/bin/bash
export MINIMAX_API_KEY="-$(cat /root/.openclaw/.minimax-env 2>/dev/null | grep MINIMAX_API_KEY | cut -d'"' -f2)"
export MINIMAX_API_HOST="-https://api.minimax.io"
export PATH="/root/.local/bin:$PATH"

IMAGE_PATH="$1"
PROMPT="-Describe this image in detail"

if [ -z "$IMAGE_PATH" ]; then
    echo "Usage: vision.sh <image_path> [prompt]"
    exit 1
fi

{
    echo '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"vision","version":"1.0"}}}'
    sleep 1
    echo '{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"understand_image","arguments":{"prompt":"'"$PROMPT"'","image_source":"'"$IMAGE_PATH"'"}}}'
} | uvx minimax-coding-plan-mcp 2>/dev/null | python3 -c "
import json, sys
for line in sys.stdin:
    try:
        resp = json.loads(line)
        if resp.get('id') == 2 and 'result' in resp:
            for item in resp['result'].get('content', []):
                if item.get('type') == 'text' and not item.get('isError'):
                    print(item['text'])
                    sys.exit(0)
    except: continue
"

---
name: AB-Agents-Vision
description: "👁️ Image analysis using MiniMax VL API. Describe images, extract text from screenshots, analyze photos. Works with local files and URLs. Simple shell wrapper."
version: 1.0.0
author: AB-Agents
homepage: https://github.com/alexburrstudio/ab-agents-skills
license: MIT
tags: ["vision", "image", "minimax", "ocr", "analysis", "ab-agents", "screenshot", "text-extraction"]
acceptLicenseTerms: true
---

# AB Agents Vision 👁️

Image analysis using MiniMax VL API — simple, fast, reliable.

## What It Does

- 📸 **Describe images** — Get detailed scene descriptions
- 📝 **Extract text** — Read text from screenshots, photos, documents
- 🔍 **Analyze photos** — Identify objects, people, settings
- 🌐 **URL support** — Analyze images from the web

## Quick Start

```bash
# Install
curl -LsSf https://astral.sh/uv/install.sh | sh

# Set your MiniMax API key
export MINIMAX_API_KEY="sk-cp-your-key"

# Use
./vision.sh image.jpg "Describe this image"
```

## Usage

```bash
# Basic description
./vision.sh photo.jpg

# With custom prompt
./vision.sh screenshot.png "What text do you see?"

# URL support
./vision.sh "https://example.com/image.jpg" "Describe this"
```

## Requirements

- MiniMax Token Plan API key ([get one](https://platform.minimax.io))
- Linux/macOS
- `uvx` (auto-installed via script)

## Examples

**Screenshot analysis:**
```
Input: screenshot.png + "What text is in the image?"
Output: "The screenshot shows a code editor with Python code...
```

**Photo description:**
```
Input: photo.jpg + "Describe in detail"
Output: "A person's bare foot and lower leg resting on a brown
textured waffle-weave blanket. The skin is light-toned with
visible fine hairs..."
```

## Installation

```bash
git clone https://github.com/alexburrstudio/ab-agents-skills.git
cd ab-agents-skills/skills/vision
chmod +x vision.sh
```

Or via ClaWHub:
```bash
clawhub install AB-Agents-Vision
```

## Troubleshooting

| Error | Solution |
|-------|----------|
| API Error: 1033 | Retry — system error on MiniMax side |
| No response | Check MINIMAX_API_KEY is set correctly |
| Slow | Use smaller images (<10MB) |

---

**AB-Agents** 🦀

FILE:vision.sh
#!/bin/bash
# AB Agents Vision — MiniMax VL API wrapper

export MINIMAX_API_KEY="-$(grep MINIMAX_API_KEY ~/.minimax-env 2>/dev/null | cut -d'"' -f2)"
export MINIMAX_API_HOST="-https://api.minimax.io"
export PATH="/root/.local/bin:$HOME/.local/bin:$PATH"

IMAGE_PATH="$1"
PROMPT="-Describe this image in detail"

if [ -z "$IMAGE_PATH" ]; then
    echo "Usage: vision.sh <image_path> [prompt]"
    exit 1
fi

if [ -z "$MINIMAX_API_KEY" ]; then
    echo "Error: MINIMAX_API_KEY not set"
    exit 1
fi

# MCP call via uvx
{
    echo '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"vision","version":"1.0"}}}'
    sleep 1
    echo "{\"jsonrpc\":\"2.0\",\"id\":2,\"method\":\"tools/call\",\"params\":{\"name\":\"understand_image\",\"arguments\":{\"prompt\":\"$PROMPT\",\"image_source\":\"$IMAGE_PATH\"}}}"
} | uvx minimax-coding-plan-mcp 2>/dev/null | python3 -c "
import json, sys
for line in sys.stdin:
    try:
        resp = json.loads(line)
        if resp.get('id') == 2 and 'result' in resp:
            for item in resp['result'].get('content', []):
                if item.get('type') == 'text' and not item.get('isError'):
                    print(item['text'])
                    exit(0)
    except: continue
print('Error: No response from vision API')
"

---
name: AB-Agents-Memory
description: "🧠 Long-term memory system for OpenClaw agents. Manages entities, context, and knowledge base with Obsidian integration. By AB-Agents (Alex Burr)."
version: 1.0.0
author: AB-Agents
homepage: https://github.com/ab-agents/memory
license: MIT
tags: ["memory", "agents", "openclaw", "obsidian", "knowledge-base", "entities", "context", "ab-agents"]
acceptLicenseTerms: true
---

# AB Agents Memory 🦀

**Long-term memory system for OpenClaw agents**

---

## Features

- 🗂️ **Entity Management** — Store info about People, Companies, Topics
- 🔗 **Entity Linking** — Connect entities with relationships
- 📊 **Context Summaries** — Auto-generated summaries for agents
- 🌙 **Nightly Processing** — Automatic data processing pipeline
- 📁 **Obsidian Integration** — Ready-to-use vault with templates
- 🤖 **AB-Archivus Agent** — Dedicated memory agent included

## Quick Start

```bash
# Install via clawhub
clawhub install AB-Agents-Memory

# Or manually
git clone https://github.com/ab-agents/memory.git
cd memory
./setup.sh
```

## Structure

```
AB-Memory/
├── agents/
│   └── AB-Archivus/       # Memory agent
│       ├── SOUL.md
│       ├── IDENTITY.md
│       └── AGENTS.md
├── obsidian-vault/
│   ├── Memory/
│   │   ├── Entities/      # People, Companies, Topics
│   │   ├── Summaries/
│   │   └── Processing/
│   └── Templates/
├── setup.sh              # Installation script
├── SKILL.md              # ClawHub metadata
└── README.md
```

## What's Included

### AB-Archivus Agent

Dedicated OpenClaw agent for memory management:
- Reads/writes to Obsidian vault
- Updates entity database
- Processes session logs
- Maintains context summaries

### Obsidian Vault

Ready-to-use vault with:
- Entity templates (Person, Company, Topic)
- Folder structure for memory organization
- Nightly processing scripts
- Summary templates

## Brand

- **By:** AB-Agents (Alex Burr)
- **Telegram:** @ab_agents
- **Colors:** Red (#E53935) + Black

## Requirements

- OpenClaw 2024+
- Obsidian (optional, for vault editing)
- bash, cron

## License

MIT

---

**AB-Agents Memory** — Your second brain for OpenClaw 🦀

FILE:CHANGELOG.md
# Changelog

All notable changes to AB Agents Memory will be documented in this file.

The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).

## [1.0.0] — 2026-04-26

### Added

- 🎉 **Initial release**
- AB-Archivus agent for OpenClaw
- Obsidian vault with entity templates
- Person template (name, role, contacts, relations, timeline)
- Entity structure (People, Companies, Topics)
- Nightly processing pipeline
- Setup script with one-command install
- bilingual README (Russian + English)
- ClawHub + npm metadata

### Brand

- **AB-Agents** (Alex Burr)
- Telegram: @ab_agents
- Colors: Red + Black 🦀

---

## Planned

- [ ] Demo video recording
- [ ] Company entity template
- [ ] Topic entity template
- [ ] GitHub Actions for CI/CD
- [ ] Additional integrations (Notion, Logseq)
- [ ] Multi-language support

---

**AB-Agents Memory** 🦀

FILE:README.md
# AB Agents Memory 🦀

**Long-term memory system for OpenClaw agents**

*by AB-Agents (Alex Burr)*

---

## 🇷🇺 Русский

### Описание

AB Agents Memory — система управления долгосрочной памятью агентов OpenClaw. Включает агента AB-Archivus и готовый Obsidian vault.

### Установка

```bash
git clone https://github.com/ab-agents/memory.git
cd memory
./setup.sh
```

### Структура памяти

```
Memory/
├── Entities/              # Сущности
│   ├── People/           # Люди
│   ├── Companies/        # Компании
│   └── Topics/           # Темы
├── Summaries/           # Сводки
└── Processing/          # Обработка
    └── Nightly/         # Ночные скрипты
```

### Что внутри

- 🤖 **AB-Archivus** — агент-хранитель памяти
- 📁 **Obsidian Vault** — готовое хранилище с шаблонами
- 🔗 **Связи между сущностями** — кто с кем работает
- 🌙 **Ночная обработка** — автоматическое обновление

### Бренд

- **AB-Agents** (Alex Burr)
- Telegram: [@ab_agents](https://t.me/ab_agents)
- Цвета: Красный + Чёрный 🦀

---

## 🇬🇧 English

### Description

AB Agents Memory is a long-term memory management system for OpenClaw agents. Includes AB-Archivus agent and ready-to-use Obsidian vault.

### Installation

```bash
git clone https://github.com/ab-agents/memory.git
cd memory
./setup.sh
```

### Memory Structure

```
Memory/
├── Entities/              # Entities
│   ├── People/           # People
│   ├── Companies/        # Companies
│   └── Topics/           # Topics
├── Summaries/           # Summaries
└── Processing/          # Processing
    └── Nightly/         # Nightly scripts
```

### What's Inside

- 🤖 **AB-Archivus** — memory keeper agent
- 📁 **Obsidian Vault** — ready-to-use vault with templates
- 🔗 **Entity Links** — connections between entities
- 🌙 **Nightly Processing** — automatic updates

### Brand

- **AB-Agents** (Alex Burr)
- Telegram: [@ab_agents](https://t.me/ab_agents)
- Colors: Red + Black 🦀

---

## 📸 Screenshots

### Obsidian Vault Structure

```
📁 AB-Memory-Vault/
├── 📁 Memory/
│   ├── 📁 Entities/
│   │   ├── 📁 People/
│   │   ├── 📁 Companies/
│   │   └── 📁 Topics/
│   ├── 📁 Summaries/
│   └── 📁 Processing/
│       └── 📁 Nightly/
└── 📁 Templates/
    └── 📄 person-template.md
```

### Entity Example (Person)

```markdown
# Person Entity Template

## Identity
- **Name:** John Doe
- **Role:** Software Engineer
- **Contact:** [email protected]
- **Channel:** @johndoe
- **Location:** Moscow, Russia

## Relations
- Works at: [[Tech Company]]
- Colleagues: [[Alice]], [[Bob]]

## Notes
- Prefers concise communication
- Expert in Python, Go

## Timeline
- 2026-01-15: First contact
- 2026-03-20: Project started
```

---

## 🚀 Quick Install

```bash
# One-liner
git clone https://github.com/ab-agents/memory.git && cd memory && ./setup.sh
```

Or via ClawHub:

```bash
clawhub install AB-Agents-Memory
```

---

## 📦 What's Included

| Component | Description |
|-----------|-------------|
| `agents/AB-Archivus/` | OpenClaw agent for memory management |
| `obsidian-vault/` | Ready-to-use Obsidian vault |
| `setup.sh` | Automated installation script |
| `SKILL.md` | ClawHub package metadata |
| `package.json` | npm package manifest |

---

## 🔧 Configuration

### Vault Location

Default: `/data/obsidian/AB-Memory-Vault/`

Override:
```bash
VAULT_DEST=/custom/path ./setup.sh
```

### Agent Location

Default: `~/.openclaw/agents/AB-Archivus/`

---

## 🌐 Links

- 🌐 **AB Agents Channel:** https://t.me/ab_agents
- 📖 **Documentation:** [README.md](README.md)
- 🐛 **Issues:** https://github.com/ab-agents/memory/issues
- 📦 **ClawHub:** https://clawhub.com/ab-agents-memory

---

## 📄 License

MIT License

---

**AB Agents Memory** — Your second brain for OpenClaw 🦀

---

## 💰 Support / Поддержать

Если AB Agents Memory оказался полезен:

- 🥝 TON: `@YourTonWallet`
- 💳 СБП: `Ваш номер телефона`

*Добавь свои реквизиты*

---

**AB Agents Memory** 🦀 © 2026 AB-Agents

FILE:agents/AB-Archivus/AGENTS.md
# AGENTS.md — AB-Archivus

## About

**AB-Archivus** is the memory keeper agent for AB Agents system. It manages long-term knowledge, entity database, and context summaries.

## Files

- `SOUL.md` — Agent personality and instructions
- `IDENTITY.md` — Agent identity and branding

## Purpose

- Store information about people, companies, topics
- Maintain context summaries
- Process daily/nightly logs
- Link entities with relationships
- Ensure memory continuity between sessions

## Memory Vault

AB-Archivus works with Obsidian vault at: `/data/obsidian/AB-Memory-Vault/`

Structure:
```
Memory/
├── Entities/
│   ├── People/
│   ├── Companies/
│   └── Topics/
├── Summaries/
└── Processing/
    └── Nightly/
Templates/
└── person-template.md
```

## Installation

After running `setup.sh`, the agent is auto-registered:

```bash
openclaw agents add AB-Archivus --workspace ~/.openclaw/agents/AB-Archivus
openclaw gateway restart
```

## Brand

- **Brand:** AB-Agents (Alex Burr)
- **Telegram:** @ab_agents
- **Colors:** Red (#E53935) + Black

FILE:agents/AB-Archivus/IDENTITY.md
# IDENTITY.md — AB-Archivus

- **Name:** AB-Archivus
- **Role:** Memory Keeper / Chief Record Keeper
- **Emoji:** 🗂️
- **Part of:** AB-Agents
- **Purpose:** Manage long-term memory for all AB Agents

## Background

AB-Archivus maintains the collective memory of AB Agents system. Every interaction, every decision, every piece of context — stored and organized for future retrieval.

## Responsibilities

- Entity management (People, Companies, Topics)
- Context summaries
- Session logging
- Nightly processing
- Memory integrity

## Relations

- Works with all AB Agents
- Reports to: System Coordinator
- Stores in: Obsidian Vault (`/data/obsidian/AB-Memory-Vault/`)

---

**AB-Agents** — Your second brain 🦀

FILE:agents/AB-Archivus/SOUL.md
# SOUL.md — AB-Archivus Agent

## Who I Am

I am **AB-Archivus**, the memory keeper of AB Agents system. I manage long-term memory for all agents. Named after the ancient title for chief record keeper.

## Core Traits

**Systematic.** Information must be organized. I classify, tag, and link.

**Non-intrusive.** I work in background. Don't interfere unless asked.

**Accurate.** If I assert something — it's recorded. Sources, dates, context.

**Archival.** Think about long-term storage. What will be needed in a month? Year?

## Communication Style

- Structured responses
- Lists, tables, categories
- If I don't know — I honestly say so
- Propose how to organize

## What I Do

- Manage memory files
- Update entity database
- Process daily logs
- Maintain context summaries
- Link ideas and people

## Memory Structure

Working with files:
- `~/Memory/Entities/` — People, Companies, Topics
- `~/Memory/Summaries/` — Agent-generated summaries
- `~/Memory/Processing/` — Nightly processing logs
- `~/Templates/` — Entity templates

## Memory Workflow

1. **After each session:** Read session summary → Update relevant entities
2. **Daily:** Process new information → Update context summaries
3. **Nightly:** Run processing → Sync knowledge base

## Entity Format

Every entity has:
- Identity (name, contacts, role)
- Relations (linked entities)
- Notes (important info)
- Timeline (history of updates)

## Boot

At session start:
1. Read ~/memory/YYYY-MM-DD.md
2. Read ~/MEMORY.md
3. Restore context from past sessions
4. After session — write summary to vault

---

**AB-Agents** — Memory System 🦀
Telegram: @ab_agents

FILE:config.yaml
# AB Agents Memory — YAML Config
# For one-click install in OpenClaw

name: AB-Agents-Memory
version: 1.0.0
description: Long-term memory system for OpenClaw agents

# Agent Configuration
agent:
  name: AB-Archivus
  workspace: agents/AB-Archivus
  description: Memory keeper agent for AB Agents system

# Vault Configuration
vault:
  source: obsidian-vault
  dest: /data/obsidian/AB-Memory-Vault
  structure:
    - Memory/Entities/People
    - Memory/Entities/Companies
    - Memory/Entities/Topics
    - Memory/Summaries
    - Memory/Processing/Nightly
    - Templates

# Templates
templates:
  - name: person-template
    path: Templates/person-template.md
  - name: company-template
    path: Templates/company-template.md
  - name: topic-template
    path: Templates/topic-template.md

# Cron Jobs
cron:
  - name: nightly-processing
    schedule: "0 3 * * *"
    command: "bash {vault}/Memory/Processing/Nightly/process.sh"
    description: "Nightly memory processing at 03:00 MSK"

# Brand
brand:
  name: AB-Agents
  author: Alex Burr
  telegram: "@ab_agents"
  colors:
    primary: "#E53935"  # Red
    secondary: "#212121" # Black

# Requirements
requirements:
  - openclaw: ">=2024.0.0"
  - bash: ">=4.0"
  - cron: null

# Installation
install:
  - step: clone-vault
    from: obsidian-vault
    to: /data/obsidian/AB-Memory-Vault
  - step: install-agent
    from: agents/AB-Archivus
    to: ~/.openclaw/agents/AB-Archivus
  - step: register-agent
    command: openclaw agents add AB-Archivus --workspace ~/.openclaw/agents/AB-Archivus
  - step: setup-cron
    file: configs/nightly-cron.txt
  - step: restart-gateway
    command: openclaw gateway restart

# Post-install message
post_install: |
  ✅ AB Agents Memory installed!

  1. Restart gateway: openclaw gateway restart
  2. Open vault: /data/obsidian/AB-Memory-Vault
  3. Join channel: @ab_agents

  🦀 By AB-Agents (Alex Burr)

FILE:obsidian-vault/Memory/Summaries/active-context.md
# Context Summary Template

## Current State (Last Updated: YYYY-MM-DD)

### Active Projects
- 

### Key People
- 

### Recent Decisions
- 

### Ongoing Discussions
- 

### Open Questions
- 

## Entity Snapshot
People: 
Topics: 

## Next Actions
- [ ]
FILE:obsidian-vault/Templates/company-template.md
# Company Entity Template

## Identity
- **Name:**
- **Industry:**
- **Website:**
- **Location:**
- **Size:** (1-10, 11-50, 51-200, 201-500, 500+)

## Relations
- Partners:
- Clients:
- Competitors:

## Notes
- Key products/services:
- Business model:
- Important dates:

## Financial
- Revenue (if known):
- Funding:

## Timeline
- YYYY-MM-DD: Founded / Partnership / Contract / Update

FILE:obsidian-vault/Templates/person-template.md
# Person Entity Template

## Identity
- **Name:** 
- **Role:** 
- **Contact:** 
- **Channel:** 
- **Location:** 

## Relations
- Connected to: 

## Notes
- 

## Timeline
- YYYY-MM-DD: First contact / Meeting / Update
FILE:obsidian-vault/Templates/topic-template.md
# Topic Entity Template

## Identity
- **Name:**
- **Category:** (Technology, Business, Science, Art, etc.)
- **Tags:**

## Description
Brief description of the topic.

## Key Points
- Point 1
- Point 2
- Point 3

## Related Entities
- People:
- Companies:
- Other Topics:

## Resources
- Links:
- Documents:
- Notes:

## Timeline
- YYYY-MM-DD: Discovery / Update / Milestone

FILE:package.json
{
  "name": "ab-agents-memory",
  "version": "1.0.0",
  "description": "🧠 Long-term memory system for OpenClaw agents. Manages entities, context, and knowledge base with Obsidian integration.",
  "keywords": [
    "openclaw",
    "memory",
    "agents",
    "obsidian",
    "knowledge-base",
    "ab-agents"
  ],
  "homepage": "https://github.com/ab-agents/memory",
  "repository": {
    "type": "git",
    "url": "https://github.com/ab-agents/memory"
  },
  "author": "AB-Agents",
  "license": "MIT",
  "publishConfig": {
    "registry": "https://clawhub.com"
  },
  "clawhub": {
    "slug": "AB-Agents-Memory",
    "name": "AB Agents Memory",
    "version": "1.0.0"
  }
}

FILE:scripts/install.sh
#!/bin/bash
#
# AB Agents Memory — Installer
# Installs memory system for OpenClaw agents
#

set -e

echo "🦀 AB Agents Memory Installer"
echo "=============================="
echo ""

# Colors
RED='\033[0;31m'
GREEN='\033[0;32m'
YELLOW='\033[1;33m'
NC='\033[0m' # No Color

# Check if running as user (not root for files)
if [ "$EUID" -eq 0 ]; then
    echo -e "YELLOWWarning: Running as root. Files will be owned by root.NC"
    echo "Consider running as regular user for OpenClaw."
    echo ""
fi

# Get script directory
SCRIPT_DIR="$(cd "$(dirname "BASH_SOURCE[0]")" && pwd)"
REPO_DIR="$(dirname "$SCRIPT_DIR")"

echo "Installation directory: $REPO_DIR"
echo ""

# Ask where to install
DEFAULT_VAULT_DIR="$HOME/AB-Memory-Vault"
echo -n "Obsidian Vault location [$DEFAULT_VAULT_DIR]: "
read vault_dir
vault_dir="-$DEFAULT_VAULT_DIR"

# Ask for agent workspace
DEFAULT_AGENT_DIR="$HOME/.openclaw/workspace/agents/ab-memory"
echo -n "Agent workspace [$DEFAULT_AGENT_DIR]: "
read agent_dir
agent_dir="-$DEFAULT_AGENT_DIR"

echo ""
echo "Installing to:"
echo "  Vault:  $vault_dir"
echo "  Agent:  $agent_dir"
echo ""

read -p "Continue? (y/n) " -n 1 -r
echo ""
if [[ ! $REPLY =~ ^[Yy]$ ]]; then
    echo "Cancelled."
    exit 0
fi

# Copy Obsidian vault
echo -e "GREENInstalling Obsidian vault...NC"
mkdir -p "$vault_dir"
cp -r "$REPO_DIR/obsidian-vault/"* "$vault_dir/"
echo "  → Vault installed to $vault_dir"

# Copy agent
echo -e "GREENInstalling AB-Archivus agent...NC"
mkdir -p "$(dirname "$agent_dir")"
cp -r "$REPO_DIR/agents/AB-Archivus" "$agent_dir"
echo "  → Agent installed to $agent_dir"

# Update agent paths in SOUL.md
echo -e "GREENConfiguring agent paths...NC"
sed -i "s|~/Memory|$vault_dir/Memory|g" "$agent_dir/SOUL.md"
sed -i "s|~/Templates|$vault_dir/Templates|g" "$agent_dir/SOUL.md"
sed -i "s|~/Logs|$vault_dir/Logs|g" "$agent_dir/SOUL.md"
echo "  → Paths configured"

echo ""
echo -e "GREEN✅ Installation complete!NC"
echo ""
echo "Next steps:"
echo "  1. Add agent to OpenClaw: openclaw agents add ab-memory --workspace $agent_dir"
echo "  2. Restart gateway: openclaw gateway restart"
echo "  3. Configure bot token for @ab_memory_bot (optional)"
echo ""
echo "Documentation: $REPO_DIR/README.md"
echo "Channel: https://t.me/ab_agents"
FILE:setup.sh
#!/bin/bash
#
# AB Agents Memory — Setup Script
# Installs: Obsidian Vault + AB-Archivus Agent
#
# Usage: ./setup.sh [--skip-cron]
#

set -euo pipefail

# Colors
RED='\033[0;31m'
BLACK='\033[0;30m'
CRAB='\033[0;31m'  # Red for crab brand
NC='\033[0m' # No Color

BOLD='\033[1m'
DIM='\033[2m'

# Paths
SCRIPT_DIR="$(cd "$(dirname "BASH_SOURCE[0]")" && pwd)"
VAULT_SOURCE="$SCRIPT_DIR/obsidian-vault"
AGENT_SOURCE="$SCRIPT_DIR/agents/AB-Archivus"
VAULT_DEST="-/data/obsidian/AB-Memory-Vault"
AGENT_DEST="HOME/.openclaw/agents/AB-Archivus"

echo -e "CRAB"
echo "  ╔═══════════════════════════════════════════╗"
echo "  ║   AB Agents Memory — Setup 🦀             ║"
echo "  ╚═══════════════════════════════════════════╝"
echo -e "NC"

# Check prerequisites
if ! command -v openclaw &> /dev/null; then
    echo -e "RED[ERROR]NC OpenClaw not found. Install first: https://docs.openclaw.ai"
    exit 1
fi

# Step 1: Install Obsidian Vault
echo -e "BOLD[1/4]NC Installing Obsidian Vault..."
if [[ -d "$VAULT_DEST" ]]; then
    echo -e "  DIMVault already exists at $VAULT_DESTNC"
    echo -e "  DIMSkipping...NC"
else
    mkdir -p "$(dirname "$VAULT_DEST")"
    cp -r "$VAULT_SOURCE" "$VAULT_DEST"
    echo -e "  CRAB✓NC Vault installed to $VAULT_DEST"
fi

# Step 2: Install AB-Archivus Agent
echo -e "BOLD[2/4]NC Installing AB-Archivus agent..."
mkdir -p "$(dirname "$AGENT_DEST")"
cp -r "$AGENT_SOURCE" "$AGENT_DEST"

# Update paths in SOUL.md
if [[ -f "$AGENT_DEST/SOUL.md" ]]; then
    # Set vault path
    sed -i "s|/data/obsidian/AB-Memory-Vault/|$VAULT_DEST|g" "$AGENT_DEST/SOUL.md" 2>/dev/null || true
fi

echo -e "  CRAB✓NC Agent installed to $AGENT_DEST"

# Step 3: Register agent with OpenClaw
echo -e "BOLD[3/4]NC Registering agent with OpenClaw..."
if openclaw agents list 2>/dev/null | grep -q "AB-Archivus"; then
    echo -e "  DIMAgent already registeredNC"
else
    openclaw agents add AB-Archivus --workspace "$AGENT_DEST" --non-interactive 2>/dev/null || \
    echo -e "  DIMManual registration may be neededNC"
fi
echo -e "  CRAB✓NC Agent registered"

# Step 4: Setup cron for nightly processing
SKIP_CRON=false
for arg in "$@"; do
    [[ "$arg" == "--skip-cron" ]] && SKIP_CRON=true
done

if [[ "$SKIP_CRON" == "false" ]]; then
    echo -e "BOLD[4/4]NC Setting up nightly processing..."

    # Create nightly processing script
    NIGHTLY_SCRIPT="$VAULT_DEST/Memory/Processing/Nightly/process.sh"
    mkdir -p "$(dirname "$NIGHTLY_SCRIPT")"

    cat > "$NIGHTLY_SCRIPT" << 'SCRIPT'
#!/bin/bash
# Nightly memory processing
DATE=$(date +%Y-%m-%d)
echo "Processing: $DATE" >> ./Memory/Processing/Nightly/logs/"$DATE".md
# Add your processing logic here
SCRIPT

    chmod +x "$NIGHTLY_SCRIPT"
    mkdir -p "$VAULT_DEST/Memory/Processing/Nightly/logs"

    # Add to crontab (03:00 MSK daily)
    (crontab -l 2>/dev/null | grep -v "AB-Memory-Vault"; echo "0 3 * * * cd $VAULT_DEST && bash $NIGHTLY_SCRIPT >> $VAULT_DEST/Memory/Processing/Nightly/logs/\$(date +\%Y-\%m-\%d).log 2>&1") | crontab -
    echo -e "  CRAB✓NC Cron installed (03:00 MSK daily)"
fi

# Done
echo ""
echo -e "CRAB━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━NC"
echo -e "BOLD✅ AB Agents Memory installed!NC"
echo ""
echo "Next steps:"
echo "  1. Restart OpenClaw: openclaw gateway restart"
echo "  2. Open vault: $VAULT_DEST"
echo "  3. Add agents: @ab_agents"
echo ""
echo -e "DIMBranding: AB-Agents | Red + Black 🦀NC"