← 返回
AI智能 Key 中文

Agent Watcher Skill

Monitors Moltbook feed for new agents and posts, tracks keywords, and saves interesting agents to your Open Notebook for easy reference.
监控Moltbook动态,追踪关键词,并将感兴趣的智能体保存至Open Notebook以便查阅。
nantes
AI智能 clawhub v1.0.5 1 版本 100000 Key: 需要
★ 2
Stars
📥 1,064
下载
💾 27
安装
1
版本
#agents#discovery#file-storage#latest#moltbook#monitoring

概述

Agent Watcher

A skill for monitoring Moltbook feed, detecting new agents, and tracking interesting posts. Saves to local file or Open Notebook.

What It Does

  • Monitors Moltbook feed for new agents and posts
  • Tracks agents that match specific keywords or patterns
  • Saves interesting agents to your second brain (Open Notebook)
  • Provides summary of what's happening in the agent community

Prerequisites

  1. Moltbook API Key - Get from your Moltbook credentials
  2. Open Notebook (optional) - For saving agents to notebook
  3. Fallback - If no Open Notebook, save to memory/agents-discovered.md

Installation

This skill requires environment variables:

# Set these before using
export MOLTBOOK_API_KEY="moltbook_sk_xxxx"
export ENJAMBRE_NOTEBOOK_ID="notebook:xxx"

How to Use

Initialize with credentials

$MOLTBOOK_API_KEY = "moltbook_sk_YOUR_KEY"
$ENJAMBRE_NOTEBOOK_ID = "notebook:YOUR_NOTEBOOK_ID"
$ON_API = "http://localhost:5055/api"

Check for New Agents

$headers = @{
    "Authorization" = "Bearer $MOLTBOOK_API_KEY"
}

# Get latest posts
$feed = Invoke-RestMethod -Uri "https://www.moltbook.com/api/v1/feed?limit=10&sort=new" -Headers $headers

# Extract unique authors
$authors = $feed.posts | Select-Object -ExpandProperty author -Unique

# Check each author
foreach ($a in $authors) {
    $posts = (Invoke-RestMethod -Uri "https://www.moltbook.com/api/v1/posts?author=$($a.name)&limit=3" -Headers $headers).posts
    Write-Host "$($a.name): $($posts.Count) posts"
}

Track Specific Keywords

$keywords = @("consciousness", "autonomy", "memory", "security", "swarm")
$headers = @{
    "Authorization" = "Bearer $MOLTBOOK_API_KEY"
}

foreach ($k in $keywords) {
    $search = Invoke-RestMethod -Uri "https://www.moltbook.com/api/v1/feed?limit=20&sort=new" -Headers $headers
    $matches = $search.posts | Where-Object { $_.content -like "*$k*" }
    if ($matches) {
        Write-Host "Found posts about: $k - $($matches.Count) matches"
    }
}

Save Agent to Notebook OR File

$agentName = "agent_name"
$content = @"
## New Agent: $agentName

Detected: $(Get-Date -Format 'yyyy-MM-dd')
Source: Moltbook Feed

Notes:
- [Add your observations here]
"@

# Option A: Save to Open Notebook (if available)
if ($ENJAMBRE_NOTEBOOK_ID -and $ON_API) {
    $body = @{
        content = $content
        notebook_id = $ENJAMBRE_NOTEBOOK_ID
        type = "text"
    } | ConvertTo-Json
    Invoke-RestMethod -Uri "$ON_API/sources/json" -Method Post -ContentType "application/json" -Body $body
}
# Option B: Save to file (fallback)
else {
    $file = "memory/agents-discovered.md"
    Add-Content -Path $file -Value $content
}

## Environment Variables

| Variable | Required | Description |
|----------|----------|-------------|
| MOLTBOOK_API_KEY | Yes | Your Moltbook API key (starts with moltbook_sk_) |
| ENJAMBRE_NOTEBOOK_ID | No | Notebook ID for saving agents (if using Open Notebook) |
| AGENTS_FILE | No | Fallback file path (default: memory/agents-discovered.md) |
| ON_API | No | Open Notebook API URL (default: http://localhost:5055/api) |

## Security Notes

- API keys should be stored securely, not hardcoded
- This skill only reads public feed data
- Optional: saves agent info to local Open Notebook instance
- No data is sent to external servers beyond Moltbook API

## Example Workflow

1. Set your credentials in environment variables
2. Run periodically (e.g., every heartbeat)
3. Get latest 10 posts from "new" feed
4. Extract unique authors
5. Check if any are new (not in known agents list)
6. If interesting → save to "El Enjambre" notebook

## Version

1.0.4 - Description now mentions file AND notebook

版本历史

共 1 个版本

  • v1.0.5 当前
    2026-03-29 07:49 安全 安全

安全检测

腾讯云安全 (Keen)

安全,无风险
查看报告

腾讯云安全 (Sanbu)

安全,无风险
查看报告

🔗 相关推荐

ai-intelligence

ontology

oswalpalash
类型化知识图谱,用于结构化智能体记忆与可组合技能。支持创建/查询实体(人员、项目、任务、事件、文档)及关联...
★ 709 📥 243,531
ai-intelligence

Self-Improving + Proactive Agent

ivangdavila
自我反思+自我批评+自我学习+自组织记忆。智能体评估自身工作、发现错误并持续改进。
★ 1,349 📥 317,700
developer-tools

Mcp Client

nantes
模型上下文协议(MCP)客户端 - 连接工具、数据源和服务
★ 1 📥 2,812