freddy-schuetz/n8n-claw

OpenClaw-inspired autonomous AI agent built entirely in n8n. Adaptive RAG-powered memory, Skills via MCP templates, Expert Agents with delegated sub-agents, proactive task management, media understanding - self-hosted with one setup script

What it solves

n8n-claw is a self-hosted AI agent that provides a centralized, private alternative to commercial AI assistants. It solves the problem of fragmented AI tools by integrating communication channels (Telegram, Slack, Teams), long-term memory, task management, and web automation into a single, self-hosted infrastructure.

How it works

The project uses n8n as the orchestration engine, PostgreSQL for data storage, and Claude (or other LLMs) as the reasoning core. The agent operates through a series of specialized workflows that handle different capabilities:

  • Communication: Interacts via Telegram or a Webhook API that can be adapted for other platforms like Slack or Teams.
  • Memory & Knowledge: Employs a hybrid search (semantic + full-text) and a knowledge graph to track entities and relationships over time.
  • Tooling: Uses the Model Context Protocol (MCP) to execute "skills" (API integrations), perform web searches via SearXNG, and read webpages via Crawl4AI.
  • Automation: Includes a "heartbeat" system for recurring tasks, proactive reminders, and background monitoring.
  • Browser Control: Uses Browser Use with Chromium to perform agentic actions on websites, including handling interactive 2FA.

Who it’s for

Users who want a fully private, self-hosted AI agent with deep integration into their own data and workflows, and those who prefer to manage their AI assistant via a familiar chat interface like Telegram.

Highlights

  • Self-Hosted: Runs on your own VPS using Docker, ensuring data privacy.
  • Hybrid Memory: Combines semantic search, entity tracking, and time-decay to maintain relevant long-term context.
  • Knowledge Graph: Automatically maps connections between people, companies, and events.
  • Agentic Web Browsing: Can sign up for newsletters, fill forms, and handle 2FA logins interactively.
  • MCP Integration: Supports installing pre-built skills or building new API integrations on demand.
  • Proactive Assistant: Manages tasks, sends scheduled reminders, and monitors background processes.
  • Multi-Channel: Accessible via Telegram, HTTP API, and MCP-compatible clients like Claude Desktop or Cursor.

Related

  • Project
    n8n-io/n8n

    n8n is a fair-code platform for building and deploying AI agents and automated workflows, connecting over 1,500 integrations with flexible model support.

  • Project
    fuyuxiang/echo-agent

    A self-hosted, long-running AI agent featuring a four-layer cognitive memory system and a self-evolution engine to improve skills from real-world usage.

  • Project
    swarmclawai/swarmclaw

    A self-hosted AI agent runtime and multi-agent framework for creating autonomous agent swarms with durable memory, scheduling, and multi-provider LLM support.

  • Project
    vercel-labs/personal-agent-template

    An open-source template for deploying a personal AI agent with long-term memory and cross-platform access via web chat, Slack, and iMessage.

  • Project
    zeroclaw-labs/zeroclaw

    A self-hosted AI agent runtime written in Rust that connects various LLM providers to multiple communication channels and local system tools while prioritizing user data ownership.