fastclaw-ai/fastclaw

Multi-Agent Framework

What it solves

FastClaw provides a lightweight, single-binary runtime for creating and managing multiple AI agents. It eliminates the complexity of manually setting up LLM connections, tool execution environments, and session management, allowing users to deploy agents with distinct personalities, memories, and skills that can be integrated into various communication channels.

How it works

FastClaw acts as an "Agent Factory" that manages agents through a centralized database (SQLite by default or Postgres) and a filesystem for skills. Each agent is defined by system files (like SOUL.md for personality and MEMORY.md for long-term facts) and can be configured with specific LLM providers and tools. The runtime handles the communication between the LLM and the agent's tools, utilizing either a local Docker sandbox or a cloud-based E2B sandbox to execute code and manage files securely.

Who it’s for

It is designed for developers and organizations that need to deploy AI agents into production environments, particularly those wanting to connect agents to IM platforms like Telegram, Discord, and Slack, or those building multi-tenant applications where agents need to be provisioned for different end-users.

Highlights

  • Multi-Agent Management: Create and manage multiple agents, each with its own model overrides, personality, and private skills.
  • Sandbox Isolation: Supports Docker and E2B sandboxes to safely execute tool calls and mirror files back to durable storage.
  • Broad LLM Support: Compatible with OpenAI, Anthropic, Ollama, OpenRouter, Groq, DeepSeek, Mistral, and any OpenAI-compatible API.
  • Channel Integration: Built-in bindings for Telegram, Discord, and Slack to allow end-users to interact with agents on their preferred platforms.
  • Extensible Tooling: Includes a plugin system (JSON-RPC), MCP server support, and the ability to install skills from ClawHub or GitHub.
  • Comprehensive API: Offers an OpenAI-compatible chat endpoint and a specialized Agent Communication Protocol (ACP) for discovery and session management.

Related

  • Project
    nextlevelbuilder/goclaw

    A multi-tenant AI agent platform built in Go that provides a unified gateway for 20+ LLM providers and 7 messaging channels with built-in multi-agent orchestration.

  • 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
    qhkm/zeptoclaw

    A fast, secure, and local-first personal AI assistant infrastructure written in Rust, providing sandboxed autonomy and multi-provider LLM support in a single 6MB binary.

  • Project
    zeroclaw-labs/zeroclaw

    A self-hosted agent runtime written in Rust that connects LLM providers to multiple communication channels and hardware tools while ensuring user ownership of data and infrastructure.

  • Project
    johnson7788/MultiUserClaw

    MultiUserClaw is a Docker‑based SaaS framework that lets you run isolated AI assistants (Hermes agents) for many users. It includes a React front‑end, a FastAPI gateway for auth, container management and LLM proxying, per‑user Hermes containers, PostgreSQL storage, and features like knowledge bases, skill store, cron jobs, multi‑channel bots, and multi‑model support.