Features
Self-improving agents that evaluate and correct responses
Multi-agent orchestration: AgentTeam, AgentFlow, AgentOS
AgentClaw dashboard for visual management
One-line install via curl, pip, pipx, uvx
100+ LLM providers (OpenAI, Anthropic, Google, Ollama, Groq)
140+ built-in tools (web search, file ops, APIs, databases)
Persistent memory and RAG knowledge injection
A2A protocol and MCP server for enterprise integration
Docker and cloud deployment support
Telegram, Discord, Slack, WhatsApp bot integration
Low-code setup in 5 lines of Python
Agent spawning and cloning for self-organization
Browser automation via CLI
Sandbox for secure code execution
Plugin and skill management system
Multi-agent orchestration with handoffs
Sandbox Agents for containerized long-running tasks
Agent-as-tool delegation (v0.15.0+)
Realtime Agents with gpt-realtime-2 voice support (v0.17.6+)
Human-in-the-loop mechanisms
Automatic session history management
Built-in tracing for debugging
Provider-agnostic LLM support (100+ models via LiteLLM)
Redis session support (optional, v0.17.6+)
Instructions, tools, and guardrails configuration
Jupyter notebook compatibility
Supports OpenAI Responses and Chat Completions APIs
