Features
Task planning: decomposes complex tasks into steps
Three-tier memory with automatic Deep Dream distillation
Knowledge base: Markdown wiki with interactive knowledge graph
Skills system: one-click install from Skill Hub, GitHub, ClawHub
Custom skills creation via natural-language conversation
Tool system: file I/O, terminal, browser, scheduler, web search
Native MCP (Model Context Protocol) integration
Pluggable models: OpenAI, Claude, Gemini, DeepSeek, Qwen, GLM, Kimi, MiniMax, Doubao, LinkAI
Multi-channel: Web, WeChat, Feishu, DingTalk, Telegram, Slack, Discord, WeCom, QQ, Official Accounts
One-line install for Linux, macOS, Windows
24/7 operation on laptop or server
Self-evolution: post-session review, dream consolidation, source-code self-update
Try online: web demo available
Supports DeepSeek V4 Flash
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
