Agents

Autonomous AI agents and the frameworks used to build them.

27 listed. Get a project listed →

All agents

OpenClaw

Other (see repo)

Open-source general-purpose personal agent that runs on your own machine and lives in your messaging apps — with tools, persistent memory, and automations wherever you already talk.

390k⤓ 2.5M/wk · npm▲ +0.1%Can run locally

Nous Research's open-source personal agent — persistent memory, tool use, and multi-channel deployment — among the most heavily used open agents by independent inference-volume measures.

248k⤓ 34k/wk · pypi▲ +0.9%Can run locally

langchain

MIT

A widely adopted framework for building LLM applications and agents, with composable abstractions for prompts, memory, tool calling, and retrieval across many model providers.

147k⤓ 40M/wk · pypi▲ +0.3%Can run locally

TradingAgents

Apache-2.0

A multi-agent framework for LLM-driven financial trading research: specialist analyst agents for fundamentals, sentiment, news and technicals feed a bull-versus-bear debate, then risk management and a trader agent synthesise a decision. Model providers and market-data vendors are pluggable.

108k⤓ 376/wk · pypi▲ +2.1%Can run locally

agno

Apache-2.0

A Python framework and platform for building, running, and managing AI agents, with built-in memory, tools, and multi-agent teams.

42k⤓ 382k/wk · pypi▲ +0.3%Can run locally

letta

Apache-2.0

A platform for building stateful agents with long-term memory that persists across sessions, letting agents learn and improve over time (formerly MemGPT).

25k⤓ 3.8k/wk · pypi▲ +0.4%Can run locally

goose

Apache-2.0

Open-source agent that runs on your own machine as a desktop app or terminal CLI and does real work rather than suggesting it — installing dependencies, editing and testing code, and driving external systems through MCP extensions, with any model you point it at. Built at Block and now governed by the Linux Foundation's Agentic AI Foundation.

55k▲ +0.5%Can run locally