AI channel
AI agents

Open-source AI agents

An AI agent is a program that uses a language model to decide what to do next, calls tools such as a browser, a code interpreter or an API, and repeats until the task is done. The projects here range from frameworks for building your own agents to finished agents that write code, browse the web or run business workflows.

How to choose

Compare three things. Which models it can use: a single provider, or any OpenAI-compatible endpoint including local ones. How it limits what the agent may do: sandboxing, approval steps and permission lists. And whether it keeps memory between steps and sessions. An agent that runs code or shell commands should run in a sandbox, so check that before giving it access to your machine.

Most-starred projects

2,689–2,712 of 3,685
526

Open source evals for physical AI. Run any LLM/VLA on any arm/humanoid against any real/sim benchmark.

Python
526

SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks?

Python
agencyenterprise

PromptInject

525

PromptInject is a framework that assembles prompts in a modular fashion to provide a quantitative analysis of the robustness of LLMs to adversarial prompt attacks. 🏆 Best Paper Awards @ NeurIPS ML Safety Workshop 2022

Python
agenvoy

Agenvoy

525

Self-hosted AI agent harness in a single Go binary — writes, sandbox-tests and repairs its own tools, and lets Claude Code, Codex and any MCP client build and share them.

Go
Claw-Company

clawcompany

525

Open source AI company OS. 38 roles, 6 templates, 4-layer memory. Your AI company remembers everything — locally. npx clawcompany

TypeScript
SearchSavior

OpenArc

525

Inference engine for Intel devices. Serve LLMs, VLMs, Whisper, Kokoro-TTS, Embedding and Rerank models over OpenAI endpoints.

Python
524

A skill for AI agents: search the web with SearXNG, browse with Camofox, bypass protections with CloakBrowser. Anti-hallucination by design. Self-hosted, free, unlimited.

JavaScript
524

透过 Claude Code 的源码,我们学到了什么?

netboxlabs

pktvisor

524

pktvisor is a dynamic network observability agent that smartly analyzes network traffic and generates opentelemetry metrics

C++
unbody-io

unbody

524

The Supabase of AI era. A modular, open-source backend for building AI-native software — designed for knowledge, not static data.

TypeScript
523

An in-the-wild benchmark for AI agents in the production harness.

Python
mco-org

mco

523

CLI-first orchestration for AI coding agents: run selected agents and models in parallel, compare raw answers, and coordinate review or implementation workflows.

Python
523

A general framework for distilling human-created multimodal resources into reusable, executable skills that AI agents can browse, compose, and run, validated across diverse domains including web, PowerPoint, Excel, Blender, CAD, Unreal Engine 5, and REAPER-based music production.

Python
addsumtech

slides_maker

522

Turn papers, code, and docs into presentation-ready, natively editable PPTX in Codex / Claude Code. Native charts and equations, speaker notes, click-build animations, and an independent critic review before delivery.

Python
kentjuno

ainovel-cli

522

CLI sáng tác tiểu thuyết AI đa agent — Bản tiếng Việt của voocel/ainovel-cli

Go
LeDat98

NexusRAG

522

Hybrid RAG system combining vector search, knowledge graph (LightRAG), and cross-encoder reranking — with Docling document parsing, visual intelligence (image/table captioning), agentic streaming chat, and inline citations. Powered by Gemini or local Ollama models.

Python

🧠 Context Engineering Research - Not just another agent collection, but using research and context engineering to function as a collective. Hub-and-spoke coordination through Claude Code.

JavaScript
522

AI Agent 工程师面试资料库 - 国内大厂面经、岗位要求、高频面试题

HTML
ZJU-LLMs

Agent-Kernel

522

A MicroKernel Multi-Agents System Framework for Adaptive Social Simulation Powered by LLMs

Python
521

🤖 AI social-media automation skill | Self-evolution workflows | AI 社交媒体自动化 skill | 自我进化工作流

Shell
caura-ai

caura

519

Caura (formerly MemClaw) — governed shared memory for AI agent fleets. Multi-agent, multi-tenant, MCP-native. Trust tiers, keystone policies, audit trails, knowledge graph, self-improving retrieval. Apache 2.0.

Python
519

Self-Programming AI Assistant. Capture, automate, and refine all your workflows.

Python
Cranot

roam-code

518

Local codebase intelligence CLI + MCP server for AI coding agents: SQLite code graph, 28 languages, 287 commands, 246 MCP tools, change-safety gates, audit evidence, zero API keys.

Python