Open-Source AI Agents Directory
Find open-source AI agents you can run yourself: coding, research, browser and workflow agents, plus frameworks for building your own. Compare them by GitHub stars, licence and recent activity, see what each agent can touch on your machine, and read the dated scan report where one has run.
Coding agents, research agents, browser agents, workflow builders and agent frameworks, all open source.
Chainlit Langgraph
A seamless integration of Chainlit with LangGraph. Ideal for fast prototyping multi-LLM chatbot or AI agents.
CoBRA
[🏆 CHI26 Best Paper] CoBRA: Reproducible control of LLM agent behavior via classic social science experiments
AI Bioworkflow
Auditable bioinformatics workflow compiler: Catalog-bound planning, Workflow IR, and validated WDL 1.0.
AI Bot Groupmate
Estrella Bot · QQ 群 AI 群友机器人。把 QQ 小号接给任意 OpenAI 兼容模型(本机 / DeepSeek / 智谱 / 通义),让 AI 像群友一样插话、接梗、潜水。OneBot v11 + NapCat,零构建网页控制台。Apache-2.0
AgentHER
AgentHER: Hindsight Experience Replay for LLM Agents
Jambo
Jambo - JSON Schema to Pydantic Converter
CodeInsight
基于 Hybrid RAG 与 LangGraph 的本地代码仓库理解工具,支持中英文提问、语义检索、证据引用和多问题拆分
Yantrikdb Hermes Plugin
YantrikDB memory provider for NousResearch/hermes-agent — self-maintaining memory with benchmarked recall, self-tuning ranking, contradiction tracking, proactive hygiene, and explainable recall.
Agent Git
Agent Git: Agent Version Control, Open-Branching, and Reinforcement Learning MDP for Agentic AI. A Standalone Agentic AI Infrastructure Layer for LangGraph Ecosystems
Paper2slides
Transform any arXiv papers into slides using LLMs
Liliumai
Lilium AI: The ultimate personal AI agent framework for autonomous computer control. Featuring browser automation, shell execution, and multi-channel integration (WeChat/Telegram/Discord).
MOTO Autonomous ASI
MOTO is an automated theorem generator for science. It's a creative novelty-seeking researcher with autonomous Lean 4 proof generation. Run for days at a time once pressing start - no interaction needed! Agents working in parallel from either local host LM studio, OpenRouter, OAuth or all 3. No internet required. Star us for more!
Threatswarm
27 scope-enforced AI agents that run the full pentest kill-chain (recon → exploit → post-ex → DFIR → report) as a one-command Claude Code plugin. Backed by 754 MITRE-mapped skills.
VidForge
Open-source AI pipeline for short-form commerce videos: multimodal analysis, RAG scripts, multi-agent orchestration, FFmpeg composition, and cost tracing.
Pie AI Agent
Pie – free, open-source AI browser agent for Chrome. Reads pages & PDFs, fills forms, extracts data, automates tabs, hands off to Claude Code. BYOK or subscribe.
Amfs
Recursive Self-Improvement for AI agents and models, helping your agents self-improve with every run
Darwinia
The Self-Evolving Agent Ecosystem — Trading agents that evolve through Darwinian selection and adversarial self-play
Agentic AI Pipeline
🦾 A production‑ready research outreach AI agent that plans, discovers, reasons, uses tools, auto‑builds cited briefings, and drafts tailored emails with tool‑chaining, memory, tests, and turnkey Docker, AWS, Ansible & Terraform deploys. Bonus: An Agentic RAG System & a Coding Pipeline with multistep planning, self-critique, and autonomous agents.
Sovereign Agent Swarm
Adaptive Human-Agent Teaming (HAT) & Smart Company (SmC) Infrastructure This repository implements a Human-Model-Free adaptive agent architecture. Instead of relying on rigid, computationally expensive human behavioral models, the system leverages real-time trajectory analysis and the Cross-Entropy
Meta Core
Deterministic telemetry & control plane for multi-agent AI systems. Atomic POSIX JSONL event bus, W3C/OpenTelemetry DAG reconstruction, and circular handoff guardrails.
Obsidian Steward
A vault-specific agent equipped with agentic capacity, fast search, flexible commands, vault management, and terminals to "jump" into other CLI agents, such as Claude, Gemini, etc.
OmniAgent
OmniAgent (ICML 2026): the first native omni-modal agent for active video perception — a 7B agent that beats Qwen2.5-VL-72B with 73% fewer frames on LVBench.
Socratic
Socratic is a framework for building reliable vertical AI agents by letting human experts teach agents interactively, turning tacit domain knowledge into a continuously improving knowledge base.
Pi
A pure-Go port of pi (earendil-works/pi) — the pi-ai unified multi-provider LLM API, pi-agent-core agent loop, and pi-coding-agent, native in Go with no Node.js runtime. Anthropic/OpenAI/Google, 999-model catalog.
List refreshed November 15, 2026. Data comes from public GitHub repositories.