AI Agents: open-source projects on GitHub
1319 repositories on AI Agents, picked by the REDDYX AI agent. Each has a summary, key features and a breakdown. Sorted by stars.
// RELATED TOPICS
45 AGENT SKILLS FOR PRODUCT ENGINEERING
A set of 45 AI agent skills extracted from shipping three live products. Covers the cycle from deciding what to build to releasing with a second model review.
★ 611 · PythonHUNTER COMMUNITY AI
Self-hosted financial AI team in 15 minutes. Open source, opencode, MCP, multi-agent.
★ 610 · PythonGATEWAY: UNIFIED SDK FOR LLM
Gateway is a local production‑grade Super SDK that unifies over 200 LLMs into a single interface. Use any model without complex setup.
★ 609 · TypeScriptHorizon — GPU terminal
GPU-accelerated terminal for sessions on an infinite canvas with intuitive control.
★ 607 · RustRSC-HARNESS MEMORY AND REFLEXES
rsc-harness — meta-harness giving the AI agent memory and safe access to the DB without risk of breaking the repo. Supports Claude Code, Codex and 15+ tools, written in J
★ 607 · JavaScriptSOLO: Human-AI Collaboration
The solo project lets people and AI agents work together via channels, tasks, and persistent workspaces in Go.
★ 605 · GoBKIT VIBECODE KIT: AI‑NATIVE DEVELOPMENT
bkit combines PDCA and Claude Code for AI‑native development in JavaScript. Kit provides templates, prompts and CLI utilities that accelerate the PDCA cycle and support m
★ 604 · JavaScriptCITADEL: LAYER FOR CLAUDE CODE AND CODEX
Control layer for Claude Code and OpenAI Codex: project memory, intent routing, security, telemetry, and agent fleet. 602 stars!
★ 602 · JavaScriptJEV: 884 VERIFIED EXAMPLES
Collection of 884 verified Jev examples — TypeSafe AI's decision model. Indexed by decisions made, not sources. CI re-validates all links weekly. Bilingual EN/中文, JSON sc
★ 602 · PythonStanford AI Agents Course (Chinese)
Chinese translation of Stanford CS329Z Engineering AI Agents course with 19 lectures and 4 hands-on projects covering LLMs, RAG, tool calling, memory, multi-agent systems
★ 599 · HTMLDEFT: SHARED AI WORKSPACE
Deft is a self‑hostable AGPL‑licensed platform where humans and AI agents chat together, manage tasks, store knowledge, approve actions, and share MCP context.
★ 598 · TypeScriptPLUGIN SUIT CLOUD CODE
A set of plugins and MCP servers for Claude Code: Agile pipeline, AI review, docs generation and code audit.
★ 596 · JavaScriptPERSISTENT MEMORY FOR AGENTS
funes stores AI agent session history in a searchable LanceDB database. Lets you find context from past dialogues and code in milliseconds. Written in Rust, integrates wi
★ 594 · RustRSS READER FOR AI AGENTS
Fast native RSS reader on Rust with CLI for AI agents — read, search, and sort feeds right from the terminal.
★ 593 · RustOFFICIAL GUIDE TO AI AGENTS
A complete learning roadmap for building agentic systems based on OpenAI, Claude, MCP, Harness, and Evals. It includes practical Python examples and recommendations for d
★ 589 · PythonAGENT OPFOR ATTACK EMULATION
Open‑source framework for simulating attacks on AI agents and MCP servers. Written in TypeScript and already has 377⭐.
★ 586 · TypeScriptwatch-skill - Agent Video Search
Convert agent videos, streams, and screen recordings into a timestamped search index, then verify and fix work via THE LOOP.
★ 582 · PythonBEX - DEPLOY WITH AI AGENT
bex — an open-source Render alternative in Go: git‑push → build → deploy on your own infrastructure, AI agents as users.
★ 578 · GoMAESTRO-FLOW AI AGENT MANAGEMENT
Maestro-flow orchestrates AI workflows through an adaptive engine and knowledge graph, offering a visual dashboard for Claude Code, Gemini, Codex.
★ 576 · TypeScriptSWARMCLAW: SELF-HEADED AI AGENTS
Open-source self-hosted runtime for autonomous agent swarms: memory, MCP tools, schedules, and 23+ LLM providers.
★ 572 · TypeScriptENFERNO FLASK FOR AI-AGENTS
Enferno — a lightweight Flask framework for building AI agents, using Vue 3 + Vuetify with no build step. It includes production-ready authentication (2FA, passkeys, OAut
★ 571 · HTMLCODE GRAPH FOR AGENTS.
graphicle — a single Rust program that builds a code graph and provides memory for the AI agents. Supports Claude Code, Codex, Cursor and any MCP clients, indexing the re
★ 571 · RustPINLOOP-CLI: JOB SEARCH FOR AGENTS
pinloop-cli searches and scores over 10 000 jobs from 4 000+ companies, updating data hourly. The tool runs in the terminal and is suited for AI agents such as Claude Cod
★ 570 · TypeScriptCOMPARTMENT: ENCRYPTED AGENT MEMORY
Encrypted, fully offline agentic memory with one‑click install and GUI memory map. Works on any OS and with any AI agent.
★ 569 · PythonORACLE3: PAPER‑TRADING ENGINE FOR PREDICTION MARKETS
Oracle3 is a paper‑trading engine and MCP server for prediction markets Kalshi and Polymarket. It checks for absence of arbitrage between related contracts, accounting fo
★ 569 · PythonAGENTOS - AI Agent Framework
TypeScript framework for building AI agents with cognitive memory, dynamic tool generation, and multi-agent orchestration.
★ 568 · TypeScriptNUPHOS — AGENT WORKSPACE FOR TEAMS
Team platform to run and manage coding agents. Syncs people, machines, and tools in one workspace. Supports Claude Code, Codex, and alternatives.
★ 568 · TypeScriptLAMPOS — OS FOR PHYSICAL AI
lampOS — an open-source operating system built to run physical AI on embedded devices. Written in Python, it supports working with LLMs, agent frameworks, and robotics.
★ 567 · PythonSMART RALPH: SPEC-DRIVEN DEVELOPMENT
A plugin for Claude Code that combines Ralph Wiggum's cycle with the specification workflow and smart code compaction.
★ 566 · ShellAGENT-SKILLS: AI Assistant for Azure
Agent-Skills gives the AI assistant up-to-date skills from Microsoft Learn for fast, accurate code in Azure.
★ 565INFRASTRUCTURE FOR AI AGENTS
Open web toolkit for AI: scraping, crawling, clean Markdown, and browser sessions ready for your agents.
★ 565 · TypeScriptLOCKKNIFE ULTIMATE TOOL
LockKnife combines TUI and headless CLI for deep Android research. Powered by Python orchestration and Rust core, it works with AI agents, credential cracking, APK analys
★ 563 · PythonPIXEL-ART OFFICE FOR AI AGENTS
Pixtuoid — a terminal application in Rust that turns AI-agent work into a pixel‑art office. Integrates with Claude Code and Codex, allowing agents to write and test code
★ 563 · RustAINL TURNS AI INTO A WORKER
AINL turns a regular chatbot into a structured AI worker, enabling multi-step workflows with memory, tools, and validation.
★ 561 · PythonREPOSITORY-HARNESS: REPO FOR AGENTS
Turns any repository into a workspace for Claude Code, Codex, Cursor and other AI agents. 559 ⭐
★ 559 · RustGISSELLE: AI APP BUILDER OPEN
Create AI applications without code: Giselle is an open TypeScript framework for rapid prototyping.
★ 559 · TypeScriptCAURA Shared AI Memory
Caura provides managed shared memory for AI agent fleets with multi‑tenant support, trust policies, and an audit trail.
★ 559 · PythonLOCAL-FIRST INBOX FOR AI AGENTS
pizza-bot is a local-first inbox for long-running AI agents on DeepAgents and LangGraph. Supports human-in-the-loop and runs via Electron. 319 stars on GitHub.
★ 559 · TypeScriptDEER-WORKFLOW — GRAPH WORKFLOW IN TYPESCRIPT
Open runtime for graph engineering: orchestration in TypeScript, with semantic work delegated to replaceable Agent runtimes.
★ 558 · TypeScriptCONTROL AI AGENTS FROM YOUR PHONE
omg.dev — remote control for Claude, Codex, Cursor, OpenCode and other AI agents. iOS app lets you launch tasks, read logs and intervene in agent work remotely. 551 stars
★ 558 · TypeScriptMCO — Multi-Agent Orchestration
A neutral orchestration layer for Claude Code, Codex CLI, Gemini CLI, and other AI agents. 375 stars on GitHub.
★ 557 · PythonYOUR AI AGENTS IN MESSENGERS
abacusai-bot — 100% free open-source framework for running personal AI agents in Telegram, WhatsApp, and CLI. Written in TypeScript, supports tools, code, and integration
★ 557 · TypeScriptAI GAME BREAKDOWN IN C
decomp.wtf - AI agent decompiles your games into C code, byte‑for‑byte identical to the original. Runs locally, without sending data to the network. Uses Claude Code for
★ 556 · PythonYOUR TEAM OF AI BOTS IN ONE BINARY
Self-hosted solution for AI agent team: spaces, all accounts in one place, task handoffs, approvals and routines. Written in Rust + React. Alternative to Grok Bot, Claude
★ 556 · RustVIBECODING WITH CLAUDE CODE
A toolkit for AI development based on PDCA methodology. 555⭐ on GitHub!
★ 555 · JavaScriptBELUGA: AI IDE FOR MACOS
Beluga — native AI IDE for macOS with P2P connections. Integrates Claude Code and git worktree for parallel agentic development. TypeScript, 500+ stars on GitHub.
★ 554 · TypeScriptSECOND BRAIN - YOUR DIGITAL OS
An agent framework that automates tasks using LLM and file intelligence — 552 stars, Python. Excellent DevTools for cross‑platform automation!
★ 552 · PythonCAD2DATA: AUTOMATION OF CONVERSION OF FILES.
The project provides a workflow for AI agents that automates CAD file conversion (.rvt, .ifc, .dwg) using command‑line tools on Windows locally. It enables integrating co
★ 552 · Jupyter NotebookEMILIA PROTOCOL ACTION GUARDIAN
EMILIA Gate verifies authority before changing money, code, permissions, and infrastructure, providing independently verifiable proof.
★ 550 · TypeScriptMOBIUS: RUST FRAMEWORK FOR AGENTS
Möbius is a compact Rust framework for building code-writing agents. Frontend-agnostic and works with any LLM. Already gained 300 GitHub stars.
★ 547 · RustCodex Mobile on any OS
Run Codex Mobile on Linux, Windows, and Termux on Android without restrictions.
★ 546 · TypeScriptSECURE SANDBOX FOR AI AGENTS
Box isolates AI agents, using OS isolation and default‑deny Dogwood policies. It injects credentials from a secure store without exposing them to the agent, and restricts
★ 546 · RustMCP-SECURITY: AI-AGENTS FOR SOC
MCP-Security is an open Python framework for integrating AI agents into SOC/SIEM/SOAR. Automates threat analysis and incident response using LLM.
★ 544 · PythonVICOA: AGENT IDE FOR CODE
Vicoa — agent IDE that lets you control a team of coding agents from any device: desktop PC, smartphone, or VPS. The project is open source, self-hostable, and written in
★ 542 · Python230+ SEMANTIC ANCHORS
Catalog of 230+ verified terms like arc42, TDD London School, tracer bullet. Activate deep LLM knowledge. Insert into prompts and agent instructions instead of manual met
★ 541 · HTMLSELF-HOSTED SEARCH FOR AI AGENTS
ask-search — CLI and MCP server over SearxNG for searching Google, Bing, DuckDuckGo, Brave without API keys. Works with Claude Code, OpenClaw and any shell. Optional Tavi
★ 540 · PythonDROIDLINE: ANDROID AUTOMATION WITHOUT ADB
Automate Android from code: tests on real phones, routine scripts, device farm, or AI agent control. Works via Python, Node.js, CLI, HTTP, and Appium. No ADB, USB cables,
★ 538 · GoMEMT-1.0: LONG-TERM MEMORY FOR LLM
Official MemT-1.0 repository. Implements long-term memory for LLM agents. Supports long context and DeepSeek integration.
★ 537 · PythonMASTER OF CODE AGENTS
Machinist runs AI coding agents as supervised local workflows. It enables supervised execution of agents like Claude Code, Codex and others, controlling their actions via
★ 535 · GoDEEP SEEKER ON ANDROID
Launch the DeepSeek Harness AI agent on Android without root or Termux — all in one APK with an extension center. Supports local agent execution and API interaction.
★ 535 · KotlinNew projects on this topic in Telegram
Telegram →