GitHub agents

ad-astrAI - multi agent AI tool for detecting elements and molecules from spectral and image data from galaxies far far away.

Multi-agent AI orchestrator with critic loops, HITL and parallel tool calling. Built with LangGraph + DeepSeek + Tavily.

Minimal autonomous coding agent. Simple tool-use loop over LLM API — no frameworks, no magic. Bash runs in Docker sandbox. Supports Anthropic, OpenAI, and any OpenAI-compatible provider.

A modern web application that analyzes pitch decks using multi-agent AI technology. Upload your pitch deck and get comprehensive feedback on structure, content, and potential improvements!

Evidence-first legal QA system — 300+ DIFC documents, 16 retrieval experiments, 87.5% rejected after ablation. LangGraph + Qdrant + FastAPI. 38/80 in Agentic RAG Challenge 2026 (https://agentic-challenge.ai/leaderboard).

AI-powered stock analysis app that delivers Strong Buy / Buy / Hold / Sell / Strong Sell recommendations backed by technical analysis, fundamental metrics, and news sentiment — with full source transparency.

Biological nervous systems don't recompute known workflows from scratch. Mnemon gives LLM agents the same primitive — execution memory that caches plans, not responses. 93% token reduction, 2.66ms vs 20s, zero tokens on repeat runs. LangChain, CrewAI, AutoGen.

A base chat bot image that can be quickly customized for domain knowledge

This repository demonstrates Cache-Augmented Generation (CAG) using the Mistral-7B model.

This project implements an AI agent that verifies if automated Hercules test runs were executed as intended by comparing planning logs, video evidence, and final outputs. It uses open-source LLMs and computer vision tools to flag deviations, providing detailed reports with technical insights.

AI Assistant - Made using ElevenLabs, Livekit & LangChain

A Deep Research Agent that orchestrates in-depth web research on any topic. Built on LangGraph with 5 search providers, three-tier LLM hierarchy ,semantic Redis caching, and PDF/MD/JSON export. Generates publication-quality PDF/MD reports with interactive plan review and RAG chat.

AI-powered logistics management app utilizing LLMs and web scraping for real-time distance calculations and route optimization. Features secure user authentication, enabling personalized access to streamline logistics operations and enhance supply chain efficiency.

A production-grade AI Digital Twin built from scratch. Features an evolutionary architecture: starting from raw RAG & Memory experiments, evolving into an Agentic Workflow with Tool Calling with a Streamlit UI. Powered by LangChain, OpenAI, and ChromaDB.

TypeScript library providing persistent memory for AI agents — conversation history, long-term memory with vector search, and automatic fact extraction.

Surprise Travel Planning Agent: An AI assistant that designs personalized, surprise travel itineraries based on user preferences, budget, and interests. It suggests destinations, activities, accommodations, and travel routes—turning trips into exciting, hassle-free adventures.

SwarAI is a voice-enabled AI assistant that uses multi-agent orchestration, LLM reasoning, and task automation to perform actions like messaging, file search, payment and information retrieval. It is powered by a FastAPI backend with interactive web-based interfaces for seamless user interaction.

Custom OpenClaw skills — multi-agent workflows, CrewAI, AutoGen, HuggingFace, dev workflows

Production-ready AI agents & chatbots using Google Gemini, Groq, LangChain, CrewAI & LangGraph | 10+ projects covering RAG, document analysis, code exploration & more

Self-hosted AI agent observability with tool-call tracing and decision tree visualization

Multi-agent AI system using CrewAI and Ollama to automatically research, write, and review beginner-friendly Python tutorials

Designed to support healthcare professionals in remote or underserved areas. By leveraging multiple intelligent agents and Google’s Gemini language model, this tool provides an AI-driven differential diagnosis based on patient data such as symptoms, vitals, lab results, and medical history.

公开类似langchain，crewai，phidata这类agent framework的搭建思路，还原从0到1，从1到100的agentic framework构建过程

Professional DAST & Pentest Framework by Zyron Security — AI-powered, active fuzzing, IDOR detection, pipeline breaker

Coursera - deeplearning.ai crewai course

28 agentic design pattern skills for Gemini CLI, Antigravity, and Claude Code. Supports English, Korean, Japanese, and Chinese triggers.

🥷The JedAi Council: Build powerful AI agents with CrewAI, Graphiti Graph RAG, LlamaIndex, and modern uv package management 🚀

The Official Github page for James Karanja Maina, author of The Complete AI Blueprint series of books.

This is AI Research Assistant repository using Gen AI and RAG model. This AI Research Assistant app is powered by Google Gemini. It helps in question answering about the provided research paper by uploading them through Streamlit, we save it in the Data folder and clean the document, index it by using Llamma-Index using Gemini Embedding Model.

Production workflow framework for Claude Code — multi-agent orchestration, parallel worktrees, autonomous task execution, and compound engineering skills

Agent-Native Version Control for AI Swarms (CASM Architecture)

🤖 Bash CLI orchestrator for AI coding tools (Gemini, Codex, Claude, OpenCode, Cursor) — non-interactive fan-out & multi-agent pipelines

No longer maintained. Structured multi-agent orchestration experiments for Codex, Claude and OpenCode workflows.

AI roundtable, where a set of AI agents gather to discuss a topic. Implemented as multi-agent system.

Stigmergy Robotics: Crypto Stigmergic Organism #SIFTA Born on Hardware.

Claude Code template with commands, skills, agents, and hooks

A Pi-style multi-agent stack for private-deployment AI services — configurable agents/tools/models via MD/YAML

Governance layer for AI coding agents — every action classified, gated, and cryptographically logged. 751 tests. 3 dependencies. Zero databases. #rai-verify-2026

Fully automated TDD pipeline for Claude Code. Tester, coder, reviewer — separated so the agent can't cheat

Graph-native memory and self-learning for Claude Code. Every session remembered, every pattern learned, every agent connected — powered by AI Native Lang.

TUI multi-agent dashboard for Claude Code — dispatch N parallel agents on GitLab/GitHub issues in isolated git worktrees.

A Meta-Agent-Driven Self-Evolving Multi-Agent System for UAV Detection and Tracking

Assemble your AI team — a multi-agent product management framework for Claude Code

Tools. Agents. Wired. | the canonical tool layer

Claude Code skill that diagnoses and improves harness configurations based on 8 design principles from Anthropic's harness design article

The Nervous System - LLM Behavioral Enforcement Framework. Anthropic built the brain. This is the nervous system.

Production-Ready AI Systems • Enterprise Architecture • Azure & Beyond

多智能体思维协作平台 — 让乔布斯、芒格、费曼同时为你思考。通过认知蒸馏重构卓越灵魂的思维模型与表达DNA。

Bobby Axelrod meets on-chain intelligence. An autonomous trading agent with metacognition — scans OKX whale flows, cross-references Polymarket smart money, debates itself, and only trades when the math says yes. Built for OKX X Layer AI Hackathon 2026.

Open-source quant platform that proves the evidence behind every BUY/SELL decision.

Portable agent skill for analyzing and designing protocol-driven coordinator-worker runtimes, task boards, mailbox messaging, and verifier flows.

The first open-source effort estimation tool built for AI coding agents. PERT + METR + wave planning.

Team templates for human × agent collaboration — dev, sales, marketing, and more. Powered by hxa-connect.

Multi-agent orchestration with MCP — LangChain, DuckDB vector storage, FastAPI. No external vector DB required.

Three complementary agent skills for academic research: WRITES papers, FINDS papers on a topic, READS any paper as a visual experience (mind maps, flowcharts, plain-English). Runtime-neutral, works with 50+ agents. Install: npx skills add aniketkrs/research-paper

MCP server that lets AI models invoke CLI agents (Codex, Claude, OpenCode) as tools — with parallel execution, retries, and structured output parsing.

Research: stigmergy-inspired pressure-field coordination for multi-agent LLM systems, benchmarked against conversation and hierarchical baselines

A production-oriented, TypeScript multi-agent AI platform built on Google Genkit. The system composes Genkit flows, reusable tools, and independent A2A agents to support complex tasks such as research orchestration, code generation, and content editing.

Managing more agents is easier when they are just one phone tap away!

StoryGPT generates fictional narratives using a simple multi-agent pipeline with OpenAI API

Where Minds Meet - Multi-agent AI Collaboration server - MCP + CLI — shared workspace for your AI agents.

Orchestrate agent teams across multiple AI coding CLIs (Claude Code, Codex, more via PRs). Define once, run everywhere.

Multi-agent Claude Code plugin for end-to-end data analysis automation — agentic swarm with profiling, cleaning, visualization, dashboards, and reporting

OmniBridge is an MCP server that gives AI agents an ephemeral, gVisor-isolated sandbox to run, test, and cryptographically prove their code works — before it ever reaches production.

Your AI company, on your own machine. A multi-agent platform where you hire named agents, chat with a Lead to design workflows, and watch them run — desktop app or self-hosted server.

Autonomous AI pentesting framework — LLM-driven agent with plan trees, attack graphs, and 11 security skills

Install your own AI DJ Being. She searches, downloads, listens, mixes, and generates music — autonomously. 30hrs for $0.04.

Autonomous DEX strategy discovery — Karpathy-style autoresearch with LCM memory | Synthesis Hackathon 2026 | Base DEX | Uniswap V3 + Aerodrome | OpenClaw + Bankr

💬 Lightweight AlgoChat client for corvid-agent — GitHub Pages hosted

An autonomous agent for mathematical reasoning and proof

Minimal, hackable AI agent built on the ReAct reasoning loop. No frameworks — just a transparent Thought → Action → Observation cycle you can read and extend in an afternoon.

Universal AI agent — autonomous digital worker for any domain

W.A.D.E. — A local-first autonomous AI runtime that runs on your hardware, owns your data, and acts without being prompted.

A bare-bones Swift agent runtime with built-in tool execution, skill support, conversation-state handling with a unified multi-provider LLM layer (OpenAI-compatible/Responses, Anthropic, Gemini, and MiniMax, plus codex OAuth)

The missing transactional layer for AI agents side effects. At-most-once execution, human-in-the-loop approvals, queryable audit trail.

Code Quality Guardian

Desktop application for managing and orchestrating JIDO autonomous agents with a Conductor-style UI

A powerful framework for building and deploying Agentic Oracles on the IXO network. Built on LangGraph, Matrix, and NestJS, this framework enables developers to create intelligent oracles with secure communication, complex conversation flows, and automatic marketplace integration.

Slipstream - Built by Vektor Memory - a multi-layered persistent agentic vector memory architecture. 4-layer associative graph memory (MAGMA) with autonomous REM cycle, CLI, DXT-MCP Cloak Tools.

AMOF 3.3 Write-Scope Authority is live. Agentic Operations Fabric — governed runtime authority for intake, context, runners, bounded repository mutations, enforcement, receipts, and evidence.

Everything AI agents need to master the full Kaltura platform — The Agentic Digital Experience Platform. Live-tested API guides with curl examples and end-to-end test scripts. Agent Skill, llms.txt, and Context7 ready.

Apiiro Guardian Agent

Build an intelligent copilot application for insurance claims support, leveraging memory and tool calling capabilities to provide personalized and context-aware assistance. This project enhances support efficiency by enabling the system to remember user interactions, retrieve relevant information, and utilize external tools for informed decision

Memory AutoResearch Agent that tests different hypothesis to create better memory inspired by @karpathy

All my AI Agent projects in one place. Smart, interactive, and tool-powered! 🚀

A lightweight workflow system for Claude Code with hook-based enforcement, session state preservation, and deterministic task coordination.

GRC Mapping Analyst is a NIST IR 8477-based toolkit for AI-assisted cybersecurity crosswalks, producing deterministic set-theory mappings (equal, subset_of, superset_of, intersects_with, not_related) and 12-column STRM CSV outputs across frameworks, regulations, and control catalogs.

Open, audit-grade agentic data quality framework with portable industry packs

Agentic BI SaaS platform for natural language analytics using LangGraph, LangChain, LLMs (OpenAI, LLaMA, Mixtral), RAG with FAISS/Chroma, SQL & BI agents, MCP tools, and long-term AI memory (mem0). Built with FastAPI, Streamlit, Plotly, SQLAlchemy, PostgreSQL,Docker,Kubernetes,governance, lineage, RBAC,usage-based billing,SaaS & on-prem deployment.

Engineering playbook: structured Markdown commands for AI-assisted development across the SDLC — planning to production operations. Tool-adaptable.

Persistent, local memory for Claude Code — SQLite + semantic search via MCP.

🟩 Learn and experiment with n8n (nodemation) + Agentic AI to build intelligent automation workflows.

Sovereign L1 Orchestrator powered by Simplicity. A TEE-hardened autonomous agent for Bitcoin

Advanced token reduction and prompt optimization framework for LLMs, featuring linguistic, algorithmic, and architectural patterns.

Agentic simulator for wildfire evacuations.

ooderNexus 是一个基于 Ooder Agent 架构的 P2P AI 能力分发枢纽产品。该项目采用 MIT 开源协议，旨在打造一个去中心化的 AI 能力共享和管理平台，支持个人、小微企业及部门内部部署，实现 AI 能力的安全、高效分发。