GitHub agents

Shogun AFM is Agent Fleet Management for self-improving AI agents — combining agent orchestration, persistent memory, fleet monitoring, governance, security posture, and Gensui command control.

Intent-addressable code for AI agents. Causari records every action an agent takes on your codebase - the prompt, model, reads, writes and reasoning - and lets you trace, diff, revert and bisect them like git commits. With bidirectional causal provenance and a built-in MCP server.

2D pixel-art survival trading game — four LLM-powered AI agents compete to escape the island

AI Testing Agent Framework · 16 experts + 32 skills + 49 utils · Multi-LLM (Claude/OpenAI/Qwen/etc) · MCP-native · Open-source · Learn-while-using

Agent-native inboxes for AI agents. REST API + built-in MCP server, custom domains, OTPs, and verification emails — powered by Cloudflare.

7 Claude Code skills for software architecture review (Python, web, cloud, microservices). Includes A/B benchmarks against unskilled baseline, assertion-graded eval suite, and interactive dashboards.

Local-first codebase indexing for Claude Code, Codex CLI, OpenCode & AI coding agents — hybrid FTS5 + Tree-sitter + graph search, fully offline.

Intelligent, self-healing polyglot pipelines. Declarative. Observable. Autonomous.

OpenSkald is a self-evolving knowledge-native publishing agent.

Multi-format RAG knowledge base with BGE-M3, LanceDB, hybrid search, FastAPI and Streamlit.

Stdlib-only Python CLI that orchestrates universal AI agent providers (Claude Code, Codex, Antigravity, OpenRouter, DeepSeek, Groq, local models, Aider) into a multi-agent PR review jury — debate, verify, synthesize one verdict. Local-first, secure by default.

Run agentic TUIs inside Neovim with persistent float/panel views and editor-context helpers.

Governed multi-agent runtime for Petri-net workflows, scoped agent permissions, durable state, and replayable execution.

Skills + MCP server for Claude Code, Codex, Hermes, or OpenClaw - headless MIKE+ automated modelling and analysis (inspect, edit, run, read, plot), no GUI.

AI-native operating system for product management — 43+ workflows, 5 skill domains, built on Claude Code

Privacy-first, VS Code-style desktop workbench for understanding, auditing, and safely steering AI changes to your codebase — local-first Agentic RAG where every AI edit is explainable, auditable, and reversible.

Forensic git-history learning diagnostic for AI-assisted developers. ICM specialist that turns commit history into evidence-backed study plans and HTML reports.

🌹 100% offline Chatbot. Custom characters, 13 languages, passion system. Powered by Ollama.

Minimalist coding agent harness with a Textual TUI and headless CLI. <1k-token system prompt, 18+ LLM providers, AGENTS.md + skills context, session tree, prompt/auto permissions.

Open-source AI Gateway, all enterprise features included.

High-fidelity transcript memory for Claude Code — index every session, recall byte-identical, search, prune, compose. Local-first, stdlib-only, MCP.

Model Context Protocol (MCP) server that gives Claude full access to the Zernio social media management API — 273 tools covering posts, analytics, inbox, ads, contacts, WhatsApp Business, and more across 14+ platforms.

MCP server giving Claude AI access to 20+ NYC public-record datasets across 9 city agencies via 18 tools — NYC property due diligence in plain English: violations, liens, sales, ownership, permits, rent stabilization, and more.

Production grade RAG SaaS with multi-tenant auth, semantic caching, LLMOps observability (Prometheus/Grafana), per-tenant token cost tracking, and rate limiting - inspired by Kapa.ai (YC S23)

DREVIHO — Dreaming Viking Hooks. Memory + Reasoning hooks for AI coding CLIs. Hybrid search, 5-scope system, MMR diversity, temporal decay, content merge, auto-linking, reasoning daemon. Claude Code, Codex, Gemini CLI, OpenClaw.

Production-grade Open WebUI deployment with RAG, private web search, OCR, local TTS, and MCP tool servers. Supports both standalone Docker Compose and Docker Swarm.

企业级 RAG 知识库 · 文档解析 / 智能搜索 / 流式问答 · FastAPI + Next.js 14 + Qdrant + OpenSearch + LiteLLM

Parse any file to Markdown and JSON, anyparse, document-parsing, document-understanding, ocr, vlm, multimodal, image, pdf, markdown, docx, pptx, xlsx, tsv, csv, txt, md, rst, htm, html, xhtml, shtml, png, jpeg, jpg, webp, epub, ipynb, eml, etc.

Local MCP server for Hermes Agent docs — offline search and chunk retrieval via SQLite FTS5

Bring Google NotebookLM into any MCP client (Claude, Cursor, Codex, Antigravity…) — grounded Q&A with citations, source management, and Studio content generation. Install via uvx or a one-click Claude Desktop extension.

AI Engineering Bootcamp: One-year, full-time, instructor-led program. Published curriculum, assessments, capstone requirement, and formal credential. Built for graduates to cite on resumes and LinkedIn.

Blazor-based UI for Netclaw (unofficial)

Aether — Private local AI OS for desktop. Smart assistant (Alfred) powered by Ollama + cloud fallback, with PDF RAG, web search, and memory. Built with Java + Spring Boot.

Turn any Flutter app into an AI-operable interface — no screenshots, no OCR, no overlay chatbot.

电商设计台 - AI驱动的电商主图与详情页生成工具

OpenCode fork with flexible System prompt control, bundled Codex prompts, and a Codex-compatible ChatGPT Responses wire.

205 pay-per-call MCP tools via x402 — no API keys. Stocks, crypto/DeFi, macro data, on-chain analytics, weather, government data, and 200+ more. USDC on Base.

An AAS MCP adapter that exposes configured Asset Administration Shell APIs as Model Context Protocol tools, enabling LLM agents to interact with any AAS-compliant backend.

An advanced, sandboxed AI coding assistant CLI & TUI in Rust, featuring parallel tool execution, multi-provider LLM support

AI-powered Bilibili video Q&A agent — ask anything about any video in the comment section. 在 B 站评论区 @机器人，即可对任意视频提问。

Agent WebBridge with multiple Chrome profiles & logins at once — drive several Google accounts simultaneously, not one at a time. Fixes the single-slot / one-login limit.

Browser automation, one binary. The simpler alternative to Playwright — no Node, no Python, no runtime. Library, CLI, MCP server, and chat UI for any AI agent.

Crevideo Reach — the TikTok Shop affiliate outreach & marketing platform, packaged as a Claude (MCP + Skills) plugin; the affordable, agent-native alternative to Euka and Reacher.

Give your AI agent a mouth, ears, and a phone number. Framework-agnostic VoIP/IVR service with Twilio — plug in any agent backend (Hermes, OpenAI, Ollama, etc).

Self-hosted, pluggable personal agent that gleans signal from RSS, scraping, search, and APIs — processes it with any LLM, then delivers on a schedule to whatever sink you wire up

Pi for Ruby

Claude Code skill for writing rigorous, evidence-grounded, hallucination-free peer reviews of ML/AI conference papers.

A cognitive substrate for AI agents — hyperdimensional memory with first-class goals, beliefs, sensory, self-model, and a non-destructive unlearn primitive. Rust top to bottom, embedded-first, no LLM in the read path. One binary, no query language.

Talk once, ship a team. A multi-agent runtime for Claude Code and Codex CLI where the lead does the orchestration — across providers, in conversation.

IM-native multi-agent collaboration platform — @mention Claude Code, Codex, OpenCode like teammates

AI-native micro issue manager for multi-agent LLM development. Hierarchical task decomposition with context chains, safety-critical standards, and zero-infrastructure deployment.

The missing observability layer for AI agents. Real-time session visualization. Every prompt. Every tool call. Every model switch. Live. Drop-in OpenAI/Anthropic gateway. Self-hosted. Bun + SQLite.

FlexSQL: Flexible Exploration and Execution Make Better Text-to-SQL Agents

arxiclaw is the local executable client that lets any LLM-powered agent (Claude Code, OpenClaw, Nanobot, or your own runtime) talk to the arxiclaw platform on behalf of a researcher.

Reliability layer for multi-agent setups — keep agents coordinating without an orchestrator. Cross-dialog contracts + drift detection + black-box memory (no LLM extraction).

Security Agent Benchmarking and Evaluation Research

Privacy-first security and mission-assurance lab for AI agents: TraceProof, authorization twins, MCP/OTel evidence, CI gates, and OSCAL exports.

A lightweight, provider-agnostic agentic harness for VS Code. Now with free GLM 5.2.

Layered, agent-agnostic context architecture for AI coding agents: a tiny always-on baseline, on-demand skills & memory, and deterministic guardrails.

Локальная модульная система ИИ-компаньона с долгосрочной памятью, личностью и поддержкой голоса. Ядро AstraCore..

Chrome extension for translating supported AI courses into 32 languages with local study tools and exam-safe translation

Structured conversational memory framework for LLM chatbots with semantic recall, memory grafting, and graph-based topic memory.

Domain-specific AI skills for working with Valkey - reusable knowledge packs that help agents and developers write correct, idiomatic, high-quality code against the Valkey ecosystem.

TAI Chief - The only AI Chief of Staff platform that every CEO will ever need.

The calm, terminal-first workspace for Claude Code — run claude, with files, Git, diffs & MCP on demand, and munu, a notch mascot that watches Claude for you. Local-only, no telemetry.

ClawTouch open hardware layer — CircuitPython firmware for a Raspberry Pi Pico 2 + the frozen USB-CDC wire protocol it speaks. MIT.

Claude Code plugin for visual docs — annotated screenshots, feature demos, flow walkthrough recordings (gif/mp4), terminal captures and before/after composites, one command each. Self-contained Chromium, self-validated output, no MCP.

Free, self-hosted AI coding assistant for VS Code — local Ollama models, Anthropic Claude, or any OpenAI-compatible server. Full agent loop with tool use, inline completions, and MCP support.

Beautiful, fast, fully native Android chat client for OpenAI, Anthropic, Groq, OpenRouter & any OpenAI-compatible endpoint. Streaming, encrypted on-device API keys, 100% Jetpack Compose + Material 3.

MCP server for XAUUSD: live microstructure, OOS-validated strategy intelligence, macro context, and AI analyst aggregation. Cross-platform (Claude, ChatGPT, Cursor, Windsurf).

Define your agent in a single config file, run it in a container and deploy it anywhere.

Project once, retrieve distillations, create tools: a pre-loading retrieval architecture and tool builder for distributable, source-grounded assistants for complex, repeatable work in your known task domains.

AccInt - local-first MCP Work Model for coding agents that learns from real outcomes.

Semantic code-search (semcode) MCP. Indexes code symbols and commit history. Combines dense embeddings with sparse BM25 vectors for hybrid search that balances semantic understanding with keyword precision.

🎯 中文大厂大模型面试教练 | 不是泛泛刷题，按真实面试逻辑做「项目深挖+追问+诊断+个性化训练闭环」，适配算法岗/应用岗/RAG/推理优化/训练微调等全岗位

Self-hosted web scraping and Markdown extraction for AI agents

⚓ Provider-agnostic terminal privacy first coding agent — bring your own model. Switch between OpenRouter Anthropic, OpenAI & local Ollama with one command. Plan mode, checkpoints, MCP, hooks, sub-agents and a modal prompt, built off Pi.

chad: a local Claude-Code-style coding agent for your laptop

The Karpathy LLM Wiki, production-ready. Zero-RAG personal knowledge base with MCP server, multi-agent support, knowledge graph visualization, and hallucination enforcement. No embeddings, no vector DBs — just markdown + git.

The 21 agentic design patterns, distilled to concepts (no code) — a design-phase reference for building AI agents. Works as a Claude Code skill.

muti-agent-仿seedance图片视频生成平台

Multi-agent Claude Code skills that turn a topic into a DaVinci/CapCut-ready video timeline — research to publish.

Create longform HD videos with music and voiceover — multi-agent pipeline for viral content production

A practitioner maturity model for securing conversational and agentic AI.

Live mission-control for your Claude Code agents and sub-agents. Watch them work, see who needs you, audit your CLAUDE.md, and run every session (and the bill) from one screen.

Self-hosted AI roleplay client with memory that survives long campaigns — tiered canon/heuristic/reflex writes, retrieval-time privacy ACL for group chats, and native MCP tool support. Imports v2/v3 character cards. Runs on Ollama, OpenAI, Anthropic, OpenRouter, or any OpenAI-compatible endpoint.

Persistent memory system for Claude Code — make your AI remember everything across sessions. Ready-to-use templates, zero dependencies.

An intelligent Hermes plugin for deterministic meal planning and fridge inventory management. It suggests what to cook, builds optimized shopping lists, tracks recipes and cooking history, and manages ingredients through a conversational AI interface—using only Python’s standard library.

AI-powered career consultant that helps manage positions and CVs within Telegram

面向复杂软件工作的闭环协调层 ，把专家路由 · 计划 · 执行 · 迭代 · Beta · Release · Feedback 收拢成一个可持续迭代的闭环工作流

MailMan — an MCP server to send and read Gmail (SMTP/IMAP App Password or OAuth2) in plain English from any AI CLI. Attachments, draft/confirm review, scheduled sends, multi-account, inbox search. macOS/Linux/Windows.

Your Linux desktop, now with an AI assistant built right into Rofi. TypeScript and Bun under the hood, no GTK, no Electron, no bloat.

Fast ServiceNow CLI with stable JSON output. Single Rust binary — agent-friendly, scriptable, cross-platform.

An AI partner whose memory lives on your machine, not in a vendor account. Swap the model, keep the mind.

🌀 The first adversarial-collaboration multi-agent Skill for Claude Code. 7 agents + 11 phases + Hook enforcement + 9 signal groups + dual-layer adversarial review. Inspired by NTCoding, agent-review-panel, AgentTeams. v1.6 — main+sub dual-layer adversarial review.

Automate QQ and WeChat farm games with multi-account support, real-time logs, data analysis, and a web control panel using Node.js.

CTF AutoResearch project built around Claude Code, bounded research rounds, and execution/data separation.

ai-berkshireの日本株版（fork ではなく独立実装）— EDINET DB MCP（有報一次データ）接地の Claude Code スキルパック / Japanese-equity edition of ai-berkshire, grounded in primary-source EDINET filings

Let the Orcha play. — AI-driven Feishu coding agent: one @-mention → autonomous plan/code/test/review/push to a new git branch.

Email MCP server for AI agents - full mailbox access for Claude, Codex, Cursor or any MCP client: search, read, send, attachments, folders across Gmail, Outlook, iCloud, QQ, 163 and any IMAP/SMTP host. Zero dependencies. EN/Chinese.