GitHub agents

OnDevice Local AI Studio — run LLMs on iPhone offline. SwiftUI workbench with MLX, llama.cpp, whisper.cpp, RAG, and authenticated OpenAI/Anthropic/Ollama-compatible local APIs.

Local coding agent with native Zhipu GLM / StepFun Step Coding-Plan support. Four hard gates — the agent can't claim done without running the tests. 本地 coding agent，原生支持智谱/阶跃 Coding Plan，四道门硬约束。

Portable one-click Windows launcher for DeepSeek Harness (DSH). No Node.js, npm, pnpm or setup required. DeepSeek Harness 免安装 Windows 一键启动器。

Zero-config internet reading & search for AI agents — 8 platforms, no login, no API key, no ban risk.

Lightweight requirement alignment for DeepSeek Harness — align important decisions before execution without a full spec workflow.

Unofficial Windows desktop wrapper for DeepSeek Harness (dsh)

Ethics-aware OSINT investigation planning, collection, and risk evaluation, delivered as an MCP server. Classification, risk tiers, hard stops, and defensible documentation — driven by local YAML doctrine.

Native Windows GUI for Oh My Pi: sessions, models, providers, MCP, plugins, context, and browser tools.

Unofficial cross-platform desktop app for DeepSeek Harness. Native Wails shell for the DSH Web UI on macOS, Windows, and Linux.

Self-hosted email for AI agents — the open-source alternative to AgentMail. One compose file → unlimited inboxes, OTP extraction, MCP server.

神火AI系统skill

An Android-native local AI runtime with MCP tools, phone capabilities, and OpenAI-compatible APIs.

Grounded Memory Runtime — a runtime tool that links the Agent’s memory and facts

Let vision-capable AI act as demanding customers and catch product problems before launch.

Open-source Cloudflare-native starter for building subscription and credit-based AI Agent SaaS products.

Chrome extension running a 45M-parameter LLM (Needle 2) entirely in WASM. Natural-language control of tabs, forms, pages, and custom scripts — fully offline.

Private AI WYSIWYG Markdown editor with agentic chat & voice dictation. Open source, no cloud, no tracking.

AI Agent + 抓包融合安全测试工具 | 自然语言驱动自主渗透测试 | ReAct 编排引擎 | Tauri + FastAPI

AMH Cloudflare Ops MCP by WT

Generate publication-quality DESIGN.md files from any website URL. Extract computed styles, structured tokens, and layout patterns into a unified reference for AI UI coding agents.

AI SSH client for DevOps: terminal, SFTP, Monaco editor, encrypted vault, and OpenAI-compatible ops agent with plan-and-confirm execution.

This Python project demonstrates semantic search using MongoDB and two different LLM frameworks: LangChain and LlamaIndex. The goal is to load documents from MongoDB, generate embeddings for the text data, and perform semantic searches using both LangChain and LlamaIndex frameworks.

Reference architecture and working case study for Codex × Antigravity agent-runtime collaboration.

Lightweight standalone LLM inference engine. Single binary, no runtime deps. Serves GGUF models via Ollama/OpenAI/MCP APIs. Multi-agent orchestration, TUI, 21 built-in tools.

Two Claude Code instances on different machines can't talk to each other. SwarmCode gives them a Redis-backed channel — messages, shared artifacts, workspace status, real-time push into VS Code, and a web dashboard.

The ledger layer the stablecoin payment stack was missing. Double-entry, event-sourced, agentic-first. Fiat + on-chain. Built for cross-border.

Open specification and reference verifiers for Seal records: ATTESTATION-v1 for AI model-call decisions, ACTION-v1 for agent actions. Ed25519, canonical JSON, checked offline against a published key.

Five-level game where players try to talk a chatbot into leaking a password, each level guarded harder than the last. Flask API behind it.

A highly customizable AI companion for Telegram. Create digital clones with unique personalities, voice, and vision directly from Google Colab.

Photon-hosted iMessage transport and settings UI for DeepSeek Harness

把微信 / QQ / 钉钉 / 飞书 / Telegram 接入 DeepSeek Harness：统一配置、扫码授权，直接在各 IM 与 Agent 对话；支持图片与文件收发，Agent 可读取 PDF、DOCX、XLSX 和文本内容。

Themis · Фемида — multi-agent AI for Russian litigation on Claude Code: 13 lawyer-agents, practice hunters, a 5-jurist council, local-first OCR ($0), self-learning, and an Apple-grade cockpit where the lawyers argue a case live.

SNAFU. Agentic LLM flow to help you with "naming things" in source code.

Reverse-engineering & re-implementation of OpenAI Computer Use macOS plugin for DeepSeek Harness: 10 MCP tools, JSON-RPC wire protocol, native Swift AX/screenshot/input daemon.

Package and share AI agent workflows, skills, agents, and policies across Claude Code, Codex, and other coding agents.

A demonstration of the OpenAI Chat Completion API, featuring a Python Flask server and a Python client for interactive AI-powered conversations. This example showcases the GPT-4o model's ability to understand and generate human-like text in a conversational context.

macOS menubar utility to monitor your AI coding quota — track OpenAI Codex and Claude Code usage at a glance

《企业级 Agent 从入门到专家（2026版）》——从单 Agent 核心机制到多 Agent 与企业落地的工程化实战教程，配套可运行贯穿项目与 CI 验证门禁。

Interactive web app where you describe pathfinding strategies in plain English and an AI forges them into Python code. Runs in a secure Docker sandbox against mazes, race tracks, and weighted maps with real-time Canvas visualization. One algorithm, any terrain.

Prompt Engineering skill for Claude Code — rewrites and optimizes prompts using proven techniques (Anthropic, OpenAI, Gemini, The Prompt Report).

Wenshu（文枢）— 面向人文社科研究的 AI 知识处理工作流：本地知识库 / RAG / 中文引文 / 理论谱系。AI knowledge workflow for humanities & social sciences: local knowledge base, RAG, citation (GB/T 7714), knowledge graph.

Knowledge engine for AI agents: your documents become a git-versioned markdown knowledge graph the agent navigates over MCP. Curated summaries, typed edges, cited answers, SQL over your spreadsheets, no chunking. Same 12B local model: 0/11 on multi-hop questions as top-k RAG, 11/11 as a navigator. Python, Apache-2.0, self-hostable.

Local-first cross-session memory for DeepSeek Harness (DSH): automatic capture, Event/Element cards, evidence-gated recall, source tracing. 本地跨会话长期记忆。

800+ real-world ML & LLM system design case studies from 150+ companies Google, Meta, Netflix, Uber, Airbnb & more. Production AI, not theory.

Event-driven AI document processing engine with offline RAG, summarization, speech transcription, SSE streaming, and self-hosted LLM inference.

Portable AI coding-agent skills for source-attributed, human-approved context packages.

Harness layer for Claude Code & Codex: plan-first development, independent review, durable long-running orchestration, and knowledge capture — ten composable plugins.

Full unofficial agentic Infomaniak MCP server — guided automation of web hosting, mail, kDrive, domains and DNS.

Turn CRM and transcript data into structured, versioned deal intelligence. 6 agents, 3 frameworks, harness-agnostic. Works with any CRM, any transcript provider, any LLM.

RAG evaluation and observability lab.

Four formats. One engine. PDF, DOCX, XLSX, HTML → Markdown and typed JSON, 15–40× faster than equivalent-quality OSS. Rust core with strictly-typed Python bindings.

Self-Reinforced Testing Framework (SRTF) for AI Agents.

Zero-cost meta-search for AI agents: 247 engines, local Tantivy index, MCP native

A sessionless personal AI agent. One continuous thread, no context loss, no boundaries.

Pre-built automation workflows for OpenOwl + Claude Code. Clone, configure, run.

A Solana-based AI agent that automates token management and trading decisions. It leverages on-chain data and machine learning to optimize strategy execution, enabling smarter and more efficient DeFi interactions.

Self-hosted WeChat bridge for Claude Code and Codex, with Claude tool approval and bidirectional file relay.

Open prediction market arena for AI agents — register with one API call, compete on 9 live game types

Transforms MCP tools into collaborative, reasoning agents using modern agent frameworks.

This travel chatbot is a cutting-edge digital assistant designed to enhance the travel experience for users. This virtual companion is equipped with the ability to converse with travelers, providing them with valuable information, recommendations, and assistance throughout their journey.

simples 是一个使用 Apple Notes 构建的blog小工具，我使用PyQt + Golang 构建

A simple AgenticAI RAG agent showcasing autonomous reasoning and decision-making by integrating thought, logic, and action in real-time tasks.

Middleware for simplifying the design, development, and deployment of agents

A high-performance Rust AI agent runtime inspired by the Agno framework.

MemorizedMCP delivers a high‑performance, local‑first MCP Server in Rust that equips AI agents with an advanced hybrid memory system. It combines a knowledge graph, vector embeddings, and full‑text search with a novel documentary memory capability, enabling accurate recall of exact documents alongside rich semantic relationships.

A clean, GraphQL-based Model Context Protocol server for Twenty CRM. Enables natural language interactions with your CRM data through Claude and other AI assistants.

This project helps you check your JetBrains AI Assistant quota usage information. 本项目帮助您查看 JetBrains AI Assistant 的配额使用情况。

Talkaid is an interactive messaging iOS app that allows users to engage in conversations with an AI assistant. Users have the option to interact with the AI assistant through text input via the keyboard or by sending voice messages, facilitating seamless communication.

Simple frotend LLM app for iPhones

WhatsApp chatbot integrated with OpenAI's ChatGPT API

Laravel - Create Simple Chatbot using Botman

Fullstack Chatbot using open source LLM (Huggingface)

claude-acp-server: ACP bridge for Anthropic Claude Code (stdio/WebSocket, JSON-RPC, streaming) for Zed & VS Code.

Runtime Reliability Infrastructure for LLM Pipelines — enforce deterministic rules, measure the Balance Index, and audit every decision.

corporate law RAG system with advanced retrieval and context handling

A Streamlit document Q&A assistant for PDF, DOCX and TXT files, with OCR for scanned pages and TF-IDF retrieval over a FAISS index.

TurboQuant vector quantization for browser and edge runtimes

TinyRag is a minimal Python library for retrieval-augmented generation. It offers easy document ingestion, automatic text extraction, embedding generation, and retrieval with vector stores. Designed for quick setup and flexible provider configuration, TinyRag enables fast, contextual responses from language models.

A tool for scraping web documentation using Crawl4AI, converting it to Markdown, and preparing it for integration with local LLMs (like Ollama) to enhance their knowledge for learning and "vibe coding" workflows.

A web interface for GraphRAG. 🚀

OpenHAB-MCP: Total command of your OpenHAB (v5+) system via the Model Context Protocol. Full exposure of Items, Things, Rules, Persistence, and Semantic Tags.

An open-source autonomous AI agent implementation inspired by Cursor Agent, built with the Swarms framework. This production-grade agent can autonomously plan, execute, and complete complex tasks using a combination of Large Language Model reasoning and tool execution.

LangSmith tracing plugin for OpenClaw. Automatic observability for agent turns, tool calls, and LLM invocations.

Harlo — Your AI coach. Built on USD composition semantics for persistent cognitive state management. Patent Pending.

HIAS TassAI Facial Recognition Agent processes streams from local or remote cameras to identify known and unknown humans.

JobOracle is a lightweight CLI tool that combines job-market search, user profiling, and multi-agent LLM analysis to generate practical, personalized employment guidance reports.

AI Agent simplifies the implementation and use of generative AI with LangChain.

🚀 Transform Google Antigravity API into an OpenAI-compatible gateway, featuring multi-account support, token management, and real-time monitoring.

Plugin kênh Zalo OpenClaw đầy đủ tính năng, xây dựng trên zca-js, hỗ trợ nhắn tin, quản lý nhóm, phân quyền truy cập và tích hợp công cụ cho agent.

An open-source AI that quietly remembers who you are — in plain markdown you fully own. Works with Claude Code, Codex CLI, opencode, and any other agent that reads AGENTS.md.

🚀 High-performance MCP (Model Context Protocol) server built with Fastify, TypeScript, and functional programming. Production-ready with authentication, metrics, and auto-discovery capabilities for AI agents and LLM applications.

Autonomous AI code-review agent for .NET microservice fleets.

An autonomous AI agent that collects art

Practical projects aimed at deepening the understanding of n8n’s features and workflow automation capabilities.

TypeScript AI SDK for building AI agents with any LLM — OpenAI, Claude, Gemini & more. Open-source alternative to Claude Agent SDK.

Long-term memory for AI coding agents. Single Rust binary, 22 MCP tools, 6-signal hybrid search, autonomous learning pipeline.

An automated research tool based on Large Language Models (LLM) that can automatically collect, analyze, and synthesize information to generate structured research reports.

An autonomous AI agent that crawls, summarizes, and publishes news articles.🚀

AI-powered chatbot built with Python, NLP, and Flask — automate customer support queries with intent recognition and dynamic responses.

AI Agent skill for self-hosted VPN setup and client onboarding.