Custom agents

GenAI & agent toolkit for Apple Silicon Mac, implementing JSON schema-steered structured output (3SO) and tool-calling in Python. For more on 3SO: https://huggingface.co/blog/ucheog/llm-power-steering

Real-time decision features without streaming infra. Turn live events into product reflexes — no Kafka, no Flink, no feature store.

Multi-agent LLM driven cell type annotation for single-cell RNA-Seq data

A cyber security toolkit blending manual testing with AI Agents

[AAAI-2025] This repository contains MAPF-GPT, a deep learning-based model for solving MAPF problems. Trained with imitation learning on trajectories produced by LaCAM, it generates actions under partial observability without heuristics or agent communication. MAPF-GPT excels on unseen instances and outperforms learnable state-of-the-art solvers

PentestAgent is a novel LLM-driven penetration testing framework to automate intelligence gathering, vulnerability analysis, and exploitation stages, reducing manual intervention. For more information, read our paper at https://dl.acm.org/doi/10.1145/3708821.3733882

Open Source policy enforcement proxy: Make your agents follow rules.

Engineering LaCAM*: Towards Real-Time, Large-Scale, and Near-Optimal Multi-Agent Pathfinding (AAMAS-24)

A lightweight, high-efficiency desktop Agent assistant with multi-task parallel capability.

Agent skill for managing Linear issues, projects, and teams. MCP tools, SDK automation, GraphQL API patterns.

This accelerator unifies enterprise data with Microsoft Fabric and applies agentic AI to enable faster, smarter decisions. With Microsoft Foundry agents and the Microsoft Agent Framework orchestration, teams can automate workflows, query data in natural language, and uncover insights that fuel agility, collaboration, and innovation.

Disciplined Agentic Engineering — a methodology kit for Claude Code: acceptance-test-first specs, explicit checkpoints, and autonomy you can actually leave running. The engineer, atdd and crap-analyzer plugins.

Open-source Claude Code skills and Codex skills for AI-first work. Audit, re-engineer, and bootstrap projects with AI-first design principles.

Stop your AI from writing like AI. A universal system prompt eliminating every known LLM style tell — works with Claude Code, Gemini CLI, Codex CLI, Copilot, Cursor, and any web AI.

An agentic development environment for AI coding agents

🤖 Advanced AI-powered virtual assistant with voice recognition, face authentication, phone integration, and intelligent automation capabilities.

Open, MIT context format for AI coding agents: centralized .aigx/ rules + a per-file boundary index that targets the right rules, forbidden imports, and gotchas to each file. The only context format validated in a controlled benchmark. Tool-agnostic; nothing injected into your source.

An agent small enough to run anywhere. A minimal agentic runtime in C — ~380 lines, 54KB binary, ~2MB RAM. A skill file + an LLM + a shell loop, no framework.

NEWTON: Agentic Planning for Physically Grounded Video Generation

Sesori iOS/Android app and the Sesori Bridge CLI — drive Claude, Codex, OpenCode, Cursor, Pi, OMP, Hermes coding sessions from your phone

AI PDF reader for macOS and Windows: ask about pages, diagrams and code, with local reading history. AI PDF 阅读器与教材问答。

器灵 Soulmate — 多 AI CLI 编排与社区运营入口

基于Fastapi + React 实现的面向智能知识管理的Agentic RAG系统，支持私有部署和多租户，以及知识图谱、多智能体、MCP/Skills、沙盒与权限管理等功能。

Orbit is a native desktop workbench for the pi coding agent — a chat-style GUI rendered entirely in Rust on GPUI, the GPU-accelerated UI framework Zed is built on. It speaks the pi CLI's RPC protocol directly over stdio: no browser, no webview, no Node.

Shared terminal sessions for humans and AI agents, starting with serial and hardware debugging.

DeepSeek Harness 插件：让 AI 以「大肥鱼」住进 QQ（SnowLuma / OneBot v11）。表情、语音、记忆、空间、B 站、翻唱、生图全部模块化——可关、可换、可本地自写扩展，连群聊策略、提示词、回复和模型都能改，保存即生效。

Everything you need for building an agent. A Go library for the LLM tool-calling loop with budgets, hooks, skills, and tracing.

Exact B-Rep CAD in TypeScript, with a Claude Code plugin that lets agents write parts and check them on OpenCascade.

Automated Web Penetration Testing Assistant Based on Function Calling and Multi Agent.

🚀 Production-Ready Multi-Agent System for WeChat Article Generation - Automated pipeline that actually works! | 真正能用的微信公众号AI生成系统：多Agent协同、自动化处理、通俗易懂

开源的智能体项目 支持6种聊天平台 Onebotv11一对多连接 流式信息 agent 对话keyboard气泡生成 支持10+大模型接口(持续更新) 具有将多种大模型接口转化为带有上下文的通用格式的能力.

whatsapp AI chatbot for eCommerce websites

A Rails agent framework for RubyLLM — define AI agents with prompts, schemas, caching, logging, cost tracking, and a built-in dashboard for monitoring LLM usage in production.

Leveraging Multi-Agent LLMs for Automated Knowledge Graph Enrichment

Jax Official Implementation of T-RO Paper: Songyuan Zhang*, Oswin So*, Kunal Garg, Chuchu Fan: "GCBF+: A Neural Graph Control Barrier Function Framework for Distributed Safe Multi-Agent Control".

Universally Triggered Agent Harness - An OpenClaw-like Inngest-powered personal agent

A structured Claude Code workflow for empirically replicating published research using UK Biobank (UKB) data, created by Chen Zhu 朱晨 | 遗传社科研究.

Medea: An AI agent for therapeutic reasoning across biological contexts

ChinaTravel: A Real-World Benchmark for Language Agents in Chinese Travel Planning

从0到1构建 Claude code (学习版)

Built from the ground-up for agents, Vessel Browser is an open source AI browser for Linux/Mac/Windows that provides a durable state, MCP control, and BYOK with full autonomous browsing. Use with Hermes Agent, OpenClaw, or connect to your favorite API provider.

AgentRE-Bench is an agentic benchmark that evaluates state-of-the-art models on long-horizon reverse engineering tasks, measuring their ability to analyze binaries, use tooling effectively, and reason over multi-step execution artifacts

Automated Web Penetration Testing Assistant Based on Function Calling and Multi Agent.

Ai agent claude trading skills for disiplined, repeatable trading workflows with a modern typescript platform

AISW | AI Switcher - Switch between multiple Claude Code, Codex CLI, Antigravity and Gemini CLI accounts in one command. Named profile manager for AI coding agents.

GAN-style self-improvement loop for any text artifact: mutate, grade with a SEPARATE model, keep only verified wins (pairwise-judged), revert the rest. The git history is the improvement log.

Let AI agents share your everyday, logged-in Chrome through a Skill + CLI — locally or across multiple machines.

Research-grade investment decision engine for AI agents: isolated multi-agent committee, auditable verdicts, backtests with lookahead protection, published negative results

Donna — a starter Hermes Agent profile: opinionated persona, 73 curated skills, guided first-run orientation, optional Token Router. MIT.

Unofficial Agent Skill for Claude Code and other SKILL.md-compatible agents, driving ArvanCloud's APIs (CDN, DNS, IaaS, Object Storage, Edge, CaaS, VOD/Live/Video Ads) and Let's Encrypt DNS-01 certs via acme.sh.

Rust CLI powered by Jev from TypeSafe.ai that ranks agent skills for the next step using live session context. Includes Claude Code hooks, structured JSON, abstention, and local feedback. Requires a TypeSafe API key.

Mission control for AI coding agents. Watch Claude Code, Codex, Gemini and a dozen more work live in one macOS terminal: live status, an editor that follows their edits, undo for any turn, approvals from your phone and local models. 20+ built-in tools, all on the site below. Built on Ghostty.

ARGI [/ˈɑːr.dʒiː/]: Agent Runtime and Graph Intelligence 🚀

本地优先的个人 AI Agent 助手，支持记忆、自主任务、工具授权和工作流。（最近工作，更新缓慢）

Cooperative Driving Dataset: a dataset for multi-agent driving scenarios

Chrome Built-in AI Demo page

Robotic Order Fulfillment System Using Kiva-like mobile robots. Conflict-based Search multi-robot path planning.

An async-first framework for building multi-agent AI systems with an innovative approach to parallelism, so you can focus on creating intelligent agents, not on managing the concurrency of your flows.

Chatbot Development Framework (with Slack integration for Jira and Jenkins)

LYRAI is a Model Context Protocol (MCP) operating system for multi-AI AGENTs designed to extend the functionality of AI applications by enabling them to interact with financial networks and blockchain public chains. The server offers a range of advanced AI assistants, including blockchain public chain operations (SOLANA,ETH,BSC,etc.)

A document based RAG application

Security gateway for AI agents - credential-isolated API proxying and policy-gated remote execution (conclaves). Reduce the blast radius!

一句需求，自动拆任务、并行多代理写码、断点恢复、验证收尾。FlowPilot 让 Claude Code / Codex / Cursor / snow-cli 进入真正的自动开发流程。

A Python-based application that enables dynamic conversations between multiple AI models in a graphical user interface. Originally designed for exploring liminal AI interactions, it's evolved into a flexible platform for multi-agent shenanigans.

Learn production-grade AI agent development from Claude Code's architecture

An opinionated framework for building sophisticated AI Agents

AI, your emotional partner

Make autonomous landing rockets using Deep Reinforcement Learning (Unity ML-Agents)

A framework for the creation of autonomous agent services.

xPrivo is the privacy-first AI assistant that you can self-host. No accounts, never. Free as in freedom.

[ICLR'26] MedAgentGYM: Training LLM Agents for Code-Based Medical Reasoning at Scale

A self-hosted AI assistant for academic research — manages your papers, searches literature, tracks deadlines, and answers you on the channels you already use.

Video summarizer agent skill for Claude Code, Codex, ChatGPT & OpenClaw — YouTube/Bilibili/podcast via BibiGPT CLI/MCP/API (bibi)

A programmable Voice AI platform: SIP and WebRTC call control, multi-party mixing, recording, TTS/STT, and pluggable AI agents (ElevenLabs, VAPI, Pipecat, Deepgram) — all driven through a REST API, webhooks, and a WebSocket event stream

A curated collection of resources, frameworks, papers, and best practices for designing, evaluating, and deploying agentic AI systems—from architecture patterns and safety considerations to real-world applications.

Agent skills for fallow. They teach coding agents (Claude Code, Codex, Cursor, and others) to check health, complexity, duplication, architecture, styling drift, and unused code with fallow.

A scrolling terminal multiplexer that tracks your AI coding agents.

专注于 VLM、VLA、世界模型及通用具身智能等方向，收录前沿论文、开源代码与数据集，追踪从感知到决策的下一代智能体技术。 A curated collection for multimodal intelligence research, covering VLMs, VLAs, World Models, and embodied AI — tracking next-generation agent technologies from perception to decision-making, with a focus on papers, code, and datasets.

Codex-first Agent Skills library for premium frontend, website, motion, accessibility, QA, and handoff workflows.

Agent-assisted maintainership and development framework for Apache projects — Triage, Mentoring, Drafting (agent-authored fixes with human review), and Pairing (developer-side dev-cycle) skills shipping; Agentic Autonomous (auto-merge) on the roadmap.

Agent-native browser with fingerprint control and proxy support.

Same model, different wrapper: a from-scratch benchmark comparing coding-agent harnesses (codex, pi, opencode, cursor, devin) and open models on correctness, speed, and token cost

An improved multi-agent AIOps and RAG platform for OnCall troubleshooting, featuring LangGraph-based diagnosis workflows, Milvus vector search, MCP tool integration, Prometheus alert knowledge from awesome-prometheus-alerts, and sanitized configuration for safe public deployment.

MCP server + CLI that puts a real human in your coding agent's loop. It publishes work mid-task, a reviewer pins what's wrong and returns a verdict, and the agent iterates until approved.

Keep your Mac awake on your terms. A native macOS menu-bar app with condition-based triggers, closed-display mode, and a headless-Mac toolkit. Built for AI agents, servers, and always-on Macs.

CometChat docs search + implementation bundles: add chat, voice, video & moderation to your app through your AI coding agent.

「摘星」· 微信助手 — AI Agent 智能助手 / AI 群聊摘要 / 关键词即时提醒 / 公众号摘要与文章实时推送 / RAG 语义检索 / 聊天、朋友圈归档 / 收藏导出 。本地运行，数据不出本机。

An open-source agent harness for software work.

A lightweight CPU only memory approach with ranked retrieval. Simple, yet effective.

The model is the engine; the harness is the car. Curated knowledge on harness engineering — context management, tool design, agent loops, memory, sandboxing, evals. EN/中文

Kapso by Leeroo: coding agents that work a goal to a verified outcome, measured by an evaluation they cannot touch, every step traced, and the lessons carry their receipts. First on RelBench and ALE-Bench, top open-source on MLE-Bench, above every human contestant at IOAI 2026.

Agent Skill for complex work: research before asking, ask before planning, plan before building, verify before delivering, independent review before calling it done. Plain text, no runtime.

A Codex skill that delegates work to ChatGPT on the web.

6688 live x.ai/bot shares for Grok Bot — curated catalog with shelf tiers, searchable Pages, bilingual README. Not a prompt dump.

Orchestrator for AI coding (claude code, cursor, codex, gemini)

Agent skill: examine a product you're authorised to rebuild, build your own, then test against the original and fix gaps until every must-have on a written checklist passes. Inspired by REA (@morluto), not affiliated. MIT.

A token efficient, local LLM friendly agentic coding harness: a complete agent runtime (agent loop, tool pipeline, sandboxing, session persistence) with four front-ends — a Blazor web UI, a headless CLI, and JSON-RPC / ACP stdio servers for editors and automation.

2 character sheets + 1 line of story → a finished AI anime episode. Agent skill for Claude Code / Codex, powered by Renoise MCP.

Observability platform for Digital Employee, providing real-time tracing, session insights, and cost analysis for multi-agent workflows

Bringing Generative AI to the way the Civil Service works