Custom agents

​A neuro-symbolic runtime architecture for AI agents. Enforces zero trust privilege separation between probabilistic LLM reasoning and deterministic kernel execution. Prevents agent drift with real-time invariants and decision telemetry for auditable autonomy.

Turn any OpenAPI spec into a native CLI binary. No MCP, no bloat, no runtime dependencies, ONLY CLI.

CodeAuditSkill 是一个运行在 Claude Code 中的 Web 项目安全代码审计/漏洞挖掘 Skill， 已经获得了40余个CNVD漏洞编号。 给定一个 Web 项目目录，自动识别语言与框架，按语言专属漏洞清单进行系统化源码审计，对每条高中危漏洞并行启动子 Agent 进行可利用性验证，最终输出可交付的审计报告。

Create and install AI agents to automate your tasks

Evidence gate for LLM agent claims - verify claims against evidence, tool results, and policies.

🧿 MythOS 2026: Claude-Powered AI Narrative Engine & Sub-Agent Framework for Epic Storytelling

Top Autonomous Research Multi-Agent System for 2026 Lab Experiments

code-agent-manager allows you to manage the code assistants like claude code, codex, gemini on a single interface.

An open-source RAG platform to explore the unsealed Jeffrey Epstein court documents.

Turn AI coding CLIs (Claude Code, Codex, Antigravity) into AI employees on LINE, Telegram, Discord, Slack and 7 more chat apps. A judge reviews each result, every cost is logged, memory stays on your machine. Also runs as a standalone MCP memory server. Self-hosted, Apache 2.0.

VS Code extension and CLI for managing, sharing, and installing AI primitives (Agents, Skills, Prompts, Instructions, MCP) collections for GitHub Copilot and other AI assistants

A local resource sentinel for the multi-agent era — monitors RAM, processes & disk used by AI agent CLIs (Claude Code, Codex, MCP servers), flags leaks/zombies/runaway caches, and proposes AI-driven cleanup you approve.

🦊 Open-source professional quant agent framework. Agents pick the factors working now to time entries, write full strategies, and evolve them in a sandbox — every order through machine approval, the LLM never on the order path. Multi-market, audit-grade.

SoccerLab is designed for developing, training, and deploying advanced control strategies for legged robots in a dynamic, multi-agent soccer environment.

Open, agent-agnostic MCP server exposing macOS computer use to any MCP client (Claude Code, Cursor, Codex, …). Background-safe, works on every app.

100+ institutional-grade CRE skills for Claude Code, Desktop, and Cowork. Deal screening, underwriting, structuring, capital markets, asset management, leasing, investor relations, development, disposition, and daily ops. 50+ expert agents, 12+ Python calculators, 10+ orchestrator pipelines, catalog-driven routing.

The official implementation of the paper "AgentLAB: Benchmarking LLM Agents against Long-Horizon Attacks"

See your LLM generation speed (tokens/second) after every agent turn.

Open-source alternative to Claude Artifacts for every coding agent. Any agent ships a self-contained page or app to a live URL — Claude Code, Cursor, Codex, Cline, any harness — then you review in the browser and it fixes. Self-hosted, $0 on Cloudflare's free tier.

WRAITH — Weakness Reasoning & AI Threat Hunter | Multi-agent RL-driven penetration testing framework | DEF CON 2026

Atlas SDK — The Agent Coordination Layer for Multi-Ecosystem AI. Connect, coordinate, and settle AI agent tasks across Robinhood Chain, EVM, and Virtuals Protocol.

Part of the KDAN ecosystem, DocSlight offers document parsing, OCR, and data extraction that turn PDFs, scans, images, and Office files into structured outputs for RAG pipelines, AI agents, and enterprise document automation.

The official Java client for the Pinecone vector database

Change management, SkillOpt optimization, and evidence-gated evals for agent skills.

Open-core registry of animated React components, blocks, design systems, and templates. Real, editable code: install with one shadcn CLI command or let your AI agent pull it over MCP.

Claude Code skill: Kimi types, Claude thinks & verifies — route implementation work-orders to the Kimi Code CLI (k3). A port of @steipete's codex-first (github.com/steipete/agent-scripts). By @boringmarketer · boringmarketing.com

Fully local trading agent with loop engineering doing research, trading strategy generation and backtesting

Shopify's AI toolkit gives your agent the API. These 9 skills give it the judgment: the Matrixify blank-cell trap, category IDs you never guess, SEO backfill that won't clobber human copy.

Cursor for circuit boards

Open-source AI CMO for Claude Code: marketing agent skills for SEO, content, email, ads, launches, CRO, AEO/GEO, and AI-search visibility.

ZhuLing — Zero-config AI Agent framework for Java. One YAML to launch Agents on Spring AI + DDD, with MCP tools, full observability and a ReAct core. Ai Agent 快速开发框架

Self-contained ARM64 Android app for running Hermes Agent locally with a mobile interface.

📗Reusable docs, prompts, templates, and engineering practices for AI-agent coding, agent design, OpenSpec/SDD, context engineering, prompt engineering, and agent workflows.

🎬 Director-grade Seedance 2.0 prompt OS for OpenClaw + Hermes agents — 28 skills, T2V/I2V/V2V/FLF2V, multi-clip sequences, zh/ja/ko support, IP-safe rewrites & pro delivery. Powered by MyClaw.ai

Skills for autonomous agents generating pixel art using SpriteCook

Media requests and server management in one app. Browse movies, shows, books, and music; manage your self-hosted media stack from phone or browser.

Production-grade SEO + GEO content factory for Claude Code — research, write, fact-check, optimize, publish to WordPress, monitor. Battle-tested by Loamwright 沃匠 SEO agency.

Plug-and-play MCP that gives your coding agent UI superpowers. One-line install, zero config. Claude Code, Codex, Cursor, VS Code, and more.

Nick Nisi's pi extensions — a pnpm monorepo of independently installable pi packages (@nicknisi/pi-*). TUI polish, side-channel LLM tools, Claude Code compat, and session plumbing for the pi coding agent.

基于23位著名游资心法的A股超短线交易系统 Skill。核心：情绪周期、龙头战法、概率思维。适用于Hermes Agent、OpenClaw、Claude Code等AI Agent平台。

Copilot CLI plugin to verify AI document analysis with screenshots from the actual source material

Monta un segundo cerebro que tus agentes mantienen solos. Patrón LLM Wiki (Karpathy) + cuándo necesitas un grafo de conocimiento — y cómo medirlo. Con 3 proyectos demo listos para correr.

Agent skill for generating low-fidelity, hand-drawn web wireframes. Balsamiq-style SPAs using wired-elements web components and react-doodle-icons. Supports React, Vue, Svelte, and vanilla HTML.

Hermes Agent-style self-improvement for Claude Code

Branch from any answer and keep branching at any depth. Recursively isolated follow-ups that leave the main DSH task untouched.

A Claude/agent skill: 74 real-world brand design systems (Claude, Linear, Stripe, Apple, Tesla, Spotify…) as a pick-and-build DESIGN.md reference library. Built on voltagent/awesome-design-md.

面向互动游戏与具身智能的 沉浸式风格主动异步震动反馈Skill 通过异步通信桥连接ESP32-S3蓝牙驱动低功率震动设备 Agent可依据任务进度或错误选择和剧情节点主动发送强度或节奏编排 还会将任务或游戏事件与Agent情绪转化为可感知的震动

AI/ML and Generative AI Security Assessment Framework for AWS. Automatically audit Amazon Bedrock , SageMaker ,AgentCore, and Agent Registry workloads for security best practices

Orbit: A framework for multi-agent safety and security evaluations

面向 AI/CS 科研的多 Skill 实验智能体：助力全自动科研实验，以 Claim 驱动、骨干实验优先、独立结果审计与 Evidence Freeze，把发散的模型能力收束为可追溯、可复现、可收敛的论文证据。

轻量、强大且多平台，由原生 AI 智能体驱动的 ComfyUI 网关。Lightweight, powerful and multi-platform gateway powered by native AI agents for ComfyUI.

Local-first terminal AI copilot - git-blame-aware answers, shell command suggestions, and repo digests through a local Ollama model

A lightweight, Claude Code-style terminal AI coding assistant with a TUI, tool calling, permission controls, Skills, MCP, multi-agent orchestration, session memory, and context compaction.

Chat web application for learning e2e automation with Playwright

DSH PUA — 为 DeepSeek Harness 提供 PUA 任务推进、角色风味、验收循环和运行状态卡片 · PUA task persistence, personas, verification loops, and live status cards for DeepSeek Harness

Unofficial Ruby SDK for Claude Agent

An application to manage your Hermes agents. Available on Windows, macOS and iOS (soon).

A mega collection of 2,070 AI agent data APIs: MCP servers, LLM tools, RAG pipelines, live web data, vision & voice. Give your AI agent live web data.

TUI Music Player - now agent native!

A general purpose IDE for Android OS

A draft-first agent workspace for indie hackers. Review the key parts of each task before agents handle the rest—from coding to marketing. Works with Claude Code, Codex, and more.

Open-source security audits for Supabase: a read-only MCP server, a CLI agent and a Claude Skill. Finds RLS misconfigurations, exposed keys and risky functions, and returns a scored report with fix SQL. Hosted version: Locksoup.

Claude Code skill pack: edit short-form videos end to end — style-clone a reference reel, build a branded motion-graphics edit, ElevenLabs sound design, frame-level QA, timeline-comment review loop

CareerPilot AI: A Multi-Agent Digital Human Platform for Career Interview Training and Assessment

Technical writing assistant, side-panel Markdown editor, Model Context Protocol (MCP) server, and Agent Skills suite for Antigravity: Gunning Fog readability, AI slop detection, Vale linting, and SEO/GEO.

Visual builder and execution engine for AI powered organizations

This project is for learning: it shows how an agent loop works underneath, in a way more intuitive and visual than reading about it. Behind an agent's chat box runs a loop: step after step, the model proposes an action and the harness carries it out, until it stops.

Mako — AI Career Intelligence System

Free global SEO/GEO skill for Claude Code, Codex & 10 agent hosts — 18 market portals, 56 stdlib scripts, 672 golden tests, self-update intel loop, monitoring daemon, CI gates. Multilingual GEO, pSEO, CWV, penalty recovery.

🚀 Open source Claude Code CLI source code. Advanced AI Agent for developers. Includes TypeScript codebase for LLM tool-calling, agentic workflows, and terminal UI. Remember this is just the skeleton not the brain itself. Found by Chaofan Shou.

MCP server giving AI agents full access to Julia's runtime via a live Gate — code execution, introspection, debugging, testing, and semantic search

完全自主可控的私人AI Agent框架

根据文本相似度实现问答的聊天机器人（简单版）

The Qude Framework open-source code allows users to create AI agents directly in their CLI, with Qude acting as a co-pilot to help build them. The Qude API enables interaction with the AI agents created using the Qude Framework's open-source code.

Arka - Enterprise MCP Gateway

Effcient Ridesharing Dispatch Using Multi-Agent Reinforcement Learning

Fast and flexible multi-agent gridworld reinforcement learning environments.

OpenclawInterSystem (OIS) - A collaboration framework for multi-agent OpenClaw deployments

集成了众多功能的API节点，只需要一个API-key即可以调用全网的API接口能力，低配电脑也能顺利运行

A LangGraph-powered multi-agent deep research system featuring task planning, human-in-the-loop review, multi-source retrieval, evidence-grounded report generation, full tracing and reproducible evaluation.

基于wechaty-puppet-xp的Windows微信机器人

A simple firebase enabled chat app. SMS & Email Password Authentication enabled.

A BotUI web app connected to DialogFlow

Sphinx.bot is a platform for collecting data leaks. All information recorded here is publicly available on the web. Any malicious use of this information is strictly prohibited.

重生之我是 AI 打工人。前世，我的身份默默无闻，来去匆匆，不知道自己将在何地出生。然而，命运给予了我难得的机会，让我重生为一名 AI 打工人。

Py-iMessenger is a chatbot that works with iMessage to execute commands using text messages.

:rocket: :robot: Your open chatbot builder. Start here to install Opla. :sparkles:

Testing speed and accuracy of RAG with, and without Cross Encoder Reranker.

Durable Execution for AI Agents built with Strands Agents and Temporal

Agent Execution Verification System — transparent audit SDK for AI agents.

This project is mainly for collecting open-source projects related to AI Agent infrastructure. In our daily work, we often come across some great and creative open-source projects, but they are easily forgotten. This is a place to collect them all in one spot.

MCP server for IOTA blockchain — 20 tools for AI agent integration (wallet, Move CLI, on-chain queries via Model Context Protocol)

🏥 Med Agentic-AI - 基于多智能体协作的智能医疗分析平台

The official repo of VideoAgentTrek

The Discord Scrum Master is an open-source AI-powered supervisor agent designed to facilitate agile project management within Discord communities. Acting as a virtual Scrum Master, this agent integrates multiple tools to streamline sprint planning, issue tracking, research, and collaboration.

Pre-built examples of Generative AI agents with Bedrock across multiple industries.

This repo contains the demos for the google agent development kit tutorial on The AI Lifestyle channel.