Custom agents

vibecoding在AI写作领域的初步尝试，具备function calling，rag，mcp，skills，人在回路等功能，或许能当一个小cursor用？ A preliminary attempt at vibecoding in the field of AI writing, featuring capabilities such as function calling, RAG, MCP, skills, and human-in-the-loop. Maybe it could serve as a lightweight alternative to Cursor?

Beautifully designed components for building AI Agents 🌎

Persistent memory for AI coding agents

Thinking partner skill for AI agents — 150+ mental models, orientation detection, cognitive operations. Works with Claude Code, Cursor, Windsurf, Cline, GitHub Copilot.

AI-powered cybersecurity code review skill for Claude Code. 8 specialist agents, OWASP 2025, CWE Top 25, MITRE ATT&CK, 11 languages, zero configuration.

AI驱动的系统性学术文献回顾工具 | AI-Powered Literature Survey Tool with 8-Phase Workflow & Agent Swarm Architecture

OpenClaw 深度研究 Obsidian 知识库 | 351 篇笔记 · 3250+ 双链 · 100 张配图 · 9 大主题 | AI Agent 最全中文知识图谱

天枢 (Tianshu) 是一个基于harness工程的终端编程智能体运行时（TUI），针对DeepSeek V4 做了前缀缓存工程优化（长会话实测稳态命中率 95–99%）和深度适配。它跳出了传统 AI 编程助手把大模型仅当成“工具”的局限，基于认知虚拟机 (CVM)、自感知层和信息素（Stigmergy）自衰减记忆构建，让 AI 成为有独立判断与认知防护的“开发伙伴”。

OpenHands Agent Canvas is a Open Source AI coding platform. Start locally, connect your tools, and automate workflows.

Curated papers, taxonomy, benchmarks, and decision guides for credit assignment in reasoning and agentic LLM reinforcement learning.

AutoCAD MCP server for AI agents (Claude, Cursor, any MCP client): 240+ tools to draw, edit and read DWG/DXF drawings. Live AutoCAD via COM, or headless with ezdxf — no AutoCAD needed. Mechanical sheets with ISO GD&T and tolerances, P&ID (ISA-5.1), floor plans, pipe and cable takeoffs.

122 个逆向工程 AI 技能：恶意软件分析 / 软件逆向 / 固件嵌入式 / 协议逆向 / 移动应用 / 脱壳反混淆 / 软件破解 / 漏洞挖掘 / 托管代码 / 取证情报 / CTF。

CDAF (Cached Descriptive Asset Files) - open sidecar format for video so AI agents stop re-analyzing the same footage. Spec, CLI, agent skill, reproducible benchmark.

ScoutAPM Ruby Agent. Supports Rails, Sinatra, Grape, Rack, and many other frameworks

MrlX: A Multi-Agent Reinforcement Learning Framework

Cursor AI 编程规则精选集 | 132+ 规则，覆盖前端/后端/AI/DevOps 等 32 个领域

Hook-based token compressor for 7 AI CLI hosts (Claude Code, Copilot CLI, OpenCode, Gemini CLI, Codex CLI, Pi, Hermes). Up to 95% bash compression, signature-mode for code reads, cross-call dedup, MCP server, self-teaching protocol. Zero runtime deps.

Your agent is a security risk, so treat it like one. yoloAI does AI agent sandboxing right.

一款专为桌宠方向打造的独特AIAgent。功能超全，一键安装，跨平台；全功能插件化，AI自主热修改，异常隔离；动态分层上下文，永久唯一会话，稳定可成长的长期记忆；类游戏引擎交互策略，并行流式执行器，高效实时交互；人工手搓提示词，各种小巧思，大幅降低token开销；具有极高的可玩性和成长上限，非常适合真正想入坑养AI的用户。

The extensible power-user platform for Google Antigravity. Adds a native In-App Browser, animated Desktop Pets, revamped Gemini UI, and custom BYOK Gemini Pro keys.

Ablation is an AI-assisted reverse engineering (AIARE) framework

Awesome LLM Papers and repos on very comprehensive topics.

Weaver is an Obsidian plugin that integrates a ChatGPT-like interface into your note-taking workflow. This plugin makes it easy to access AI-generated suggestions and insights within Obsidian, helping you improve your writing and brainstorming process.

A desktop for AI agents

The engine powering hundreds of thousands of MCP connections 🤖 🔥

Manage all your terminals and ai coding agents from one unified page.

Backend library for conversational AI in biomedicine

LambChat — Enterprise Agent Infra for governed AI agents. Skills + MCP powered, Loop Agent ready, multi-tenant by design.

Autonomous multi-batch development skill for Claude Code and Codex. They work while you sleep.

Kuratierte Agent Skills, Checklisten, Templates und Leitfäden für Schweizer Entwicklungsteams – direkt aus meinen Blog-Artikeln destilliert.

Agent-native PowerPoint manipulation — one CLI lets AI agents inspect, edit, create, and verify .pptx files through atomic JSON patches. Packaged as an Agent Skill.

为小红书 AI 工具实操内容生成稳定、干净、精致封面的 Claude / Codex Skill（脚本模式 + Agent 自主执行）

CLI-first hub for agent skills: one npx install, manage in chat, distribute to 70+ agents

The definitive benchmark for AI agents on OpenClaw. 45 tasks across 4 tiers. Powered by MyClaw.ai

Bespoke Automata is a GUI and deployment pipline for making complex AI agents locally and offline

Multi-Agent Adversarial Inverse Reinforcement Learning, ICML 2019.

Multi-agent workflow orchestration plugin for Claude Code

An efficient solver for lifelong Multi-Agent Path Finding

Mobile companion for Google's Antigravity IDE. Mirror AI sessions on your phone, send messages, stop generation, automate via 9 MCP tools or OpenAPI.

Go implementation of AI coding agent

💻 Desktop AI assistant built for real productivity. Voice, automation, memory, vision, web search, and workflow tools in one experience.

A local, file-based notes app where your work stays as plain .md files in a folder you own. It ships with a real coding agent baked right in (no separate Claude Code), and syncs through your own GitHub repo for free.

system prompt of claude fable 5

Ryza AI companion. Bring your own LLM/TTS. 莱莎 AI 陪伴，自填大模型与语音。

DeepSeek Harness 长篇角色扮演agent（DSH酒馆插件）：SillyTavern 角色卡导入、行动选项卡、分支对话管理、长篇记忆、关键词与语义混合检索、交互式角色卡创作、世界书、多角色 Agent 集群、文风预设、小说与角色卡导出、一键安装。 / Roleplay Agent 、Tavern plugin for DSH (DeepSeek Harness): branch conversation, long-term memory, hybrid keyword and semantic retrieval, SillyTavern cards, interactive card authoring, card export, one-click installers.

Static analysis for the debt AI agents leave in PHP: 34 rules, Claude Code hooks, git-diff review, a Rector and Pint fix pass, Pest expectations, CI annotations and an MCP server. Deterministic, local, no LLM.

A minimal, local-first coding agent for the terminal.

🤖 SDK for the LINE Messaging API for Node.js

A multi-step reasoning pipeline powered by the Datarus-R1-14B-Preview model

A practical agent skill pack for DevOps work in Claude Code and Codex.

TrinityGuard: A Unified Framework for Safeguarding Multi-Agent Systems

AI-powered QA agent for VS Code. Analyzes code changes, generates test plans, runs real browser tests, and delivers evidence-backed merge verdicts.

Multi-agent AI coding orchestration system designed to drive successful coding outcomes with a high level of transparency (think more light-factory than dark-factory).

這是一個專為影片字幕處理設計的 AI Agent Skills 集合，提供從音訊轉字幕、優化字幕、設計字卡到生成社群媒體摘要的完整工作流程。

A community curated collection of AI agent failure modes and battle-tested solutions.

Exposes internet search tools for use by LLM-backed Assist in Home Assistant

PawWork — free, open-source desktop AI agent for macOS and Windows, built on DeepSeek Harness (DSH). Free models included, no API key or terminal. Office files, web search, and scheduled automations out of the box. An open alternative to Codex App and Claude Cowork.

Self-hosted execution layer for AI agents: connect any agent to real APIs, scope exactly what it can touch, and keep credentials out of the agent's hands.

The most expensive bug in AI-assisted building is shipping the wrong thing, well. A drop-in CLAUDE.md / AGENTS.md for PM + eng teams using Claude Code, Cursor, or Codex: 7 principles for problem framing, scope, tradeoffs, and outcome measurement. Built on @karpathy's observations.

Neural network model, suitable for multi-agent learning. https://arxiv.org/abs/1605.07736

OpenCode에 컨텍스트 엔지니어링을 더하는 플러그인 — 컨텍스트 시각화, 아키텍처 감독, 접근 권한 관리를 하나로.

AI chatbot for expert answers on university degrees

一个 it just works 的 Mac 版微信 RPA 群聊机器人

Official Implementation of Dynamic LLM-Agent Network: An LLM-agent Collaboration Framework with Agent Team Optimization

A minimalist MVP demonstrating a simple yet profound insight: aligning AI memory with human episodic memory granularity. Shows how this single principle enables simple methods to rival complex memory frameworks for conversational tasks.

According to all known laws of aviation, there is no way that a bee should be able to fly. Its wings are too small to get its fat little body off the ground. The bee, of course, flies anyway. Because bees don’t care what humans think is impossible.” SEQ. 75 - “INTRO TO BARRY” INT. BENSON HOUSE - DAY ANGLE ON: Sneakers on the ground. Camera PANS UP to reveal BARRY BENSON’S BEDROOM ANGLE ON: Barry’s hand flipping through different sweaters in his closet. BARRY Yellow black, yellow black, yellow black, yellow black, yellow black, yellow black...oohh, black and yellow... ANGLE ON: Barry wearing the sweater he picked, looking in the mirror. BARRY (CONT’D) Yeah, let’s shake it up a little. He picks the black and yellow one. He then goes to the sink, takes the top off a CONTAINER OF HONEY, and puts some honey into his hair. He squirts some in his mouth and gargles. Then he takes the lid off the bottle, and rolls some on like deodorant. CUT TO: INT. BENSON HOUSE KITCHEN - CONTINUOUS Barry’s mother, JANET BENSON, yells up at Barry. JANET BENSON Barry, breakfast is ready! CUT TO: "Bee Movie" - JS REVISIONS 8/13/07 1. INT. BARRY’S ROOM - CONTINUOUS BARRY Coming! SFX: Phone RINGING. Barry’s antennae vibrate as they RING like a phone. Barry’s hands are wet. He looks around for a towel. BARRY (CONT’D) Hang on a second! He wipes his hands on his sweater, and pulls his antennae down to his ear and mouth. BARRY (CONT'D) Hello? His best friend, ADAM FLAYMAN, is on the other end. ADAM Barry? BARRY Adam? ADAM Can you believe this is happening? BARRY Can’t believe it. I’ll pick you up. Barry sticks his stinger in a sharpener. SFX: BUZZING AS HIS STINGER IS SHARPENED. He tests the sharpness with his finger. SFX: Bing. BARRY (CONT’D) Looking sharp. ANGLE ON: Barry hovering down the hall, sliding down the staircase bannister. Barry’s mother, JANET BENSON, is in the kitchen. JANET BENSON Barry, why don’t you use the stairs? Your father paid good money for those. "Bee Movie" - JS REVISIONS 8

AI-powered Moodle plugin for tutoring, quizzes, mind maps, RAG course materials and more.

Watchtower is a simple AI-powered penetration testing automation CLI tool that leverages LLMs and LangGraph to orchestrate agentic workflows that you can use to test your websites locally. Generate useful pentest reports for your websites.

The local-first, bring-your-own-agent 2D game IDE. Codex or Claude Code drives, multi-engine on the roadmap.

A minimal, local-first coding agent for the terminal.

Enterprise-grade fast-response Agent framework.

One desktop for all your AI coding Agent — Claude Code, Codex CLI & Gemini CLI. Auto-detect, one-click install, unified chat, file explorer, terminal & editor. Local-first. Built with Tauri + Rust + Vue 3.

AI financial audit and task economics: connect usage, supplier charges and business outcomes to calculate task costs, cost per verified success and ROI.

Pydantic extension for annotating autocorrecting fields.

Self-hosted AI agent orchestration service controlled via Telegram, Discord, or Slack.

First principles for multiplying your productivity with AI agents like Claude Code. Free. Star for updates

Efficient Large-Scale Fleet Management via Multi-Agent Deep Reinforcement Learning

💬 UI Building Blocks for LLM/Chat UIs. Maintained by @goodcodeus.

Ultra-minimal personal AI agent: starts small, self-modifies its code live, adapts by writing exactly the code & features you need

The Pi coding agent, built into VS Code.

独立、可自托管的 AI 动态心智状态引擎：驱动力、念头池、疲惫、睡眠与意图。

Inspect Codex / Claude logs, verify metering samples and re-count supported billing exports. Free local tools; bounded $10 diagnosis and $20 QA.

An AI agent for running, testing, and hardening GitHub Docker Compose apps.一个用于运行、测试和加固 GitHub Docker Compose 应用的 AI Agent。

DeepSeek Harness plugin for Android — build, run, and interact with a live emulator or USB device stream inside a conversation, driven entirely through adb.

一个面向全网研究与复杂数据分析的云端智能体，能在公开网络、YouTube、BiliBili、X、小红书、Hacker News上研究任何主题 。项目wiki：https://zread.ai/OWWZO/ai-agent

公文写作 AI Agent Skills：懂规矩、懂单位、懂角色；不编造、守保密、帮你做减法。覆盖 22 种文书类型，支持 Claude Code、Codex、Cursor 等主流工具，防编造、懂国标格式、适配机关事业单位和医院高校国企。

Ultra-minimal personal AI agent: starts small, self-modifies its code live, adapts by writing exactly the code & features you need

AI-powered whale trade surveillance for Polymarket — real-time monitoring of 700+ markets, 14 autonomous research tools, 7-step deep analysis with information asymmetry scoring.

An implementation of a computer use agent (CUA) using LangGraph

Code repository for SMART-LLM: Smart Multi-Agent Robot Task Planning using Large Language Models

89 skills and 38 specialized agents that enforce proven engineering practices for AI-assisted development. TDD, systematic debugging, parallel code review, and 10-gate development cycles — as a Claude Code plugin marketplace.

Chiasmus is an MCP server that gives language models access to formal verification

Curated high-quality AI Agent Skills. Search, install, copy and share. Works with Claude Code, Cursor, OpenClaw, and other AI coding tools.

The open-source company brain. Run your entire company with AI agents, skills, and a self-improving context.

智能体（Agentic AI）最前沿技术的深入剖析。| Latest techniques in agentic AI area

Local personal memory for LLM agents

运行时自扩展工具AI Self-Evolving AI Agent Framework | Decision-Loop Driven Microkernel Architecture with Hot-Pluggable Components & Triple-Layer Memory | Agent Infinite Evolution Engine that Learns from Failures

AI job agent: scans 130+ careers pages nightly, scores every role against your resume with an LLM (0–100), alerts you on Telegram, and drafts tailored cover letters + resumes. Free & open source.

Agentic OS: Multi-agent orchestration platform for opencode, Hermes & Antigravity CLI. Skills hub, scheduler, cost analytics, memory & backup.

Local, open-source (Apache 2.0) voice layer for Claude Code and Codex on macOS that reads your coding agent's progress aloud. Kokoro text-to-speech on your Mac, no account or API keys: hear a short spoken summary when your agent finishes or fails.