Custom agents

OpenART is an open-source framework designed to evaluate the safety and robustness of autonomous AI agents in dynamic, long-horizon, and stateful environments. It stress-tests agent runtimes against multi-step state poisoning, privilege escalation, and tool-use vulnerabilities across 10,000+ benchmark scenarios.

Open-source, license-free MCP server for RTL waveform debug: reads FST waveforms (VCD/FSDB auto-convert) plus a SystemVerilog netlist, with 34 tools covering driver analysis, value/X tracing, pass-fail waveform diff, a browser wave viewer the agent drives, and pre-simulation static analysis.

An mcp connector to evaluate anything fast and cheap. Give your AI agent direct access to typesafe ai's jev model.

大 A免费开源AI插件FQGate-agent（原插件名 tonghuasun-agent）：为 Codex、Claude Code、DeepSeek 等 AI 工具提供本机 A 股实时行情、K 线、Level2、资讯、账户查询与可选交易能力。

A workspace and framework to build domain-specific systems for agentic knowledge work.

AI lifecycle platform where engineers and agents track experiments, train models, and ship to production.

PYA (Protect Your Ass): security audit plugin for Claude Code. 9 specialist agents, a dated threat feed, verified findings, offline by default.

EveryInfra MCP server & agent plugin: scraper API for TikTok, Instagram, Amazon, YouTube, Xiaohongshu & Douyin. 573 data APIs across 106 platforms, from $0.56/1K calls. For Claude Code, Cursor, Codex.

ArmorIQ intent enforcement plugin for OpenClaw AI agents

Browser-local AI code Q&A engine. Chat with your codebase, ensure privacy, and optimize LLM collaboration.

Code/Data for the paper: "LLaVAR: Enhanced Visual Instruction Tuning for Text-Rich Image Understanding"

An engineering team in a box for Claude Code — 12 specialized agents, 15 automation hooks, and the P7/P9/P10 methodology.

AI Agent: Open-source headless CMS built with ASP.NET Core (C#) and React, featuring REST APIs, GraphQL, and a GrapesJS page designer.

Ticket-Driven Automated Software Engineering. OpenASE is an all-in-one platform that turns tickets into working code — AI agents automatically pick up tickets, execute workflows on your machines, and deliver results with full traceability.

A modular framework for building intelligent AI agents powered by Claude, with rule-based workflows, skill management, and clean JS, TS architecture.

Repo to accompany Jon Krohn and Ed Donner's workshop on Agentic AI in Action at ODSC East

Organization-as-Code for autonomous AI agents. Brain-inspired memory that grows, consolidates, and forgets. Multi-model (Claude/Codex/Gemini/Cursor/Ollama).

A fast, native browser automation CLI built from the ground up for AI agents, powered by Chrome DevTools Protocol. 63 commands covering navigation, interaction, screenshots, accessibility, network mocking, visual diffing, test generation, and more — all from a single binary.

基于ragflow二次开发的后台管理系统，可以独立运行，支持批量管理知识库、聊天、智能体、用户等，解决ragflow在后台管理上的痛点。

Agent-readable docs, verified examples, and skill manifest for @zapier/zapier-sdk

中文 Claude Desktop / Claude Code 桌面客户端：中文界面、本地 Code 工作区、GitHub 导入、权限守卫和 Artifact 预览。

Extract design tokens (colors, typography, spacing, border radius, shadows) from any public website. Generates JSON and CSS custom properties for local projects. Available as an AI agent skill (Claude, Cursor, Codex) and standalone CLI.

The Fastest AI Memory Model - Your Second Brain

A local-first, long-term memory for anything you'd otherwise forget.

Jev DSH 决策引擎｜面向 Agent Harness 的结构化决策插件。原生支持 DeepSeek Harness，通过 iPolloWork 支持 OpenCode、Codex Harness。

[Public preview] Externally scored agentic ML research benchmark: 60 tasks, real competition ground truth. Open protocol, operated evaluation.

MCP server that lets AI agents run maigret, an OSINT tool that searches a username across thousands of sites

Self-hosted browser workspace for Claude Code and Codex CLI, with Docker sessions, terminal, files, Git, and usage tracking.

A General AI Agent That Thinks and Acts

An open-source Vibe platform similar to Claude Cowork / Manus / Openclaw, with professional rich image document generation.

A Collection of Multi-Agent Reinforcement Learning (MARL) Resources

AI Copilot for Vim/NeoVim

Knowledge Management for Humans and Agents

Asks what to optimise and how to measure it, then runs experiments on its own branch.

Open-source memory runtime for AI agents — reproducible, provenance-tagged context bundles instead of query-time retrieval. Apache-2.0, self-hosted on Postgres + pgvector, Python + TypeScript SDKs.

System audio capture + multi-provider ASR + local-first AI review workspace. Floating live captions, 12 ASR backends, 60+ languages, AI summary/chat/mindmap, Open API, MCP server, and Agent Skill.

Agent skills for Qdrant vector search: scaling, performance optimization, search quality, monitoring, deployment, model migration, version upgrades, and SDK usage across Python, TypeScript, Rust, Go, .NET, Java

Let your agents read, send, and react to email autonomously, without human involvement

A modular multi-agent framework for video-based pain localisation and adaptive exercise recommendation. This prototype implements the core elements from the OpenRehabAgent paper, including pose synthesis, heuristic pain estimation, a Q-learning agent, safety rules, and feedback tracking

distill any book down to its spine

Put the world's smartest AI agents & plugins in your pocket

Auto Research with UI. Autonomous Generalist Scientist / AI Scientist / Agent Scientist / Robot Scientist, across all Scientific Fields.

Multi Agent Reinforcement Learning using MalmÖ

A browser coding agent interface for selecting elements and sending instructions directly to Claude Code.

Directory of OpenAPI specifications for thousands of public APIs. Every API is machine-readable and scored for AI agents. CC0.

专为CTF设计的可扩展AI Agent，可自动解CTF题，也能与用户协作交互解题~

BlacksmithAI is an OPEN-SOURCE advanced penetration testing framework that leverages multiple AI agents to automate security assessments. The system orchestrates specialized agents through a complete penetration testing lifecycle, from reconnaissance to post-exploitation using professional security tools in a controlled environment.

Give your AI agent a 3D body: text or a photo to a rigged GLB avatar, an LLM brain, a Solana wallet and a one-tag embed. Keyless MCP server for Claude, Cursor and Codex. 50k+ models forged.

Evidence-chain whiteboard + MCP server for AI research agents: every claim traces back to the run and code that produced it. 讓 AI agent 做研究時，每個結論都能追到產生它的程式碼。Claude Code · Codex · Cursor · Gemini CLI

A package for parsing PDFs and analyzing their content using LLMs.

DB-GPT WebUI，LLM to vision.

structural variant calling and genotyping with existing tools, but, smoothly.

:loudspeaker: Ptt 文章通知機器人！Notify Ptt Article in Realtime

Run a team of local coding agents in your terminal. Launch multiple Claude Code, Codex or Gemini agents, switch between them, review their changes and add comments, and push PRs all from one streamlined terminal UI. An exercise to push how fast development can happen with multiple parallel agents. Mostly vibe-coded

Codebase for paper: RoCo: Dialectic Multi-Robot Collaboration with Large Language Models

[NAACL 2025] KnowAgent: Knowledge-Augmented Planning for LLM-Based Agents

Official Implementation of MultiWorld: Scalable Multi-Agent Multi-View Video World Models

🎼 Turn Claude Code into a production powerhouse. 33+ AI agents automate bug fixing, security scanning, and dependency management. 19 slash commands, 6 MCP configs (600-5000 tokens), quality gates, and health monitoring. Ship faster, ship safer, ship smarter.

Fixing GRPO training collapse in long-horizon multi-tool agents. A lightweight PRM-Lite + LATA joint approach achieves +37% over vanilla GRPO on τ-bench airline (50-task, multi-turn).

More than just Karpathy’s LLM Wiki, 100% local with Ollama. Drop Markdown notes → AI extracts concepts → your Obsidian wiki auto-links and grows. Zero sharing. Your notes stay yours.

UmaDev: A coding agent that works like a real dev team, commanding the Claude Code / Codex / OpenCode you already use.

Recursive Experiential–Working Memory Evolution for Long-Horizon Agent Harnesses

Validate Skill content against Agent Skill specification, with additional content density and quality checks.

Free Domain Name For Everyone

App Desktop integration Zalo, Facebook & Telegram personal for Multi-Account management, CRM, Marketing, POS, ERP, Workflow and AI Assistant integration. Built on top of zca-js, fbchat-v2, telegram-api.

Harness is an AI Agent development guardrail Meta-Skill that establishes four layers of defense for any project in one command: knowledge management, architecture constraints, feedback loops, and entropy management.

Local LLM Powered Recursive Search & Smart Knowledge Explorer

Generate "Rules for AI". Quickly ✨

A distributed Spark/Scala implementation of the isolation forest and extended isolation forest algorithms for unsupervised outlier detection, featuring support for scalable training and ONNX export for easy cross-platform inference.

A library of shared system prompts for creating customized educational GPT agents.

A comprehensive statusline dashboard for Claude Code — session info, quota bars, agent tracker, MCP health, message history, and more.

tmux-native dashboard for managing Claude Code agents

A Site Reliability Engineer AI agent that can monitor application and infrastructure logs, diagnose issues, and report on diagnostics.

SEOBuild Onpage - The first AI agent that writes pages Google ranks AND LLMs cite. One command in, ranking page out. Built on DeerFlow, powered by 2026 SEO + GEO strategies tested / working. Forensic competitive analysis, 500-token chunk architecture, entity consensus, verification tags. BYOK GSC, DataforSEO. Works w/ OpenClaw, Claude Code, Codex

Realistic examples of building evals and optimizing agents with Harbor

🧬 Symbio（共生）- AI Infra 级多 Agent 协同框架 | 33 Killer Features: 动态 DAG, 本体化记忆, 防过早完成, 数据飞轮, 神经符号安全, OpenTelemetry 可观测, HITL 人类介入, MCP 原生支持, 语义缓存, PromptOps, 边缘计算, 隐私计算 | Next-Gen AI Infrastructure for Multi-Agent Orchestration

Adaptive Python web scraping toolkit + MCP server for AI agents. Self-healing selectors that survive site changes, TLS-fingerprint stealth to bypass anti-bot filters, CSS/XPath parsing, and 24 built-in scrapers, clean, structured, LLM-ready data from any URL.

Toward Autonomous Scientific Discovery

LumiCore local-first personal AI runtime and desktop agent platform

Slack CLI for humans and AI agents. Read, send, search and reply across one or many Slack workspaces from the terminal. No Slack app needed.

Cross-platform remote desktop for macOS and Windows with hardware acceleration, multi-monitor support, clipboard transfer for text, images, files and folders, and MCP integration for AI agents to connect to and control remote computers.

面向开发者与 AI Agent 的开源流量调试工作台，让业务异常可筛选、调试修改可追溯、排查现场可分享。

一个功能完整的微信公众号内容创作与管理平台，支持文章采集、AI改写润色、风格分析、微信发布等全流程。 ##

[ICML-2026] Official implementation of "SEAgent: Self-Evolving Computer Use Agent with Autonomous Learning from Experience"

Install, manage, and share skills across every major coding agent — Claude Code, Cursor, Copilot, and more.

Bring professional data and tools to the AI you already use: open-source MCP, CLI, SDKs, skills, and docs to find services, call them, and audit usage.

Automate Ghidra reverse engineering from the command line — headless analysis, decompilation, and structured JSON output for AI agents like Claude Code

The AI agent memory cloud you’ll wish existed yesterday. Manage agent memory like a GitHub repo: switch devices, switch agents, switch sessions, and your AI still remembers who you are, what it has learned, the rules you taught it, where the project left off, and what to do next.

Official code for the paper "Memora: A Harmonic Memory Representation Balancing Abstraction and Specificity"

本地优先的 AI 投资研究工作台（AI investment research workbench），支持深度调研、证据化投资论点、投资组合风险分析、自选股与 Pi Agent 副驾驶。

Gugu (咕咕): personal assistant, chat and device client — source release

Harnessclaw is a powerful, Electron-based desktop application designed to manage, chat with, and operate AI agents and skills seamlessly.

Use Reinforcement Learning to train an autonomous driving agent in CARLA Simulator

800+ pure-markdown skills for autonomous AI research. Non-linear orchestration with backtracking, 4-layer military hierarchy (Campaign → Strategy → Tactic → SOP), 5 MCP integrations. The AI is the researcher — you set the direction.

Empower Your Dev Ecosystem with AI Agents, Automate with Your AI Copilot, Elevate Your Code.

Finetuning of Falcon-7B LLM using QLoRA on Mental Health Conversational Dataset

a chatbot based on sklearn where you can give a symptom and it will ask you questions and will tell you the details and give some advice.

A textual adventure game

Kane CLI by TestMu AI (Formerly LambdaTest)

CLI tool for Linear.app with JSON output, smart ID resolution, and optimized GraphQL queries. Designed for LLM agents and humans who prefer structured data.