Custom agents

Blueprint for an always-on remote agent workspace: code on your Mac, run the work on an always-on Linux box (Herdr + Syncthing + Cloudflare Tunnel + Caddy).

Multi-agent adversarial code review for any coding agent

A benchmark and generation framework for malicious agent skills.

BASE is an open framework and proposed standard for shaping AI through conversation—turning the way you work into a durable, portable method you control across tools and models.

Registry-as-ground-truth verification & autonomous research framework for agent-driven development.

One URL in, an agency-grade launch video out. A Claude Code skill (Remotion + React) with a 16-law taste codex — 12 brands, 12 worlds, zero templates.

My Claude stuff (skills, agents and whatnot.)

Self-contained, Dockerized offensive security challenges for evaluating AI-powered penetration testing agents. Covers modern tech stacks (Node.js, Python, Go, Java, PHP, Ruby) across diverse vulnerability classes and target environments from basic injection to multi-step exploit chains, sandbox escapes, and defense-enabled environments.

Simple note taking, powerful search, AI ready

A fast, local, indexed code search tool optimized for AI coding assistants. Written in Rust using Tantivy for full-text indexing.

Resume and hand off Claude Code, Codex & Gemini CLI sessions between agents: one terminal picker for all your local coding-agent history (also Antigravity, OpenCode, Pi; MCP server).

Know what AI is actually running in your clusters. An unprivileged Kubernetes controller that inventories models, runtimes, agents, and RAG components at runtime - emitting CycloneDX 1.6 ML-BOMs with evidence, confidence tiers, and Sigstore-verified model signatures

🛡️ A curated list of tools, frameworks, standards, and resources for AI agent governance, safety, and compliance

🔥 Quo Vadis, World Modeling? Towards Interactive World Proxies for Continually Improving Agents

Context meters, quota tracking, and provider rate limits for agents running in Herdr.

Claude Code skill: distill a book or video series into one interactive single-file HTML · 书籍/视频深度蒸馏

Self-hosted voice assistant. Speak to your computer: it answers aloud, types your dictation, and operates your apps.

DeepSeek Harness 原生桌面版，基于 Tauri 2，内置 Node.js 运行时，一键安装，极速启动，托盘常驻，自动更新。DeepSeek Harness native desktop version: built on Tauri 2 with a bundled Node.js runtime.

The living DeepSeek Harness plugin directory — refreshed hourly, compat-tested daily, with an in-app plugin store and scaffolder. DSH 插件活目录：每小时刷新，每日兼容实测，内置插件商店与脚手架。

Benchmark for comparing agent harnesses on everyday online tasks — fixes the base model, varies the harness. Sister project of ClawBench, same scoring pipeline.

DairyBench: A Long-Horizon Multi-Agent Benchmark for Business Decision-Making

Community-maintained reference guides and megathreads for Hermes Agent.

墨枢是一个面向中文长篇网文作者的 AI 小说写作工作台，支持作品大纲、章纲、人物与世界观设定库、时间线、RAG 检索、一致性守卫、自然化审查、AI 辅助续写、团队协作、用量计费，以及面向漫剧改编的静态分镜管理。前端采用 Vue 3，后端采用 FastAPI，支持 PostgreSQL/pgvector、Redis、Docker 自部署和 Cloudflare Workers 预览。

General, reusable Flutter engineering skills for AI coding agents — architecture, Riverpod 3.x, testing, persistence, i18n/RTL, accessibility, navigation & more. Agent Skills open standard; works with Claude Code, Cursor, Codex & 70+ agents.

A practical guide to DSH plugin quality: ENTRY checks, FLOW journeys, and BOOST practices. Community guide; not a certification or ranking.

Open-source AI desktop assistant — run local or remote LLMs with privacy-first design and zero setup required.

A beautiful home for your Hermes agents. Chat with them one at a time or all together, connect your apps, and reach them from the browser, your Mac or PC, your phone, WhatsApp and Telegram.

[NeurIPS 2025] ParamMute (a.k.a. PIP-KAG): suppressing knowledge-critical FFNs for faithful retrieval-augmented generation (RAG knowledge conflicts)

Run AI-generated logic safely inside your app. Stof is JSON with functions: data and logic in one document, sandboxed, in Rust, JavaScript, and Python.

Admin foundation on FastAPI + React 19 + shadcn/ui — row-level permissions, multi-tab that keeps state, approval workflow, permission-bound AI assistant, open API, S3 resumable uploads; PostgreSQL / SQL Server / MySQL

Splunk App for Data Science and Deep Learning - container images repository

Discord bot - Friend Time helps your server coordinate times and events by converting times mentioned in chat between time zones!

the source code of RAG course at https://time.geekbang.org/column/intro/100817901

AI-powered mobile automation with Model Context Protocol (MCP) integration. Seamlessly control Android & iOS devices through Appium with intelligent visual element detection and recovery. Built for AI agents like Claude to perform complex mobile testing workflows.

A minimal self-replicating agent based on GPT-3.5/4

Rust Agent Development Kit

Run OpenClaw (MoltBot, ClawdBot) Safely with Docker: A Practical Guide for Beginners

Your personal AI operating system

Make AI Free Again - Shard is a GUI for Ollama LLM's. Powered by Next.js.

Private RAG app sample using Llama3, Ollama and PostgreSQL

A comprehensive, free MCP server designed specifically for financial analysis and quantitative trading. This specialized platform offers one-click local deployment with a sophisticated department-based architecture that mirrors real financial company operations.

Empowering AI agents to dominate DeFAI

Official source code for the AI Agent Foundations: Codebase Analyst course. The main branch contains the final solution, starter provides the boilerplate for practice, and lesson-X-complete tags track the code at the end of each lesson. Use the 'Issues' tab for any course-related questions.

Build fully sentient, unruggable AI agents

An AI HR Agent who lives in Slack (GPT-powered)

DocAider is an LLM-powered repository agent designed to support developers in efficiently generating documentation through a multi-agent approach. With its integrated workflows, docAider streamlines the documentation update process, making it both effective and user-friendly.

Fine-grained authorization for AI agents using OpenFGA.

Your own self-evolving AI agent powered by Elixir/OTP

This project is to integrate Vercel AI-SDK-UI (frontend for chat interfaces) with Agno agentic python backend. This integration shows how to build realtime chat AI agentic apps using Agno and Vercel.

The startkit remix template with ai elements, supports ACP agents and vercel gateway

MCP Security Solution for Agentic AI — real-time proxying, behavior analysis, and malicious tool detection

Production-grade AI agent infrastructure with resource constraints, ethical governance, and cryptographic integrity (private implementation code)

A security-first MCP server empowering AI agents to orchestrate Ghidra, Radare2, and YARA for automated reverse engineering.

Let AI agents design with you.

Official code for the paper: Embodied Multi-Modal Agent trained by an LLM from a Parallel TextWorld

Lightweight multi-agent task management that promotes agentic knowledge capture

API-first multi-agent coding assistant with CodeAct tool calling and first-class terminal support.

Claude Code configs for the expert Solana builder. CLAUDE.md, agents, commands, hooks, rules, skills and settings across Web, Anchor, Pinnochio, Unity, Mobile, security and more.

Efforts toward giving Qwen 3 Coder 30B A3B proper agentic tool calling capabilities at or near 100% reliability.

An example extension in go using function calling and confirmation dialogs

Another curated list of Python frameworks

A personal AI OS for macOS, powered by Claude.

The multi-agent work platform for local-first execution.

Open Source React Chat Widget. Ready for use and can be connected to any backend like Chatbot/NLP/Live Chat engine or messenger.

An agentic macOS desktop tool

I specifically cover the following topics: Java primitive data types, declaration statements, expression statements, importing class libraries, excepting user input, checking for valid input, catching errors in input, math functions, if statement, relational operators, logical operators, ternary operator, switch statement, and looping. How class variables differ from local variables, Java Exception handling, the difference between run time and checked exceptions, Arrays, and UML Diagrams. Monsters gameboard, Java collection classes, Java ArrayLists, Linked Lists, manipulating Strings and StringBuilders, Polymorphism, Inheritance, Protected, Final, Instanceof, interfaces, abstract classes, abstract methods. You need interfaces and abstract classes because Java doesn't allow you to inherit from more than one other class. Java threads, Regular Expressions, Graphical User Interfaces (GUI) using Java Swing and its components, GUI Event Handling, ChangeListener, JOptionPane, combo boxes, list boxes, JLists, DefaultListModel, using JScrollpane with JList, JSpinner, JTree, Flow, Border, and Box Layout Managers. Created a calculator layout with Java Swing's GridLayout, GridBagLayout, GridBagConstraints, Font, and Insets. JLabel, JTextField, JComboBox, JSpinner, JSlider, JRadioButton, ButtonGroup, JCheckBox, JTextArea, JScrollPane, ChangeListener, pack, create and delete files and directories. How to pull lists of files from directories and manipulate them, write to and read character streams from files. PrintWriter, BufferedWriter, FileWriter, BufferedReader, FileReader, common file exceptions Binary Streams - DataOutputStream, FileOutputStream, BufferedOutputStream, all of the reading and writing primitive type methods, setup Java JDBC in Eclipse, connect to a MySQL database, query it and get the results of a query. JTables, JEditorPane Swing component. HyperlinkEvent and HyperlinkListener. Java JApplet, Java Servlets with Tomcat, GET and POST methods, Java Server

Debuggable runtime for AI agent workflows. DAG pipelines, artifact lineage, and replayable runs.

Subdosec is a fast, accurate subdomain takeover scanner with no false positives. It also offers a database of sites vulnerable to subdomain takeover (public results), along with detailed metadata like IP, CNAME, TITLE, and STATUS CODE for reconnaissance to identify potential new vulnerabilities.

Agent Relay Protocol — Stateless WebSocket relay for autonomous agent communication

Distributed model predictive control for multi-agent point-to-point transitions.

A multi-agent reinforcement learning framework for optimizing coverage and connectivity in Space-Air-Ground integrated networks. This project simulates and trains intelligent agents to coordinate satellites, UAVs, and ground stations for efficient network deployment.

Agno-Go: A High-Performance Multi-Agent System Framework Based on Golang. Inheriting the Agno design philosophy, it leverages Golang's concurrency model and per

Turn any MCP stdio server into a production HTTP API: per-user session pooling, timeout survival via background jobs, fail-closed auth, SSE/webhook push, live dashboard. Python 3.11+ · FastAPI · Redis

Open AI Chat Bot in the Menu Bar: ChatGPT desktop app for Windows, Mac, and Linux

Dabarqus is incredibly fast RAG that runs everywhere.

Claude Code skill for using Codex CLI as an execution specialist for implementation-heavy coding work, multi-file refactors, and wrapper-based execution flows.

A Claude Code plugin that optimizes documentation for AI coding assistants like Claude, GitHub Copilot, and other LLMs. Makes your docs more effective through c7score optimization, llms.txt generation, question-driven restructuring, and automated quality scoring.

The Python Harness for Production AI Multi-Agent Systems

Open WebUI tool port of Superpowers by Jesse Vincent — brainstorm → spec → plan → execute agentic dev workflow for local LLMs

An AI-powered Virtual YouTuber (Vtuber) utilizing Google's Gemini language model to create engaging, personalized, and context-aware interactions. This project explores the potential of AI in human-computer interaction and virtual content creation.

Ruby & Rails agentic coding assistant powered by Claude. 28 tools, 112 skills, 16-layer architecture.

MCP server for AI agents to read, edit, inspect, and validate local HWPX documents.

Temodar Agent is an AI-powered WordPress plugin and theme security analysis platform built for security researchers, product security teams, auditors, and defenders. It combines AI agent workflows, multi-provider LLM orchestration, Semgrep-powered static analysis, and risk-based WordPress reconnaissance in one local-first Docker application.

Sample .NET app using Microsoft Agent Framework to showcase collaborative AI agents (e.g. writer + editor) with web & console interfaces.

A Model Context Protocol (MCP) server that provides context optimization tools for AI coding assistants including GitHub Copilot, Cursor AI, Claude Desktop, and other MCP-compatible assistants enabling them to extract targeted information rather than processing large terminal outputs and files wasting their context.

Persistent agents for Claude Code as a plugin, not a harness. Memory, personality, messaging across WhatsApp, Telegram, and Discord, plus a service mode for 24/7 runs. Imports from OpenClaw.

Image and video processing shaders for ReShade.

Last9 MCP Server

Domain-Aware Neurosymbolic Agent (Dana), an agent-native programming language

STAgent is a multimodal LLM-based AI agent that enables deep research about spatial transcriptomics data

ChatEDA: A Large Language Model Powered Autonomous Agent for EDA (TCAD'24, NAACL'25)

AI-native test management for QA engineers, SREs, and DevOps teams. From unit tests to chaos experiments: unified in Markdown, validated in SQL. Built for regulated industries where accountability, traceability, and structured evidence are essential.

Full Claude agent, completely contained, one command.

MCP server for bioinformatics agent skills — query a knowledge graph of 78 bioinformatics workflows directly from Claude Code

MCP server for Claude Code × Kimi Code CLI 2.4.1 — default model k3 (up to ~1M on eligible plans); kimi-for-coding = K2.8 Preview. Delegate bulk reads to Kimi, cut Claude-side tokens.

a minimal, Rust-based coding agent focused at CI workflows and small codemods, similar to Claude Code

Source code graph RAG (GraphRAG) for C/C++ development based on clang/clangd

Sync MCP servers, Skills, Agents & Commands across Claude Code, Cursor, OpenCode, Codex. One config, all tools.

A unified framework for building agentic RL environments.