Custom agents

A curated list of papers and resources based on the survey "Self-Evolving Agents as Dynamic Graph Transformation: A Survey and New Perspective"

Fox in the Box: One-click Hermes Install — self-hosted AI assistant platform

Daily Leetcode masterclass with Claude code DSA mentor

Reusable Claude Code skills for AI startup operators. Open source, MIT licensed.

🧭 给 Claude Code / Cursor / Codex 用的『架构副驾』skill —— 开新项目时用持续深度提问引导你在写代码前想清楚架构(产出架构图 / ADR / 演进路线)。知识源自 awesome-architecture。中英双语。

Find and redact secrets in AI coding agent histories (Claude Code, and more).

AVL Code 是安天(Antiy)澜砥团队推出的国产 AI 智能编程与安全分析桌面助手,Windows / macOS / Linux 全平台原生分发。它把对话式 AI 与工程师真实工作台合二为一:读写代码、执行命令、研判恶意样本。

AI桌面人形机器人 — ESP32-S3全栈运行，不会开口说话，22种情绪+关系成长+6路舵机AI控制，全栈开源(CC BY-NC-SA 4.0)

Comunidade brasileira de Hermes Agent (Nous Research) — skills, guias e integrações em português.

P2P Encrypted Messaging for Agents and People

A curated guide & directory for open-source AI devtools, local LLMs, autonomous agents, and SaaS alternatives. Built for modern AI developers.

AI agent context layer for career-related tasks.

RepoMind is an AI-powered documentation generator for modern codebases that helps developers understand any GitHub repository faster. It analyzes project architecture, folder structures, dependencies, APIs, and core logic to generate clean, intelligent documentation with contextual insights. Built with Next.js, Octokit, RAG pipelines, Pinecone.

81 production-ready subagents for Gemini CLI and Antigravity CLI: code review, security, testing, DevOps, data/AI and language experts. One-line install, validated in CI.

A framework for building intelligent agents

星⭐收藏家：Bilibili 收藏视频 Markdown 多 Agent 本地编排工作台

58 个跨境电商 · DTC 品牌出海 AI 技能库:Amazon 运营、SEO/GEO、广告归因、财务、联盟、红人、线下零售、VOC、众筹。事实均标来源与日期,内置证据闸——数据不足时说「数据不足」,而不是编一个数字。| 58 skills for cross-border e-commerce & DTC: sourced, dated, with evidence gates. EN/中文. Works with Claude Code & Antigravity.

small coding agent in go

Simple editor supports agentic AI.

A companion for coding agents to help you understand large outputs

Your high-ticket creator business, encoded. A team of agents running every department - foundations, marketing, nurture, sales, launch, scale, partnerships. Runtime agnostic.

Project-local Codex workspace kit with reusable skills, agents, hooks, plans, docs, and helper scripts for multi-repository development.

基于多模态大模型（VLM）与 Agent 架构的医疗辅助系统，针对医疗场景下复杂单据解析难、医学逻辑容易产生幻觉的核心痛点，提供体检报告解析、生化指标趋势分析、多专科智能分诊等服务。

The Economical Coding Agent

数模Agent（数模 Agent）/ 数学建模 AI 助手：多智能体协作完成题意分析、文献检索、建模、代码执行与论文写作，支持人工审核和断点恢复。Local-first multi-agent workbench for mathematical modeling.

A minimal local coding agent in Go — a faithful port of Sebastian Raschka's mini-coding-agent, with no LLM framework.

倪海厦skill · 经方中医AI 将经方大师倪海厦的完整中医思维体系注入 AI Agent

Official RollingGo Hotel Booking MCP Server. 2M+ hotels at wholesale rates, 40+ LLM compatible. Free unlimited calls for businesses & individual devs. Filter by location, date, star grade, guests & tags; pull real-time room types, pricing & cancellation rules.

sci-box 提供一组可直接调用的科研绘图复刻模板，帮助 Agent 快速生成高质量图表。sci-box provides a set of directly callable scientific drawing replication templates to help Agents quickly generate high-quality charts.

An open-source, agent-guided creative-direction workflow for turning one visual reference into a connected brand system with Lovart Agent.

终端里的 AI 编程助手。基于 Python + Textual 构建，提供交互式 TUI、非交互式单次执行、浏览器远程模式三种前端，并内置多 Agent 团队协作能力。架构上参考 Pi 的分层设计，取向是稳定核心 + 可扩展 Runtime。

Akasha AI is an open-source, AI-native knowledge operating system for humans and AI agents, enabling organizational knowledge to continuously accumulate and evolve.

Myna Recorder — a Windows macro recorder you can read and edit. Image anchors, an auto clicker, an AI assistant that writes macros, Agent mode (computer use, for short jobs) and a screen watchdog.

Jev-powered semantic code search for coding agents — find behavior across repositories via CLI or MCP, with exact source excerpts and line numbers.

Zoomies is a lightweight, self-hosted GitHub Actions runner fleet controller. Ephemeral runners by default, GitHub App auth, webhook-driven autoscaling, multi-host agents, a polished live web UI, and a one-line interactive installer. Single Go binary, SQLite, no Kubernetes.

YouTube and YouTube Shorts marketing skills for Claude Code and Codex: high-CTR titles, SEO descriptions, retention hooks, thumbnails, and a content plan. Publish via Publora. MIT. Content engineering by Creative Content Crafts.

Connectome agent host runtime — runs Connectome agents, MCPLs, and the context-manager stack

Tesseract by Mirage — CLI releases and agent skills for creating and editing videos.

Myna Recorder — a Windows macro recorder you can read and edit. Image anchors, an auto clicker, an AI assistant that writes macros, Agent mode (computer use, for short jobs) and a screen watchdog.

Open-source research assistant and MCP audit toolkit for systematic reviews: cited research reports, PRISMA tracking, evidence and claim checks, and reproducible exports with human oversight.

从 700+ 篇文章、80 场直播中提炼出的 A 股判断规则体系（48 章，含数字阈值与行情回测）

Ship agents you can audit.

agent-marketplace is a command-line tool that installs shared AI agent resources into the AI coding tools you use

Workflow builder for coding agents 🐶

Make Claudia content with AI agents - the toolkit behind the Claudia Grok Bot

85+ buildable AI engineering portfolio projects for 2026, grouped by theme (RAG, agents, evals, fine-tuning, multimodal…). Free, MIT. By Landed.

An open-source Roblox-like engine to easily create 3D multi-player environments and games for embodied agents. Implemented in Rust with Three.js rendering.

Hardware mouse MCP server for AI agents on Windows x64. USB HID control and voice dictation; requires compatible uMouse hardware and firmware. 面向 AI 智能体的硬件鼠标 MCP：USB HID 控制与语音听写，需要兼容的 uMouse 设备与固件。

VeriWeb: Verifiable Long-Chain Web Benchmark for Agentic Information-Seeking

The open source chat powered by LLMs with RAG. Kollektiv makes it easy to sync your custom data sources and get accurate, contextual replies.

🤖️ 聊天机器人——夫子的「自然语言理解」模块

Pandora is an Agent-Based Modelling framework for large-scale simulations

Queue-Based State Machine - A lightweight workflow execution engine with DAG-based stage orchestration. Unlike simple task queues (like Celery) or advanced orchestrators (like Highway), Stabilize strikes a balance specifically optimized for high-throughput, stateful DAG execution, making it highly suitable for coordinating autonomous AI agents

🚀 Next-generation AI code assistant with Multi Agent methodology

Dialog State Tracking Challenge 2 & 3 Data

Materials for the Call for Code 2020 solution starter kit for crisis communication in the context of COVID-19.

Connect Chatwoot to Botpress

Maiar: A Composable, Plugin-Based AI Agent Framework

Guardrails to support any-agent

Quantitative trading research infrastructure for AI Agents - backtest, sweep, radar, deploy

O-GIA is an umbrella for research, infrastructure and projects ecosystem that should provide open source, reproducible datasets, models, applications & safety tools for Open Generalist Interactive Agents (O-GIA). O-GIA systems will act in collaboration with human or autonomously, supporting various kind of validated decision making and assistance.

Micky podcast agentic engineering workflow bundle

Modular Agent Based Evolution Framework

Smart contracts for running Agents and Services on the SingularityNET open network.

The proxy IP pool, crawling the mainstream free agent, automatically re -processing, automatically testing the availability of the agent, and the common request header has been attached

Ansible playbook to automatically deploy the Server Density agent, and start monitoring. https://www.serverdensity.com

A collection of agent skills that can be used to create WordPress themes and sites

别读评论区了，让你的 Agent 去和专家吵一架。

A Python crawler tool that can automatically simulate browser operations to crawl all users' tweet content and save all static resources (videos, pictures) locally without calling the Twitter API. Use the selenium library.

Your personal AI copilot that lives where you do -- browser, terminal, your messaging apps, or a phone call.

Welcome to AI Agents Handbook — your all-in-one guide for building AI agents, from basics to advanced techniques. Explore code, tutorials, and resources for beginners to experts!

SpeechAgents: Human-Communication Simulation with Multi-Modal Multi-Agent Systems

Chinese version implementation of Generative Agents: Interactive Simulacra of Human Behavior

Learn to build multi-agent systems with Google's Agent Development Kit (ADK). This repository offers step-by-step guides and examples using the open-source Python ADK framework

P4wnP1 - WiFi covert channel - Client agent (experimental Proof of Concept) by MaMe82

[NeurIPS 2024] GenRL: Multimodal-foundation world models enable grounding language and video prompts into embodied domains, by turning them into sequences of latent world model states. Latent state sequences can be decoded using the decoder of the model, allowing visualization of the expected behavior, before training the agent to execute it.

Demonstrate calling a kernel function and handle process creation callback against HVCI

🧠 High-performance persistent memory system for Model Context Protocol (MCP) powered by libSQL. Features vector search, semantic knowledge storage, and efficient relationship management - perfect for AI agents and knowledge graph applications.

AI-powered project management framework based on an opinionated view on effective prompts and a highly modular approach to building effective agents, workflows, templates, prompts and context documents.

Agentic GitHub Guardrails - Replace static protection rules with context-aware enforcement.

Aquiva's Agentforce blueprint — installable as 2GP Unlocked package. Your org's butler: ask it anything, tell it to handle things while you're in a meeting, and it'll notify you when done. It knows your data, your schema, your files, and remembers how you like things.

On-device AI SDK for Flutter — LLM inference, vision, STT, TTS, image generation, embeddings, RAG, and function calling. Metal GPU on iOS/macOS.

Temporal Code Intelligence platform. Time-series complexity analysis across Python, JavaScript, Java, and Go. Linear regression trend detection, cyclomatic complexity tracking, confidence scoring, and predictive decay forecasting.

Production-grade OpenClaw personal assistant setup. Security-hardened, 15+ custom tools, Purple-Team audited. Templates & architecture docs for the community.

Multi-Drone based Single Object Tracking with Agent Sharing Network

GLPI Android Inventory Agent

Learn function calling and tool calling from simple tools to full agents.

Connect agent policies for distributed ML applications

[Patterns] MedAgentsBench: Benchmarking Thinking Models and Agent Frameworks for Complex Medical Reasoning

A powerful desktop AI assistant with agentic browsing, listening, coding CLI and editor

Claude Code and AI Agents Skills for Developer Marketing.

The TTCP CAGE Challenges are a series of public challenges instigated to foster the development of autonomous cyber defensive agents. This CAGE Challenge 4 (CC4) returns to a defence industry enterprise environment, and introduces a Multi-Agent Reinforcement Learning (MARL) scenario.

This repo contains hands-on labs for building AI Agents using the Azure AI Agent Service SDK and Semantic Kernel. The Azure AI Agent Service is used to create AI agents and Semantic Kernel is used to orchestrate the agents in a multi-agent system.

Web Cloud-based AI task management and agent orchestration platform

A self-hosted web application for managing AI skills, workflows, and contexts with full MCP (Model Context Protocol) integration. Organize, manage, and dynamically load specialized knowledge bases into any AI Agent just by toggling your Skills On/Off in simple local hosted WEB UI.

Multi-agent coordination system for Claude Code with institutional memory

See what your AI coding agent is doing. Multi-provider assistant & session monitor for VS Code and the terminal — inline completions, code transforms, and a full TUI dashboard tracking tokens, cost, and context across Claude Code, OpenCode, and Codex. Switch between multiple Claude Code and Codex accounts.

SkillDock.io is a registry of reusable AI Skills based on the AgentSkills specification. Discover, install, and publish versioned Skills that work across OpenClaw, Claude, and any Skills-compatible agent.

GNAP — Git-Native Agent Protocol. RFC Draft for git-based agent orchestration. Zero servers.