Custom agents

[NeurIPS'26] SkillOrchestra: Learning to Route Agents via Skill Transfer

基于 Django + Vue 的 RAG 知识库问答系统，支持 Agent 推理、知识图谱、Wiki 自动生成和跨会话记忆。

Offline PII firewall for AI agents and LLM apps: fast local detection and redaction, Claude Code hook, LiteLLM guardrail. Zero network calls, one dependency.

Carto creates a portable AI container of a codebase & every AI tool then knows what breaks before it changes anything. Move fast with AI without losing confidence, context, or control.

The open-source legal AI agent. Documents stay on your disk and redlines arrive in Word as tracked changes. Fork of OpenCode.

A best-practices template for AI-assisted development at scale. It turns your repository into institutional infrastructure where AI agents can work sustainably — converging around documentation as a stable attractor across sessions, long cycles, and multi-role collaboration

Hand off tasks to Devin — a plugin/skill for Claude Code, Codex, Cursor, and any coding agent

Lunon Deep Research: an AI research agent built for institutional-quality analysis. Evaluated on DeepResearch Bench.

Code repository for Manning's Build an AI Agent From Scratch

Pareto-optimal models for cleaning the web — fast, encoder-based main-content extraction from HTML.

生意参谋 sycm.taobao.com 只读数据 CLI + AI 店铺分析 Skill · Windows / Mac · 共用取数插件，在已登录的 Chrome 里取数

Build polished interactive maps and report-ready figures from spatial data with AI agents.

[ARCHIVED — superseded by skillos] Browser-native markdown agent workspace. Not related to llm_os.

Build a coding agent from scratch, one harness primitive at a time. 15 chapters (ch-00..ch-14), each green on two gates.

Memory is a long-term memory module for AI agents, providing capabilities for memory extraction, storage, retrieval, and migration for agents running on the openJiuwen framework.

AI-native construction estimating platform for intake, takeoff, pricing, scheduling, knowledge, and quotes.

Facebook Pages marketing skills for Claude Code and Codex: high-share posts, hooks, engagement replies, and a weekly plan. Publish via Publora. Open source, MIT. Content engineering by Creative Content Crafts.

Long-running fleet orchestration and memory infrastructure for AI agents. Enables persistent context, shared memory, task execution, and governed coordination across multi-agent systems.

[ICML 2026] Evaluating AI Agents on Continuous Software Evolution

Dev-only Rails engine that plugs AI coding agents into your running app: MCP introspection tools + stack-matched skills from companion gems

Open-source, white-label AI agent platform for agencies to deploy and manage branded WhatsApp agents across client accounts, with human handoff, knowledge bases, custom tools, and self-hosting.

Anymotion is an open source AI motion graphics agent that turns product stories into polished SaaS explainers. Give it your idea or website, and let AI handle the branding, storyboard, keyframing, and animations.

A tiny CLI that gives AI agents a complaint box — log the papercuts they hit during work

Tu asistente personal, directamente en tu ordenador. Habla contigo por voz, gestiona WhatsApp, atiende llamadas, consulta la web y recuerda lo importante. Una inteligencia artificial diseñada para trabajar de forma local, privada y autónoma, sin depender de la nube.

MCP server for Indian APIs — GSTIN, IFSC, PAN, UPI, pincode, HSN/SAC. Zero auth. Offline-first. For AI agents.

Official COROS MCP server and Agent skills for integrating sports and health data into AI clients.

Self-hosted PaaS for one machine: apps, local AI models and databases, operable by coding agents over MCP.

Your AI-powered Autonomous DevOps, Infrastructure Monitoring & Deployment Engine

A VisualBasic(.NET) language kernel and runtime for scientific data computing, deep learning, LLM inference, GPU acceleration, visualization and command-line data-science applications — running on .NET (net10.0) across Windows, Linux and macOS.

面向企业运维场景的 AIOps 智能体系统，基于 LLM 与 LangGraph 构建，融合 RAG 知识检索与多工具调用，实现运维问答、异常分析、故障诊断与报告生成。

ship-it - an idea-to-merged-PR delivery pipeline for any coding agent (Claude Code, Kiro, opencode, Codex, Gemini CLI, Cursor...): 7 skills, 6 shared contracts.

Local AI for Omarchy: the model validated for your GPU, one button on the bar. Start serves it, any coding agent opens on it, one click shares it on your tailnet.

A sovereign, zero-trust local AI agent and micro-IDE.

Native macOS video editor that coding agents can drive: everything the UI does, Claude Code and Codex can do through the bashcut CLI or MCP server. Layered timeline, captions, voiceover, plugins.

Autonomous X/Twitter posting pipeline for your AI agent. No API key, no paid scrapers: researches your niche, writes in your voice, publishes through a real browser, and verifies its own posts.

All-in-one board for DeepSeek Harness: MCP manager + Skills + WorkBuddy experts (attach/switch/detach any time) + tool-usage tailor + three direct markets (WorkBuddy connectors, SkillHub 190k+ skills) with one-click install · DSH 一体化看板：MCP/技能/专家随时挂换/工具账本/三大市场直连安装

Describe the dashboard you want. Your coding agent builds it — live, over your own database.

跨 Agent、多语言面试准备 Skill | Codex · WorkBuddy · Claude | JD analysis, tailored answers, role practice & adaptive mock interviews

🎮 AI-powered game studio in your browser. Turn ideas into playable browser games via multi-round dialogue — from single-file HTML to Vite/React and Phaser projects. Describe, iterate, playtest, and publish — zero coding required. 中文：浏览器里的 AI 游戏工作室。通过多轮对话生成可玩浏览器游戏，支持单文件 HTML 与 Vite/React、Phaser 等工程。描述、迭代、发布——无需写代码。

简明技术中文 / Simplified Technical Chinese: a controlled language for Chinese technical writing, modelled on ASD-STE100. 40 rules, a one-meaning-one-word vocabulary, a zero-dependency checker, and an agent skill for Claude Code, Codex and Cursor.

RL observability for LLM agents: behavior maps, native trainer metrics, rollout investigations, and Python/SQL analysis.

You type. It already knows. Local LLMs that read and think while you type.

A composable framework for reproducible evaluation of long-running agents.

Free, open-source connector agent chatbot AI browser to many App via MCP.

StepFun's flagship model for agentic work

SGLang model provider for Strands Agents.

AI-powered red teaming, and pentesting for web applications via opencode : defensive code audit (17 vulnerability categories) + offensive penetration testing (63 attack categories, 15 agents, 11 domains).

联想 ThinkSmart View（CD-18781Y）语音助手系统镜像：基于 rickx/ubuntu-thinksmart，开机直接进语音界面；含中英文刷机教程（写给人和 AI 助手）、出镜像脚本和设备修复服务｜Voice-assistant system image for Lenovo ThinkSmart View (CD-18781Y), based on rickx/ubuntu-thinksmart: boots straight into the voice UI; flashing guide in Chinese and English (for people and AI agents), image build scripts and device fixes

A skilll for agent to add a Civilization-style ceremonial quote when a substantial multi-step task is truly complete.

🐊 Multi-Team Intelligent Orchestrator v5.3.1 — DNA-driven multi-agent collaboration with 9 genes, 11-step pipeline, 60+ roles, 6 self-evolution gears. Built for OpenClaw.

telegram bot with gpt

All the class codes, projects, and assignments for the SMIT Batch 2th.

This repo contains the code for our paper entitled "Multi-Agent Trajectory Prediction with Heterogeneous Edge-Enhanced Graph Attention Network".

The MVC framework for chat apps built on a platform designed for communication systems.

Simple planner for AutoGPT that can create tasks and checklists to execute commands based on its strategy.

This guide demonstrates how to build an Agentic AI system using Google's Agent Development Kit (ADK) and the Model Context Protocol (MCP).

PhysiCar AI - GitHub Codespaces environment for Agent, ROS2 and DeepRacer

面向 iPhone/iPad 的 Python & JavaScript 开发环境。Python 3.13、JS 运行、HTML 预览、语法高亮、分屏编辑、AI Agent（24 个工具）、SSH 服务器管理（终端/SFTP/部署/监控/AI 运维）。Write, Run, Deploy on iOS。

Confidential is the confidential computing stack. We run your AI workloads (inference, training, agents) in hardware-encrypted Trusted Execution Environments (TEEs). You can provide on-prem security and end-to-end privacy guarantees on your off-prem hosted infrastructure. Your data and code stay private. You can independently verify these claims.

This is a repo for studying the application of LLM Agents on Games

🤖 language modeling sft demos

Official Channel Web SDK Loader

The Unified Agent Kit for AI-Driven Development - A comprehensive toolkit for software development powered by AI Agents.

Distributed Heuristic Multi-Agent Path Finding with Communication - ICRA 2021

无需付费的控制手机并完成你指令的Agent！让Agent帮你操控手机点外卖，打网约车，给微信或者QQ等好友发消息.......解放你的双手！并且无需调用GPT-4o等付费api！

Code for reproducing the ACL'23 paper: Don't Generate, Discriminate: A Proposal for Grounding Language Models to Real-World Environments

⚡️ Devex — A Fast, Secure, and Scalable Repl-as-a-Service Platform built for Developers 🚀

一个使用Flutter开发，支持诸多云平台AI大模型API调用的智能工作生活助手应用。除了常规大模型应用，还有极简记账、随机菜品、猫狗之家、waifu图片、MAL动漫排行、BGM动漫资讯、饮食健康等生活日常工具。

😎 Sagentic.ai Agent Framework - Sagentic.ai is a unified platform for building, running and scaling autonomous agents.

2024 Alibaba Global Mathematics Competition AI Track Global 2nd Place Project (Agent Universe)

Multi-modal & multi-domain customer service agent with real time text, voice and soon video

OpenClaw Agent Skill — 对任意 GitHub 项目进行多源深度分析，输出结构化研判报告

Django BM25 full-text search using PostgreSQL - a lightweight Elasticsearch alternative

An unofficial curated list of resources for Amp, an AI coding agent.

KUNO CMS A full-stack blog application with Go backend and Next.js frontend, containerized with Docker for easy deployment.一个i18n的CMS系统

Conversational agent that fuses chat data with live web results through Tavily search, extract, and crawl.

Possibly futile attempt at grounding hype with theory and fundamentals

A collection of software tools to aid myself (and optionally others) in improved agentic ai development

Chemist AI Agent for Developing Materials Datasets with Natural Language Prompts

[IEEE Intelligent Systems] Awesome-Graph-augmented-LLM-Agent (GLA)

Safe Multi-Agent MuJoCo benchmark for safe multi-agent reinforcement learning research.

Training LLMs to reason and analyze data with notebooks

Codes for paper "SafeAgentBench: A Benchmark for Safe Task Planning of \\ Embodied LLM Agents"

Meta Prompting to generate AI instructions, agents, skills & prompts

AI-native macro investment research infrastructure with native MCP, terminal CLI, agent runtime, and disciplined decision workflows. 一个 AI-native 的宏观投资研究基础设施，内建原生 MCP、Terminal CLI、Agent Runtime 与纪律化决策工作流。

Loom is an AI harness, that can be used with local or cloud LLMs, for complex tasks. It decomposes work, drives execution through a verification harness, and keeps models on track with structured state instead of history. It can route between thinking and acting models, verifies outputs, and exposes an APP/TUI/API/CLI/MCP for both humans & agents.

🏛 An agent-first web browser

The Freepik enables popular agent Model Context Protocol (MCP) to integrate with Magnific APIs through function calling.

local-first analytics for AI agent skills

Turn your AI coding sessions into knowledge.

[ICRA2025] MARVEL: Multi-Agent Reinforcement Learning for constrained field-of-View multi-robot Exploration in Large-scale environments

Template repository with AI agent guardrails, safety protocols, and sprint task framework. For Claude, GPT, Gemini, and all LLMs.

An open source implementation of the paper: "Sequential Diagnosis with Language Models" From Microsoft Built with Swarms Framework

Plain-Markdown persistent memory for AI coding agents. Long-term facts, daily logs, scratchpad, and semantic search — works with pi, Claude Code, and any LLM agent.

Self-hosted agentic AI platform powered by local models

An open specification for Zero Trust governance of autonomous AI agents. Five core elements, four maturity levels, enterprise compliance mappings.

Free MIT-licensed Claude Code plugin for SEO, AEO and GEO. Audit websites, plan content, generate schema and check links. Includes 12 skills, 5 agents and 3 commands.

SGLang model provider for Strands Agents.