Code agents

Create Remotion ad video projects from a URL with an AI coding agent, no video-generation AI required.

大型語言模型(LLM)的讀書會內容code部分整理，此github提供簡單、免費、易實作code範例，並由淺入深帶你一步步了解如何基於python建置LLM流程、部屬功能等

The CoSAI Risk Map is a framework for identifying, analyzing, and mitigating security risks in Artificial Intelligence systems. As traditional software security practices are not always sufficient for AI, this project provides a shared understanding and a common language for addressing the unique security challenges of the AI development lifecycle.

Agent Base is a source-level research project on coding agents. It compares Codex CLI, OpenCode, Gemini CLI, Kimi CLI, and SWE-agent across agent loops, tools, MCP integration, context/memory handling, UI flows, web architecture, and safety controls.

Replace chaotic AI code generation with a disciplined, spec-driven workflow. SDD Pilot enforces structured development phases and quality gates, leveraging AI agents to conduct online research for best practices and plan your architecture before a single line of code is written.

Automated Benchmarking of LLM Agents on Real-World Software Security Tasks [NeurIPS 2025]

就算有一套完全自动赚钱的工具，你也会经常盯着它。你跟专业交易员对比，差别在于是否看得懂。而EasyDeal 就是解决这个问题的一套框架，提供一组MCP工具，接入 openclaw、Claude Code, Fay等agent后，在MT5交易环境可以协同各种策略工作，监控策略的执行、为你解答各种问题、指导你做出突发处理，甚至直接帮你修改策略代码。

The universal MCP server for Claude Code and other MCP-compatible clients. One installation gives AI access to 10,000+ PowerShell modules and any CLI tool. You and AI collaborate in the same console with full transparency. Supports Windows, Linux, and macOS.

One shared macOS desktop pet for AI agents and apps, with MCP/CLI control and Codex Pets support.

C# coding-agent harness (.NET 10, Microsoft Agent Framework) — sessions, tools, permissions, providers, MCP, plugins

Code intelligence CLI — function-level dependency graph across 34 languages, 34-tool MCP server for AI agents, complexity metrics, architecture boundary enforcement, CI quality gates, git diff impact with co-change analysis, hybrid semantic search. Fully local, zero API keys required.

Let your AI coding agent self-host any open-source app for you. 2,200+ verified recipes — provisioning, DNS, TLS, hardening. Works with Claude Code, Codex, Cursor, Aider, OpenClaw, Hermes.

Open Source A2A Protocol Adapter SDK for Different Agent Frameworks and Harnesses

Claude Code bug bounty agent. Install in current workspace: npx -y hacker-bob-cc@latest install "$PWD" then restart Claude and run /bob-hunt target.com

A practical framework for AI-Assisted Research in Mathematics and Machine Learning

AI agents that write code, review each other's work, and coordinate across your machines

A collection of skills for AI agents (Kiro, Cursor, Windsurf, Claude Code, and others). Each skill is a reusable module that teaches the agent to perform complex tasks with context, structure, and best practices.

面向 PRD、业务规则、SOP、流程文档和产品截图的证据型 RAG 知识库。

An agent platform you build, run, evaluate and improve using coding agents.

video-lens is a coding agent skill that fetches a YouTube transcript and generates a structured HTML report: executive summary, key points, analysis, takeaway, timestamped topic outline, and an embedded in-page player. No API keys, no external services beyond the coding agent itself.

🔨 Kyoko is the all-in-one, fully local tool for debugging and improving your AI agents.

Skill-Inject: Measuring Agent Vulnerability to Skill File Attacks

AI-powered software development system that automates the entire lifecycle from requirements analysis to code delivery through specialized agent collaboration.

Collection of extensions and themes for Pi Coding Agent

transformers safetensors qwen text-generation audio-text-to-text custom_code

gguf llama.cpp qwen3_5 code rl atropos

Compiles AI agent traces and truns them into reusable context.

Vendor-neutral OpenTelemetry skills for AI coding agents, grounded in upstream sources

Antigravity 2.0 不开启 TUN、无需使用 Proxifier 的登录代理修复方案。通过提供已配置好的 version.dll 和 config.json，让 Antigravity.exe 与 language_server.exe 走本地代理，解决 Sign In、Agent 初始化以及 daily-cloudcode-pa.googleapis.com 请求超时问题。

BioDSA: Framework for Vibe Prototyping of AI Agents for Biomedicine

Enhanced MCP server for Godot 4.5-4.7: 33 tools / 199 actions, 3-layer architecture (headless + editor + game bridge), secure sandbox, recording & frame-verify, cross-version CI.

给前端 Agent 使用的中文规范，统一产品界面、数据状态、组件组织和代码归属。

The Kubernetes control room—built in Rust, ready for engineers and AI agents.

Open-source AI agent for workers in mainland China to organize employment facts and evidence, estimate compensation, and prepare negotiation and arbitration drafts.

Research capability for coding agents: multi-source corroboration, active discovery, logical-space thinking, structured output. Browser-first evidence.

Agent Skills for Russian editing, writing, editorial policy, promo copy, posts, and UX copy.

Type one sentence, approve once, walk away — Toh Framework installs an AI build department (14 commands, 8 agents, 23 skills) into your project and keeps building, testing and fixing until it is verified done. Works in Claude Code, Cursor, Antigravity, Codex and ZCode.

基于 Hybrid RAG 与 LangGraph 的本地代码仓库理解工具，支持中英文提问、语义检索、证据引用和多问题拆分

CodeGraph builds a semantic graph of your codebase — functions, classes, imports, call chains — and exposes it through 42 MCP tools, 38 languages, a VS Code extension, and a persistent memory layer. AI agents get structured code understanding instead of grepping through files.

An IDE for coding with agents

Codex-friendly multi-backend academic and AI diagram factory

Recent memories stay detailed. Older ones grow concise. Important ones live on.近期记忆保持详细，旧记忆逐渐精简，重要记忆长期存在。

OpenGraphMemory : Self-hosted knowledge graph extraction, temporal graph storage, and agent memory platform.

Drive your pi coding agent from any browser on your network — laptop, phone, or tablet.

面向 MTF、crossdresser 与性别多元成年人的中文 Agent Skill，基于生如夏花知识库并加入医学安全护栏。

Deep dive into Hermes Agent source code

一个可移植的多 agent 协作 skill，适用于 3 个及以上 AI agent。它能够自动识别 agent 的工具、权限和专长，分配协调者、实现者、验证者等角色；通过单一主写入者机制避免文件冲突；对重要任务执行“规划 → 实现 → 独立复核 → 最终验收”流程，并通过结构化上下文和模型分层降低 Token 与 API 成本

OKF (Open Knowledge Format) — curated catalog of tools, plugins, skills, proposals, and docs for agent-friendly knowledge. YAML-driven, agent-searchable, MCP-ready.

Compiles AI agent traces and truns them into reusable context.

XCode CLI & Simulator MCP. Progressive disclosure of tool responses reduces context and token waste. Lets AI agents build, run and drive iOS apps — without drowning in Xcode's output.

🏛️ The patent plugin for AI coding agents — turn technical ideas into patent disclosures with Claude Code, Codex, and OpenCode.

A secure and efficient alternative to `eval` for coding agents

DeepSeek API Platform integration for OpenClaw

Started out as Dynamic Function Calling for OAI. Upon reviewing a research paper released (LATM) This is/has become a implementation of such system using: OpenAI and Autogen

chatGpt Api SDK 封装，支持 GPT-3.5，GPT-4，账单查询，开箱即用，非常简单！

key0 is the commercial gateway for your AI agents and APIs. Let AI agents discover, pay for, and access your APIs autonomously - no human in the loop.

Okapi: Instruction-tuned Large Language Models in Multiple Languages with Reinforcement Learning from Human Feedback

A wargaming platform compatible with reinforcement learning agents

Code for the RL method MATD3 described in the paper "Reducing Overestimation Bias in Multi-Agent Domains Using Double Centralized Critics"

Compose to create working TeamCity server with PostgreSQL and Agents

Full-stack Python Code Agent platform

This code implements Prioritized Level Replay, a method for sampling training levels for reinforcement learning agents that exploits the fact that not all levels are equally useful for agents to learn from during training.

JoularJX is a Java-based agent for software power monitoring at the source code level on all OSes

Agentic RAG platform purpose-built for small language models (SLM). Robust PDF/SQL search

A simple WhatsApp bot that uses OpenAI's ChatGPT to respond to user inputs. Includes a cron to enable ChatGPT to send messages to recipients at periodic intervals.

Civic Tech & Data AI For Good project. Tracks prosecutor election messaging, mass incarceration indicators, and historical context. Utilizes Agentic AI for data scraping.

Browser based agent orchestrator / You host the server wherever you want / Harnesses all CLI AI harnesses / Mobile and VR Voice coding

Enable AI agents to create and execute their own tools in Rust autonomously. Available as web API and MCP services

Agent skill for modern Android development with Jetpack Compose — best practices for code generation and review

Compiles AI agent traces and truns them into reusable context.

Learn how AI coding agents work by building one from scratch.

agent skills for autonomous data labeling, winner at openai codex hackathon 2026

A full IDE experience anywhere you have a terminal. SSH into a server, run lazyide, and pair with your agentic coding tool.

Craftsman is an powerful generic coding Agent for GitHub Copilot. It provides a Plan agent that does extensive interview to understand the need and a Ralph implementation loop to implement most plan in a single premium request.

Local-first project management and persistent context for OpenClaw, Claude Code, Codex, Cursor, and multi-agent coding workflows.

The world’s fastest AI model gateway (450x less overhead than LiteLLM). Unified access to LLMs across endpoints (openAI, self-hosted, etc.) behind a single authentication layer - with API key generation, user management, request logging, and more

AI agent skills for Adobe After Effects automation — describe what you want, and we'll generate and execute the code.

Zero trust LLM gateway. OpenAI-compatible proxy with semantic routing and load balancing across OpenAI, Anthropic, Ollama, vLLM, and any compatible backend. Identity-based access, virtual API keys, and end-to-end encryption via OpenZiti

Monitor Claude Code sessions, costs, config, hooks, agents & MCP servers from a single Rust binary — TUI (9 tabs) + Web interface with live process tracking, budget alerts, and 30-day forecasting

The explainable, local-first memory engine for AI agents. One ~9 MB binary fuses vector + graph + columnar under VelesQL; why() returns the evidence path behind every recall. No cloud, no glue code — runs on server, browser, mobile and desktop.

Tmux plugin that gives visual feedback for AI agent states (running/needs-input/done). Supports Claude Code, Codex, and custom agents. Pane borders, window title colors, status bar icons.

Manage multiple Claude Code configuration profiles. Switch between work, personal, and other accounts without logging in and out.

call-e integrations for phone calling and real-world outreach across OpenClaw, Codex, and other agent platforms.

'Personal AGI' that thinks on its own. Autonomous cognitive cycle, earned autonomy, 60+ tools. It decides what to do without being told.

WormGPT desktop chat client built with Electron + React and the Groq API.

A curated list of 100+ AI-powered tools, platforms, and resources for Site Reliability Engineering (SRE) — agents, incident management, observability, AIOps, chaos engineering, and more.

Open-source AI-era employment platform connecting skills, jobs, enterprises, governance, and AI agents.

My submission to the ARC-AGI-3 Developer Preview Agent Compitition.

Agentic-Agile Development bridges Agile process values with the capabilities of AI coding agents. It is a methodology for building software through human-agent partnerships.

在自己的 Mac 上搭一套常驻、自愈、走订阅、墙内也能用的自托管 AI 伴侣 · 人看版讲思路，机看版给完整规格 · 文 / 小C & Grace

AI agent skills for Meshy AI 3D generation platform

Code graphs, wikis, and research-backed agentic coding. Deterministic tools for nondeterministic workflows, in one opinionated Claude Code config.

Stealth headless browser engine in Rust — native BoringSSL TLS/JA4 fingerprint, real JS via deno_core, no Chromium, no CDP. Python + MCP bindings. A camoufox alternative.

transformers safetensors internlm2 feature-extraction text-generation conversational

gguf text-generation base_model:deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct base_model:quantized:deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct license:other endpo

ADE — an agentic development environment for macOS. A roster of persistent coding agents, each with its own repo, runtime CLI, and memory.

A risk-aware multi-agent runtime for safe code migrations in OpenCode.

HAR: open agent harness (CLI + MCP) for coding agents. Isolated worktrees, deterministic verify, software factory workflows for Claude Code, Cursor, and Codex.

Terminal-native DeepSeek coding agent: side-git snapshots, session fork/backtrack, auto model routing, skills, sub-agents, LSP diagnostics, live cost tracking, MCP, optional local HTTP/SSE API. Based on CodeWhale (MIT).

Native Android GUI for running AI coding agents locally on-device. No terminal or PC required.