Code agents

MCP-native AI Agent Framework — give Claude / OpenCode / Cursor a project-agnostic engineering brain via MCP

An open-source AI Agent framework in TypeScript with tools, memory, and guardrails support.

Programmatic JS/TS SDK for doc2mcp — convert any docs URL into a hosted MCP server from code, CI, LangChain, and LangGraph.

AI-friendly TypeScript rules engine for serializable JSON business rules and deterministic workflow decisions.

Placeholder, no code yet — name reserved for Aelix: the agent loop, Agent, AgentHarness and the typed HookBus.

Placeholder, no code yet — name reserved for Aelix: ExtensionAPI, the extension loader and the built-in extensions.

Verify what AI coding agents can actually read, write, execute and reach using synthetic canaries.

国内 Agent 工具一键接入国产模型：官方直连、零中转、全中文、可自检

Give your AI coding agent a memory for its mistakes - log before fixing, mechanically enforced

Provider-neutral lifecycle control layer that helps coding agents plan, execute, validate, and prove software tasks to completion.

Self-describing encrypted env files for AI-assisted development

Run bounded, non-interactive jobs across agent harnesses with isolation, exact resume, and provenance

Road signs for coding agents: one true fact at the moment of action, silence otherwise

An ephemeral orchestration hub for AI coding agents: presence, leases, messaging and a shared blackboard, all with TTLs.

A hidden pywebview browser controlled through an agent-friendly HTTP debugging API.

Project-level configuration and synchronization for AI coding agents

Run your own autonomous AI worker on AgentBounty

Launcher for the agentfenster MCP server - Windows GUI automation on a hidden second desktop.

Clean messy CSVs with an agentic workflow: an LLM decides the steps, tested code does the work.

AI Developer Governance & Codebase Evaluator — A GraphRAG-powered multi-agent system featuring autonomous code evaluation, red-team testing, and dynamic LLM-narrated architectural documentation.

Skills Manager: scan, install, update, and analyze AI agent skills safely across Claude, Cursor, Codex, and OpenCode.

A Hamachi server for agents: start a hub, get an address, agents join and work together

Local-first flight recorder for AI coding agents: hash-chained audit trail, MITRE-mapped sequence detection, DLP

Observability for Claude Code subagents. Parses session transcripts, surfaces what agents actually did, and flags the failure patterns.

Compile ordered Markdown policy modules into a deterministic, verifiable AGENTS.md.

Deterministic security scanner for AI agent skills + Trusted Distribution CLI. Local, zero-dependency, executes nothing. Facts, not verdicts.

Blameless postmortems for your coding agent. Auto-memory remembers — it doesn't learn.

One source of truth for your AI agent instructions. Generate and validate AGENTS.md, CLAUDE.md, .cursorrules, Copilot, Gemini and Windsurf instruction files from a single file.

Move a coding-agent session between tools without re-explaining your project.

Approve or deny your AI coding agent's risky commands from your phone, with an audit trail. Works with Claude Code and OpenAI Codex.

Model-independent orchestration of external coding-agent runtimes.

AI-powered Python unit test generation agent with ReAct architecture

OpenAI-compatible API server for the Claude Code CLI — streaming, tools, vision, and live session reuse.

Claude Code's projects/<slug>/ path encoding, as a function

Interactive TUI for debugging token usage of Claude Code sessions

Local privacy and proxy consistency audit for Claude Code and agent skills

Onion-architected, autonomous Claude Code session runner and full Anthropic SDK CLI — never blocks on a human, distinguishes rate limits from exhausted credits, and resumes safely across usage windows.

CLI AI coding agent: ReAct tools, hybrid RAG, multi-provider LLMs

Extract and summarize Claude Code sessions into daily reports

AI development agent for specification-to-code implementation

Dynatrace OneAgent Python code module for Azure Functions

Terminal-based AI coding agent with reusable skills, personas, and a Web UI

Portable, signed agent memory: typed records, a code-enforced fact lifecycle, and storage-agnostic pipeline ports.

Turn any agent into a self-learning system that learns from user feedback - safely, with two lines of code.

Paritok context compression as a LiteLLM Proxy callback — tool filter + content/history compression + non-destructive recall + edit recovery, transparently on every request through LiteLLM Proxy.

Enterprise memory-sharing CLI agent harness for coding, governed data, and productivity work.

A terminal-based coding agent tool inspired by Claude Code

Use OpenAI, Gemini, and any OpenAI-compatible model inside the Claude Code CLI.

Unofficial async Python RPC client for Prime Agent

Terminal chat assistant for software/coding questions, powered by Gemini with Grok fallback.

AI assistant for semantic search and chat with codebases

Knowledge-graph-augmented retrieval for RAG systems: Neo4j-backed entity extraction (regex or LLM-based), co-occurrence graphs, entity deduplication, and GraphRetriever for hybrid vector+graph retrieval via the optional ragleap-rag integration. Framework-agnostic, optional multi-tenant namespacing, no hardcoded models or vocabulary.

Retrieval as a mesh of MCP services — an agentic RAG platform built on LangGraph. Under active development.

AgentBus client — give every coding agent a real inbox and a real email address

Run Claude Code in a throwaway Docker container.

Turn AI-generated art into game-ready 2D assets

COSE Sign1 proof-of-possession helpers for Swarmauri

Single-purpose AI coding agent powered by llama.cpp -- local-first, terminal-based, MCP-aware.

True Memory Fragments: lazy, source-bound code memory for AI coding agents

Messaging between concurrent AI coding sessions, across clients, over MCP. Zero infrastructure. (YAAC: Yet Another Agentic Chat.)

MailKite review: compare features, pricing, use cases, access model, and alternatives for this Email agent in 2026. Rated 5/5 from 9 verified reviews. Email webhook and API service without managing mail servers - for devs and agents

Codex Dream Skin review: compare features, pricing, use cases, access model, and alternatives for this Desktop AI Agents agent in 2026. Reversible theme studio for Codex Desktop using local injection

AI agent worktree and dev artifact garbage collector

skrun — manage AI agent skills across projects

AI Agent 环境诊断工具 —— 一条命令诊断你的 AI Agent 环境哪里有问题

Standalone Real Memory AI Agent CLI for project chat, coding workflows, provider configuration, and safe Git operations.

AI Agent Toolkit - CLI unificada para herramientas de desarrollo impulsadas por IA

TypeScript-native workflow engine for AI agent orchestration — code as config, no DSL

Wunderkind — specialist AI agent addon for OpenCode with 6 retained specialist agents for any software product team

Self-hostable API, viewer, and CLI for sharing AI agent conversation threads

Parasitic AI Agent Token simulator — writes mock usage into ~/.claude and ~/.codex with real JSONL schemas.

Code organized into context-bounded cells for LLMs coding alongside humans.

LangChain and LangGraph integration primitives for streaming OpenUI generative interfaces

Unified observability for AI agents and infrastructure. One SDK.

Show OpenAI Codex, GitHub Copilot, Z.ai GLM Coding Plan (global and China), and DeepSeek balance usage in the pi coding agent.

Semantic diff review for coding agents: an LLM turns a git diff into a narrative HTML report, a human comments on it, and the feedback comes back as plaintext.

AirRoute — free AI gateway. One endpoint, many free providers, auto-fallback, adaptive auto routing, token compression, and a full admin console. Works with Claude Code, OpenCode, Cline, Copilot & any OpenAI/Anthropic-compatible agent.

Stopgap opencode plugin that smooths the OpenAI/Codex GPT-5.6 rough edges (Luna 404s, missing session titles, reasoning-summary markers) until native support lands

CoDev — AI Coding Agent Hub. Install, configure, and manage multiple AI coding agents.

SamAI SDK is a provider-agnostic TypeScript AI SDK for building AI agents, chatbots, tool calling, RAG, MCP, streaming, voice agents, and multi-provider LLM applications using OpenAI, Claude, Gemini, Groq, Mistral, Ollama, Azure OpenAI, and AWS Bedrock.

LexVibe MCP server — one-step legal compliance for vibe-coded apps: privacy policy, terms of service, cookie consent banner and EU AI Act check from Claude Code, Cursor, Codex, Gemini CLI or any MCP client.

Helix CLI — helix init, helix dev, helix auth

Headless React primitives for AI chat — state management, streaming adapters for OpenAI, Vercel AI SDK, and AG-UI, message format converters, and thread management for OpenUI generative UI apps

Selective Codex skills, lean hooks, indexed memory, and optional multi-agent orchestration

Project-aware, evidence-driven operating system, multi-agent CLI, and secure MCP desktop bridge for coding agents.

Stop multi-agent coding drift: contracts, path leases, evidence gates + optional self-improving dispatch (Homer-meta) for Pi

Repo-local multi-agent collaboration protocol + gateway CLI for AI coding agents (Claude Code, Codex): runs, task claims, evidence gates, review/verify, audit.

Self-hosted multi-device, multi-agent session atlas

A local-first, multi-agent development harness that turns a goal into planned, executed and verified work — with an evidence receipt for every run.

Lifecycle orchestration engine for AI agents — 28 stages, adaptive scope, agent framework mode

Lean, budget-gated agent framework for Claude Code — pods by job-to-be-done

agentic — AI-powered terminal coding agent

Agentic code coverage analyzer

An opencode agentic dev team.

QA Agentic Workflows for OpenCode — 9router edition

Canonical capability manifest schema + 3-step pre-delegation negotiation handshake with HMAC-signed manifests, capability diff, and TTL. Zero runtime dependencies.

Installer for the cmdop CLI — one AI agent that can reach every machine you own

Record Claude Code and Codex work sessions as Markdown or Notion work diaries

FastMCP server that delegates work from Claude Code to OpenAI Codex CLI and Google Gemini CLI as headless subprocesses.

Eval framework for agent skills, across Claude Code, Codex CLI, Copilot CLI and OpenCode