Coding Agents tools for MCP

Open-source terminal-native AI coding agent with 131K+ stars and 2.5M+ monthly active developers. Provider-agnostic architecture supports 75+ LLM providers plus native LSP auto-configuration, multi-session parallel agents, and MCP extensibility. The build/plan agent split and client/server architecture make it the most complete open-source reference for a terminal-first coding harness

The official open-source AI agent that brings the power of Gemini directly into your terminal. Features context-aware coding assistance, file manipulation, and command execution capabilities

High-performance editor in Rust with native AI features, Zeta edit prediction, and Agent Client Protocol support. ~77K+ ⭐

Block's open-source, extensible AI agent donated to the Linux Foundation's Agentic AI Foundation in April 2026. Its MCP-native architecture treats every capability as an MCP server, making it a practical reference for building vendor-neutral, extensible harnesses where tool integration is the primary extension mechanism rather than framework-specific plugins

Continue is an open-source autopilot plugin for VS Code and JetBrains, enhancing coding with LLMs through features like task and tab autocomplete, natural language edits, file generation, and customization options, available under the Apache 2.0 license github | website

Local-first code intelligence graph for MCP and CLI. Builds a persistent map of your codebase so AI coding tools read only what matters, with benchmarked context reductions on reviews and large-repo workflows.

程序员鱼皮的 AI 资源大全 + Vibe Coding 零基础教程，分享 OpenClaw 保姆级教程、大模型玩法（DeepSeek / GPT / Gemini / Claude / GLM）、最新 AI 资讯、Prompt 提示词大全、AI 知识百科（Agent Skills / RAG / MCP / A2A）、AI 编程教程（Harness Engineering）、AI 工具用法（Cursor / Claude Code / TRAE / Codex / Copilot）、AI 开发框架教程（Spring AI / LangChain）、AI 产品变现指南，帮你快速掌握 AI 技术，走在时代前沿。本项目为开源文档 aiguide，已升级为鱼皮 AI 导航网站

High performance coding agent harness written in Rust, built for heavy multi-agent dev workflows

Gentle-AI configures the AI coding agents you already use: Claude Code, Cursor, OpenCode, Codex, Pi, and more. Choose persistent memory, Organic-Driven Development, curated skills, MCP servers, personas, and optional bounded review. Open source, no agent lock-in.

The most comprehensive Claude Code guide: agentic workflows, hooks, skills, MCP servers, quizzes, and production-ready templates. 430K+ lines.

A Socratic interview gates the spec on an ambiguity score, then one command drives execution, a staged evaluation gate, and a budgeted evolution loop that runs the semantic stage only

Local-first CLI and TUI coding agent that runs open-weight models entirely on your machine through a llama.cpp fork. No account or API key required. 56 built-in tools (browser, filesystem, git, memory, vision), MCP support, five-layer local memory, macOS/Linux/Windows

Cross-platform desktop coding agent built on a Rust runtime, with Git-aware workflows, Deep Review, MCP/Skills/Hooks, remote workspaces, and state-bound Mini Apps

⚡ 需求分析效率提升 200%！全球首个为 AI 编程时代设计的团队协作 MCP 服务器，自动分析需求自动编写前后端代码，下载切图

The ripgrep of AI context: a zero-dependency C++23 CLI + MCP server for coding agents. Find what you want without reading the repo, then check you built what you meant — blast radius, tests-to-run, quality deltas. Signatures at 74.7% fewer bytes than bodies; every guess labelled, every loss published. Paddle out with a map.

Open-source AI agent desktop app for Windows & macOS. One-click install Claude Code, MCP tools, and Skills — with sandbox isolation, multi-model support, and Feishu/Slack integration.

Markdown knowledge graph for you and your AI agents — editor LSP plus CLI and MCP server so agents can search, retrieve, and refactor plain-text notes

CLI token optimizer and AI context generator. Scans codebases to extract routes, schema, components, and dependencies for Claude Code, Cursor, Copilot, Codex, and Windsurf. 9x–13x token reduction, built-in MCP server, zero runtime dependencies

Local first, desktop companion platform with animated pets, plugin SDK and coding-agent integrations.

Open-source CLI that scans your codebase and generates AI agent configs (CLAUDE.md, .cursorrules, skills, MCPs) for Claude Code, Cursor, and Codex. Scores your setup 0-100 and recommends MCP servers

Code quality analyzer for vibe-coded Python. Detects dead code, clones, complexity issues, and coupling problems with MCP integration for AI assistants

CodeAgent is an open-source coding agent that lives in your terminal. It understands your repository, plans multi-step changes, writes and edits files, runs commands and tests, and iterates until the task is done — all with a transparent, extensible tool loop.

A DeepSeek coding agent harness: persistent shell, Ultra subagents, auto approval, Chrome MCP and session telemetry

Whale — blazingly fast, terminal-first AI coding agent for DeepSeek. ~98% prompt cache hit rate, 1M context, MCP tools, dynamic workflows.

A fast coding agent, rpc plugin, sub-agents, sloppy edits, and mcp

Self-hosted AI runtime that orchestrates Claude Code, Codex, Gemini CLI, OpenCode, Copilot CLI, Cursor Agent, Goose, Qwen Code, and Droid with heartbeats, schedules, delegation, runtime skills, and an org chart view. Ships as a desktop app (Electron) and CLI. MIT licensed, TypeScript

Terminal UI that runs Gemini CLI alongside Claude Code, Codex, OpenCode and other coding-agent CLIs, each in its own persistent tmux session. One list shows every session's live status, prompts go into a pane without attaching, a session can spawn into its own Git worktree, and a full-file diff review sends line comments back to the agent. Launches your installed Gemini CLI unmodified, so login, c

Open-source parallel-agent harness: run coding agents on your own computer or a hosted one, controlled from a single web UI with a mobile client. Claude Code, Codex, Grok, Cursor, OpenCode, Copilot, Pi.

Autolith is a self-modifiable general purpose Lisp AI agent

Native macOS app running Claude Code, Codex, Cursor, Gemini, and shells in parallel across git worktrees or remote hosts, with live status, session persistence across restarts, a menu-bar rollup, and an MCP server for agents to spawn others.

Turn DeepSeek Harness into a self-hosted enterprise Agent platform with knowledge, deep research, content generation, vibe coding, browser automation, governance, and admin controls.

Desktop workspace for running Claude Code, Codex, Gemini CLI and other CLI agents side by side on Windows and macOS, with up to 8 parallel tasks each in its own git worktree, hunk-level diff adoption, approval gates for agent tool calls, agent-to-agent channels and sessions that survive reboots

support cc/codex/grok build/kimicode/pi/antigravity cli/opencode sessions viewer, skills manage, token usage statistics, global search, resume, and export to html

Open-source desktop app to orchestrate multiple AI CLI agents (Claude Code, Codex, Gemini) simultaneously with automations, Kanban management, remote control, and MCP servers

Native macOS terminal built on libghostty with an approval inbox for permission prompts, a working/blocked/idle sidebar, and in-terminal diff review.

Production-grade Claude Code and Codex skills for Spring Boot developers

Claude Code PRO — полный курс и гайд на русском: CLAUDE.md, settings, skills, subagents, hooks, MCP, плагины, CI. 8 лабораторных, библиотека промптов и готовый starter-kit. Здесь бесплатные курсы по chatgpt -

The black box for AI-written code. Passive decision graph linking every commit to the agent prompts behind it: drift detection, DLP audit of what agents read, evidence-driven risk, MCP memory for coding agents, signed provenance attestations. One command, offline by default.

The OpenAI Agents API, open source and self-hosted. Run Codex, Claude Code or MiniMax Code on your own machines with your own models, through the official OpenAI SDK.

Native macOS terminal with GPU rendering, durable sessions, coding-agent workflows, and SSH connectivity. Session host, daemon, and CLI support Linux; optional MCP and AI summaries.

Open-source, local-first AI agent for coding and real work. BYOK models, MCP, skills, plugins, workflows, and private knowledge bases.

CLI-first AI coding agent runtime in Rust. One binary, any model provider, MCP-native; runs interactively, piped in CI, as a daemon, over WebSocket, or as an ACP agent inside editors

VS Code extension that bridges mobile devices to Google's Antigravity IDE. Mirror active AI chat sessions on your phone, send messages, upload files, stop AI generation, and automate workflows via a local HTTP API or 9 MCP tools. Listed in the official MCP Registry

Ask GPT, Gemini, Grok, or OpenRouter for a second opinion or a fix, as expert subagents over MCP

A living agent organism in Lisp. It becomes yours and evolves over time

Multi-agent orchestration harness for Kimi K2.6. Parallel coding teams in git worktrees, DAG scheduling, live quality gates, terminal HUD, and MCP skill-hooks

Provider-neutral, local-first CLI coding agent (Chat/Cowork/Code modes) built for governance: org policy engine, enforced network egress, and a verifiable hash-chained audit trail. Works with any OpenAI-compatible endpoint (DeepSeek, OpenAI, Claude via OpenRouter, Groq, Ollama). Zero runtime deps. MIT

Local workspace that runs coding-agent CLIs (Claude Code, Codex, OpenCode, Kimi, Qwen, Hermes, Aider, and others) side by side in tmux, driven from a TUI, a loopback-only Web UI, or the CLI, with an opt-in stdio MCP server that lets one agent spawn, read, steer, and stop the others. Apache-2.0

Open-source browser extension and MCP server for annotation-first UI pair programming with AI agents. Annotate any webpage and bridge UI context to Claude Code, Cursor, Windsurf, or Copilot

Zero-dependency CLI to install AI agent skills as roles & behaviors from any source. Works with opencode, claude-code, cursor, and all spec-compliant agents

Open-source platform for deploying AI agents as production services. Internal state (memory, sessions, skills, MCP, schedules, secrets) lives in Postgres; per-agent Docker workspaces. Fleet ops via single commands (hermit skills enable ... --all). CLI, Web UI, Telegram/Discord/Slack. MIT, TypeScript

Open-source, context-aware terminal coding agent written in Go with multiple providers, Turn Memory for controllable cross-turn tool-output retention, skill-driven MCP integration, subagents, Agent Skills, and hashline edits

Open-source AI agent control plane and software factory: every tool and model call allowed, denied or held for a human, budgeted and recorded, and agents run from tickets, webhooks and CI in hosted containers or on your own runners, landing work as pull requests

Open-source AI terminal and agent runtime for Android with a native PTY, bash/git/Python/Node bundled in the APK (no Termux, no PC), the Codex CLI built in, scheduled on-device agents, local LLMs and an MCP server

Terminal AI agent runtime in Rust with a ratatui TUI, parallel subagents, MCP, and plugins; works with Claude, OpenAI, Gemini, or local models

Cross-platform terminal workspace that runs the real Claude Code in one pane per project, with a live sub-agent view, diff review, Git, and permission prompts that pop out of the Mac notch.

Open source MIT CLI for Anthropic Claude Code. Caps verbose terminal tool output (git, npm, grep, ls, cat, find, tsc), dedups duplicate file reads within a session via a PreToolUse hook, and surfaces per-session cost stats via 15 MCP tools. 1,317 tests, zero network calls

Native macOS IDE that seals each agent in its own repo clone under Seatbelt or Docker, serves each a shared symbol and call-graph index over MCP, and gates every step on tests or your approval. Claude Code, Codex, Cursor, OpenCode.

Native macOS terminal where a manager agent runs the tree — spawning workers and sub-managers, detail flowing down, summaries rolling up — while every node stays the CLI's own TUI you can take over. Claude Code, Codex, OpenCode.

Native macOS app that runs each agent's own CLI and sign-in side by side, keeping sessions in the agent's own store, with handoff between tasks, usage limits across accounts, shared MCP connectors, and a browser the agents drive. Claude Code, Codex, OpenCode, Cursor Agent, Antigravity CLI, Pi.

Python terminal coding agent where the model writes its own multi-agent workflows (Gigacode). Provider-agnostic, local-first, 15+ model providers, MCP client, browser agent

Git-native decision memory that stores constraints and ruled-out alternatives as git trailers and notes, giving coding agents only the guidance still in force for the file being edited

Open-source agentic workspace on the VS Code core that runs agent chats side by side, each optionally in its own Git worktree, with subagents, hunk-level keep or undo of agent edits, and a Git checkpoint per turn. Claude Code, Codex, OpenCode, Cursor, Grok, Antigravity, Kimi, and any ACP agent, plus local models through Ollama.

Orchestrates Claude Code, Codex, OpenCode, and Cursor across machines and cloud workspaces, syncing memory, skills, MCPs, plugins, tools, and configuration into each workspace; its desktop app, CLI, and MCP let developers or agents coordinate the same thread fleet from their preferred environment.

Native Apache-2.0 macOS execution agent with a replaceable model backend and an open runtime for code, documents, presentations, and authorized computer actions

Invisible engineering infrastructure for AI agents

Self-hosted, MIT-licensed, mobile-first web UI that drives the native Claude Code and Codex CLIs on your own machine, with server-side sessions, full-text search, an IDE panel, an MCP board the agents read, and an installable PWA. Single Docker container

Terminal-first Tauri desktop app that runs the real agent CLIs in PTYs with per-turn snapshots, diff review, and rewind, and witnesses every prompt, approval, and tool result into a hash-chained ledger exportable as a signed proof packet or an IETF audit trail anyone verifies offline; multi-engine rooms, push-to-talk voice with local TTS, and an MCP server so agents can drive the console. Linux…

CLI for structured software specs. Write .dog files, compile to .dag graphs, query via MCP server. Validates completeness, detects drift

Local-first AI coding assistant desktop app with MCP integration, multi-provider AI support, and skills marketplace. Built with Electron, React, and LangGraph. MIT licensed