能力标签
genie MCP工具
🛠
AI工具

genie MCP工具

基于 TypeScript · 开源 AI 工具,GitHub 社区精选
英文名:genie
⭐ 315 Stars 🍴 50 Forks 💻 TypeScript 📄 MIT 🏷 AI 8.2分
8.2AI 综合评分
MCP工具AI代理代码生成PR自动化开发者工具
✦ AI Skill Hub 推荐

genie MCP工具 是 AI Skill Hub 本期精选AI工具之一。综合评分 8.2 分,整体质量较高。我们强烈推荐将其纳入你的 AI 工具库,帮助提升工作效率。

📚 深度解析

genie MCP工具 是一款基于 TypeScript 的开源工具,在 GitHub 上收获 0k+ Star,是MCP工具、AI代理、代码生成、PR自动化领域中的优质开源项目。开源工具的最大优势在于代码完全透明,你可以审计每一行代码的安全性,也可以根据自身需求进行二次开发和定制。

**为什么要使用开源工具而非商业 SaaS?**
对于个人开发者和有隐私需求的用户,本地部署的开源工具意味着数据不离本机,不受第三方服务商的数据政策约束。同时,开源工具通常没有使用次数限制和月度费用,一次安装即可长期使用,对于高频使用场景的总拥有成本(TCO)远低于订阅制商业工具。

**安装与环境准备**
genie MCP工具 依赖 TypeScript 运行环境。建议通过 pyenv(Python)或 nvm(Node.js)管理 TypeScript 版本,避免全局环境污染。对于新手用户,推荐先创建虚拟环境(python -m venv venv && source venv/bin/activate),再安装依赖,这样即使出现问题也可以随时删除虚拟环境重新开始,不影响系统稳定性。

**社区与维护**
GitHub Issue 和 Discussion 是获取帮助的最快渠道。在提问前建议先检查 Closed Issues(已关闭的问题),大多数常见问题都已有解答。遇到 Bug 时,提供 pip list 的输出、完整错误堆栈和最小可复现示例,能显著提高开发者响应速度。AI Skill Hub 将持续追踪 genie MCP工具 的版本更新,及时通知重要功能变化。

📋 工具概览

genie MCP工具 是一款基于 TypeScript 开发的开源工具,专注于 MCP工具、AI代理、代码生成 等核心功能。作为 GitHub 开源项目,它拥有活跃的社区支持和持续的版本迭代,代码完全透明可审计,支持本地部署以保护数据隐私。无论是个人使用还是集成到企业工作流,都能提供稳定可靠的解决方案。

GitHub Stars
⭐ 315
开发语言
TypeScript
支持平台
Windows / macOS / Linux
维护状态
轻量级项目,按需更新
开源协议
MIT
AI 综合评分
8.2 分
工具类型
AI工具
Forks
50

📖 中文文档

以下内容由 AI Skill Hub 根据项目信息自动整理,如需查看完整原始文档请访问底部「原始来源」。

genie MCP工具 是一款基于 TypeScript 开发的开源工具,专注于 MCP工具、AI代理、代码生成 等核心功能。作为 GitHub 开源项目,它拥有活跃的社区支持和持续的版本迭代,代码完全透明可审计,支持本地部署以保护数据隐私。无论是个人使用还是集成到企业工作流,都能提供稳定可靠的解决方案。

📌 核心特色
  • 开源免费,支持本地部署,数据完全自主可控
  • 活跃的 GitHub 开源社区,持续迭代更新
  • 提供详细文档和使用示例,新手友好
  • 支持自定义配置,灵活适配不同使用环境
  • 可作为基础组件集成进现有技术栈或进行二次开发
🎯 主要使用场景
  • 本地部署运行,保护数据隐私,满足合规要求
  • 自定义集成到现有系统,扩展技术栈能力
  • 作为开源基础组件进行商业化二次开发
以下安装命令基于项目开发语言和类型自动生成,实际以官方 README 为准。
安装命令
# 方式一:npm 全局安装
npm install -g genie

# 方式二:npx 直接运行(无需安装)
npx genie --help

# 方式三:项目依赖安装
npm install genie

# 方式四:从源码运行
git clone https://github.com/automagik-dev/genie
cd genie
npm install
npm start
📋 安装步骤说明
  1. 访问 GitHub 仓库页面
  2. 按照 README 文档完成依赖安装
  3. 根据系统环境完成初始化配置
  4. 参考官方示例或文档开始使用
  5. 遇到问题可在 GitHub Issues 中查找解答
以下用法示例由 AI Skill Hub 整理,涵盖最常见的使用场景。
常用命令 / 代码示例
# 命令行使用
genie --help

# 基本用法
genie [options] <input>

# Node.js 代码中使用
const genie = require('genie');

const result = await genie.run(options);
console.log(result);
以下配置示例基于典型使用场景生成,具体参数请参照官方文档调整。
配置示例
# genie 配置说明
# 查看配置选项
genie --config-example > config.yml

# 常见配置项
# output_dir: ./output
# log_level: info
# workers: 4

# 环境变量(覆盖配置文件)
export GENIE_CONFIG="/path/to/config.yml"
📑 README 深度解析 真实文档 完整度 44/100 查看 GitHub 原文 →
以下内容由系统直接从 GitHub README 解析整理,保留代码块、表格与列表结构。

简介

<p align="center"> <img src=".github/assets/genie-header.png" alt="Genie" width="800" /> </p>

<p align="center"><strong>Wishes in, PRs out.</strong></p>

<p align="center"> <a href="https://github.com/automagik-dev/genie/releases"><img alt="signed release channels" src="https://img.shields.io/badge/releases-signed%20channels-00D9FF?style=flat-square" /></a> <a href="https://github.com/automagik-dev/genie/stargazers"><img alt="stars" src="https://img.shields.io/github/stars/automagik-dev/genie?style=flat-square&color=00D9FF" /></a> <a href="LICENSE"><img alt="license" src="https://img.shields.io/github/license/automagik-dev/genie?style=flat-square&color=00D9FF" /></a> <a href="https://discord.gg/xcW8c7fF3R"><img alt="discord" src="https://img.shields.io/discord/1095114867012292758?style=flat-square&color=00D9FF&label=discord" /></a> </p>

<br />

Genie is a planning-and-execution layer for AI coding agents. You describe what you want in one sentence; Genie interviews you into a plan, dispatches agents to build it in parallel, reviews the result against acceptance criteria, and hands you something ready to merge.

The whole thing is a lightweight body: a set of skills, plain-markdown documents in git, and a single per-repo SQLite file. No daemons, no Postgres, nothing resident. A command opens the database, runs one transaction, and exits.

What's inside

  • Skills carry the methodology — brainstorm → design review → wish → plan review → work → implementation review, authored once in runtime-neutral form and delivered to every agent skill home.
  • Documents in git. Wishes, designs, and brainstorms are plain markdown under .genie/wishes/<slug>/ and .genie/brainstorms/<slug>/; you diff, review, and version them like any other code.
  • One file of state. Tasks, boards, dependency edges, and wish-group execution state live in a single per-repo SQLite file (.genie/genie.db), on Bun's built-in engine.
  • Small. 15 CLI commands, 4 runtime dependencies (@inquirer/prompts, commander, zod, nats) — nats initializes only when the omni runner starts. A ~0.9 MB single-file bundle. Bun-powered.
  • Spawn-context contract. genie context --wish <slug> [--group g] [--plan] emits one line of versioned JSON — composed branch + resolved base SHA + ready tasks — that a spawn consumes. --plan previews the same payload without side effects; the wishless form resolves the repo's integration branch for plain spawns.
  • Zero daemons, no Postgres. Nothing runs in the background between invocations.

Install

curl -fsSL https://raw.githubusercontent.com/automagik-dev/genie/main/install.sh | bash

Every release is cosign-signed (keyless OIDC) with SLSA provenance; the installer verifies the binary — via gh attestation verify, falling back to cosign verify-blob — before it runs.

The repository-hosted .well-known/latest.json and dev.json manifests are the authoritative channel pointers. GitHub's /releases/latest route and prerelease badge are deliberately not channel authority: a promotion advances only a monotonic manifest and never rewrites already-published assets or channel-significant draft/prerelease/latest metadata.

Genie ships exactly three surfaces, and nothing else:

  1. The signed binary — installed and updated by install.sh and genie update.
  2. The skills — delivered by the skills.sh channel. genie install and genie update run the pinned skills CLI over the tree the signed release put on disk, then record what landed in ~/.genie/skills-install.json. Without the binary, the same skills install with npx skills add automagik-dev/genie (add -g --all for every agent home).
  3. The Orca plugin — an optional lifecycle integration you register with Orca yourself (see below).

There is no Claude marketplace plugin, no Codex plugin, no Genie-installed hooks, and no role-agent profiles.

--integrations auto|codex|claude|all|none (or --skip-integrations) is the consent scope for the skills channel. Any value other than none installs to every detected agent skill home, because the skills CLI already installs per agent; none skips the channel entirely, writes no record, and reports skills: skipped (consent: none). A failed skills install never rolls back the promoted binary — it prints the exact remedy command and sets a non-zero exit code.

Upgrading from a plugin-era release? genie update runs a one-shot, backup-first retirement of what that era left on the host — Codex and Claude plugin registrations and caches, the stamped Claude workflow, role-agent TOMLs, managed skill mirrors, Hermes and pi links. Every asset is classified before anything is touched, and only the ones provably Genie-owned and unmodified are removed; modified, unmanaged or ambiguous assets are preserved and reported. Backups land under ~/.genie/state-backups/integration-retirement-<timestamp>/. The window covers assets written by releases 5.260711.6 or newer; older hosts follow the manual steps in the docs.

From inside a trusted initialized repo, run genie init to scaffold state and retire proven-owned historical MCP routes. Then run genie doctor to confirm the install: it reports one skills: <agent> <present>/<total> @ <ref> line per known agent skill home, not detected for a home this host does not have, and a warning naming genie update when skills are missing or older than the running binary.

Installing the plugin in Orca

genie setup --orchestration-mode orca selects Orca as Genie's lifecycle authority. It does not register the Genie plugin with Orca — that is a separate, Orca-side install. Orca accepts exactly two kinds of source:

  • a marketplace source: a git repo whose root holds orca-marketplace.json;
  • a plugin source: a git repo whose root holds orca-plugin.json, or a local folder containing orca-plugin.json.

The genie repository root can never be the plugin tree. Orca's loader rejects any install tree containing a symlink ("unsafe file path or symlink") and caps an install at 2000 files / 50 MB. This repo has docs -> .docs-vendor/genie, runs to roughly 14,000 files in a dev checkout, and keeps its manifest nested at plugins/genie/orca-plugin.json, which a git plugin source never looks at. So the plugin is published as a **tree-only git ref whose root is plugins/genie** — symlink-free, ~132 files, ~1.3 MB:

RouteWhat to give Orca
Marketplace sourcehttps://github.com/automagik-dev/genie.git, ref main — the index; the plugin itself resolves to ref orca-plugin
Plugin git sourcehttps://github.com/automagik-dev/genie.git, ref orca-plugin (stable) or orca-plugin-dev (pre-release)
Local folder~/.genie/plugins/genie (what genie install/genie update ships)

.github/workflows/orca-plugin-ref.yml republishes those refs: every push to main that touches plugins/genie force-pushes a parentless commit carrying that subtree to refs/heads/orca-plugin, and every such push to dev does the same to refs/heads/orca-plugin-dev. They are tree-only by design — no history, no shared ancestry with main, never merged back. Orca pins the commit it fetched, so a republish cannot retroactively change an existing install.

The repo root carries only orca-marketplace.json, a source-only index no release tarball contains. scripts/orca-manifest-parity.test.ts fails the build if the index drifts from the plugin's identity, or if plugins/genie ever grows a symlink or crosses Orca's file cap.

Install, update, rollback, and uninstall

Signed release tarballs include plugins/genie/orca-plugin.json and the compiled Orca entrypoint on every supported platform. The normal installer stages and verifies that payload; authority remains standalone until the explicit setup command above. genie update preserves the selected mode and lifecycle history, verifies the replacement payload, and refreshes a prior Genie ownership claim only after an Orca compatibility probe. Run genie doctor after installation or update before resuming lifecycle mutations.

genie update --rollback checks the retained rollback state and prints signed-version reinstall guidance when a safe in-place rollback is unavailable; follow that guidance, then run genie doctor. A failed update, rollback, or mode preflight leaves the prior configuration and authority unchanged. genie uninstall removes only ownership-proven Genie artifacts and registrations. Modified or unproven files are preserved, and neither local Genie history nor Orca records are deleted. Review the command's backup/recovery output before removing any retained files manually.

Quickstart

The lifecycle is shared by every agent the skills channel reaches. Claude Code invokes a skill as a slash command; Codex and the rest invoke it by name or in plain language:

1. /brainstorm or "brainstorm this"   an idea → DESIGN.md → mandatory design review
2. /wish or "turn that into a wish"   accepted DESIGN.md → a scoped WISH.md
3. /review                            mandatory plan review; persist APPROVED or concrete gaps
4. /work                              native role subagents build each approved group
5. /review                            independent implementation review: SHIP, FIX-FIRST, or BLOCKED

Skills are discovered from the agent's own global skills home, so there is no owner-qualified selector and no plugin tier to disambiguate against. The starter cards shipped inside each skill stay selector-free for the same reason.

Re-run genie board any time for a current snapshot of task state on the kanban. The plan documents land in git as you go; the operational state lives in .genie/genie.db.

🎯 aiskill88 AI 点评 A 级 2026-05-20

genie创新性强,将需求对话转化为PR的全流程自动化方案独特且实用。代码质量和维护活跃度良好,适合现代AI驱动开发场景。

📚 实用指南(长尾问题)
适合谁
  • 需要让 Claude / Cursor 操作本地工具的 AI 工程师
  • 构建多智能体协作系统的 Agent 开发者
最佳实践
  • 配置 MCP 服务器时建议使用 stdio 传输 + JSON-RPC,避免暴露公网
  • Agent 任务先做 dry-run 验证工具调用链,再开启自主执行
常见错误
  • API key 直接提交到 git 仓库(请用 .env 并加入 .gitignore)
  • MCP 配置路径拼错或权限不足,重启 Claude Desktop 才生效
部署方案
  • CLI:直接 npm install -g / pip install,命令行调用
  • 云端托管:可放在 Vercel / Railway / Fly.io 等 PaaS 平台
相关搜索
genie 中文教程genie 安装报错怎么办genie MCP 配置genie Agent 工作流genie 与同类工具对比genie 最佳实践genie 适合谁用

⚡ 核心功能

👥 适合谁
  • 需要让 Claude / Cursor 操作本地工具的 AI 工程师
  • 构建多智能体协作系统的 Agent 开发者
⭐ 最佳实践
  • 配置 MCP 服务器时建议使用 stdio 传输 + JSON-RPC,避免暴露公网
  • Agent 任务先做 dry-run 验证工具调用链,再开启自主执行
⚠️ 常见错误
  • API key 直接提交到 git 仓库(请用 .env 并加入 .gitignore)
  • MCP 配置路径拼错或权限不足,重启 Claude Desktop 才生效

👥 适合人群

AI 技术爱好者研究人员和学生开发者和工程师技术创业者

🎯 使用场景

  • 本地部署运行,保护数据隐私,满足合规要求
  • 自定义集成到现有系统,扩展技术栈能力
  • 作为开源基础组件进行商业化二次开发

⚖️ 优点与不足

✅ 优点
  • +MIT 协议,可免费商用
  • +完全开源免费,无授权费用
  • +本地部署,数据完全自主可控
  • +开发者社区支持,遇问题可查可问
⚠️ 不足
  • 安装和初始配置可能需要一定技术基础
  • 功能完整性通常不如成熟商业产品
  • 技术支持主要依赖开源社区,响应速度不稳定
⚠️ 使用须知

AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。

建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。

📄 License 说明

✅ MIT 协议 — 最宽松的开源协议之一,可自由商用、修改、分发,仅需保留版权声明。

🔗 相关工具推荐

📚 相关教程推荐
📰 相关 AI 新闻
🍿 AI 圈相关吃瓜
🗺️ 相关解决方案
🧩 你可能还需要
基于当前 Skill 的能力图谱,自动补全的工具组合

❓ 常见问题 FAQ

作为MCP工具主要支持主流语言,具体取决于项目配置和Anthropic模型能力。
💡 AI Skill Hub 点评

经综合评估,genie MCP工具 在AI工具赛道中表现稳健,质量优秀。如果你已有明确的使用需求,可以直接上手体验;如果还在评估阶段,建议对比同类工具后再做决策。

📚 深入学习 genie MCP工具
查看分步骤安装教程和完整使用指南,快速上手这款工具
🌐 原始信息
原始名称 genie
原始描述 开源MCP工具:Wishes in, PRs out. CLI agent that interviews you, plans the work, dispatches pa。⭐315 · TypeScript
Topics MCP工具AI代理代码生成PR自动化开发者工具
GitHub https://github.com/automagik-dev/genie
License MIT
语言 TypeScript
🔗 原始来源
🐙 GitHub 仓库  https://github.com/automagik-dev/genie 🌐 官方网站  https://docs.automagik.dev/genie

收录时间:2026-05-18 · 更新时间:2026-05-19 · License:MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。

📺 订阅 AI Skill Hub Daily Telegram 频道
每天 8 条精选 AI Skill、MCP、Agent 与自动化工具推送
加入频道 →