npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

autoace-cli

v1.0.11

Published

autoace-cli — MCP CLI for Kuaiyou Master Android automation. Lets Cursor/Claude control the phone, push skills, and deploy domain-coach learning plans.

Readme

autoace-cli

The Agentic Skills - 快游大师电脑端 MCP CLI(npm 包名:autoace-cli)。让 Claude Code / Codex / Cursor 等连接 Android 上的「快游大师」App,编写并推送智能体技能领域教练学习计划

配套 Agent Skill:npx -y skills add kuaiyou-app/kuaiyou-open-source --skill autoace -g -y。CLI:npm install -g autoace-cli@latest

  • npm:https://www.npmjs.com/package/autoace-cli
  • 源码:https://github.com/kuaiyou-app/kuaiyou-open-source/tree/main/autoace-cli
  • 官网文档:https://kuaiyou-app.github.io/docs/
  • 接入教程:https://github.com/kuaiyou-app/kuaiyou-open-source/blob/main/docs/mcp-ecosystem-tutorial.md

兼容旧命令名:kuaiyou-mcp-server(同一二进制)。

旧 npm 包 kuaiyou-mcp-server 已废弃,不再更新。请改用 autoace-clikuaiyou-mcp-server 命令本身仍随本包安装,无需改动现有 MCP 配置。

要求

  • Node.js ≥ 20、npm ≥ 10
  • 手机已安装快游大师,并开启 MCP 服务
  • 局域网模式:手机与电脑同一 Wi‑Fi;点击「MCP 服务」可复制给 Agent 的完整连接配置

安装 / 配置

npm install -g autoace-cli@latest

autoace-cli 是 stdio MCP server。请在 AI 客户端中注册,而不是只在普通终端前台运行:

serverName: autoace
transport: stdio
command: autoace-cli
args: []
env:
  KUAIYOU_DEVICE_IP: "192.168.1.100:41899"
  KUAIYOU_MCP_PAIRING_CODE: "482917"

也可用 command: npx + args: ["-y","autoace-cli@latest"]。env 可省略,首次把 App 复制文案交给 pair_device 即可。

如果客户端提供 TLS 地址,可改用 KUAIYOU_DEVICE_URL=https://host:port。当前仅提供 IP 的客户端仍通过兼容的局域网 HTTP 通道连接;此时应使用可信、隔离的网络,因为配对码和屏幕数据不会获得传输层加密保护。

端口与配对码每次开启 MCP 服务都会变化;配对码不要写入项目仓库、日志或 Git 提交。本机 device.json 权限应为 600。连续错码返回 429 时,按 Retry-After 等待后再试。

契约获取

CLI 不内置技能或学习计划 Schema。

  • 技能:get_kuaiyou_schemaGET /api/mcp/schemavalidate_kuaiyou_skill / push_reactive_skill 运行时拉取同一端点。
  • 生成规则:get_kuaiyou_promptsGET /api/mcp/prompts(设备权威;勿缓存/写入仓库;404 请升级 App,禁止用仓内 markdown 兜底)。
  • 领域教练计划:plans_schemaGET /api/mcp/plans/schemaplans_validate / plans_deploy 运行时拉取该端点。禁止learning-plan.schema.json 镜像进本仓当权威。

客户端是唯一契约源;未连接设备时不能执行完整契约校验。计划路由需 App 版本支持(与 feat/mcp-plan-import 联调);若设备返回 404,工具会提示当前 App 尚未暴露该路由。

配对与会话开场

pair_device 成功后会综合展示:

  1. 配对成功POST /api/mcp/pair
  2. 用户提供的连接/配对材料connectionInfo:App 复制文案中的地址与 设备:品牌 · Android · 分辨率 · App 等;也可传结构化 host/port/code/deviceLabel)——设备画像来自用户材料,由 CLI 假定 pair 响应体结构
  3. CLI 能力摘要

传入的地址/配对码会覆盖本 MCP 进程的目标,并写入本机配置(默认 ~/.config/autoace/device.json,可用 KUAIYOU_CONFIG_DIR 覆盖)。下次冷启动优先使用该记录,无需改 mcp.json。换设备或 App 重新开启 MCP 时再 pair_device 即可覆盖。

设备工具在请求前会探活 GET /api/mcp/health(免鉴权)。超时、连接拒绝或网络失败会立刻停止,提示设备已断开、需要重新配对,并清掉进程内的旧地址覆盖与 pair 缓存——不会继续对失效 IP 截屏/拉 schema。HTTP 401/429 是配对码或限流,不是掉线;业务路由 404 是 App 过旧,请升级。

配套 Agent Skill(autoace)要求:配对后先展示该综合上下文(默认自动化主路径,不要把 plans_* 当并列主功能);若用户同条消息已给出编写任务则立即继续,否则等待指示。写技能前优先调用 observe_screen(截屏 + 可交互节点),不要一上来拉取完整 get_ui_tree。调试时 push_reactive_skill 可带 run: true(或随后 run_skillwait: true),CLI 会等到技能结束或失败并返回 log 摘要,失败附带截屏;手机确认框仍须用户点,CLI 不能跳过。

计划 MCP tools(仅当用户明确要求领域教练 / 学习计划)

| Tool | 设备路由 | |------|----------| | plans_schema | GET /api/mcp/plans/schema | | plans_list | GET /api/mcp/plans | | plans_get | GET /api/mcp/plans/{id} | | plans_validate | schema + POST /api/mcp/plans/validate | | plans_deploy | POST /api/mcp/planspendingConfirm=true,须手机确认) | | plans_delete | POST /api/mcp/plans/delete { "planId" } |

部署成功仅表示进入确认流;同 id 覆盖含进度;新 id 无保有槽返回 HTTP 409。

也可全局安装:npm install -g autoace-cli,客户端 command 使用 autoace-cli

License

Apache-2.0