从当前安装包查询 Cherry Studio 产品信息并排查运行问题。当用户询问功能、路由、快捷键、Provider、语言、Agent、频道、定时任务、Code CLI、当前版本,或报告运行错误、连接失败、配置异常并需要诊断时触发。
日本語の概要は準備中です。原文の説明を表示しています。
Runs Antigravity CLI headlessly for repository analysis and coding tasks. Use when the user asks to delegate work to Antigravity CLI or compare its result with another coding agent.
インストール方法を見るインストールする前に、エージェントに与えられる指示の中身を確認できます。
command -v agy. If it is missing, stop and ask the user to install Antigravity CLI in Code Mate.agy -p "<prompt>" --output-format json
Pass the prompt as one quoted argument. Print mode waits up to --print-timeout (5m by default); pass a shorter value when the caller's own timeout is tighter.
Parse the JSON result and read its status field: anything other than a success status is a failure, and error carries the reason. Treat unparseable output as failure too. Never start interactive Antigravity or an authentication flow.
Antigravity refuses to run headlessly until it is signed in — it returns status: "ERROR" with authentication failed or timed out and tells you to run agy to log in. Do not do that: stop and ask the user to finish Antigravity setup in Code Mate. Never request, read, print, or copy credentials.
Headless mode inherits the permission mode persisted in user settings, and reading and writing files inside the active workspace are auto-approved by default — the default is not a read-only boundary, and a prompt cannot make it one. Do not run it against a writable project for a read-only task. Use a disposable or read-only copy instead. Run against the real project only when the user explicitly requests workspace changes, constrain the directory, and inspect the diff afterward. Never pass --dangerously-skip-permissions, which auto-approves every tool request including outside the workspace.
Example: for an approved edit, ask Antigravity to touch only named files, run it in that project, parse the response and any reported error, and inspect the resulting diff before reporting success.
まだレビューはありません。使ってみた感想をお寄せください。
概要と使いどころ
从当前安装包查询 Cherry Studio 产品信息并排查运行问题。当用户询问功能、路由、快捷键、Provider、语言、Agent、频道、定时任务、Code CLI、当前版本,或报告运行错误、连接失败、配置异常并需要诊断时触发。
日本語の概要は準備中です。原文の説明を表示しています。
Interact with the user's visible Agent browser in Cherry Studio. Use for page navigation, authenticated websites, screenshots, forms, clicks, and browser debugging. Check live browser tools first; browser control requires the Browser setting and an available Agent pane.
日本語の概要は準備中です。原文の説明を表示しています。
Develop, fix, and profile Cherry Studio in a tracked Electron instance. Use for everyday implementation, UI and interaction work, bug fixing, runtime debugging, DevTools inspection, lag or jank investigation, CPU and memory monitoring, leak checks, and startup-performance analysis; reuse a verified workspace instance across instructions and launch or replace one only when required.
日本語の概要は準備中です。原文の説明を表示しています。
Test Cherry Studio PRs by resolving and checking out a PR, statically inspecting its changes, running interactive UI tests against a safely tracked Electron instance through CDP, producing a structured report, cleaning up only the owned test instance, and restoring the original branch.
日本語の概要は準備中です。原文の説明を表示しています。
Run Cherry Studio critical-path system regression tasks through the repository-owned Playwright E2E workflow. Use for full regression, release acceptance, development-branch system validation, or a named cherry-regression-test task on GitHub-hosted macOS and Windows runners.
日本語の概要は準備中です。原文の説明を表示しています。
当用户明确要求搜索、安装、查看、卸载或创建 Skill,或内置 Skill / 工具出现能力缺口、无法完成当前任务时触发。通过 `mcp__skills__search_skills` 搜索并用 `mcp__skills__install_skill` 安装;已安装 Skill 的查看和删除通过产品清单导航到 Skills UI;没有合适结果时调用内置 `skill-creator` 创建并验证自定义 Skill,再继续原任务。普通任务仍先尝试内置能力。
日本語の概要は準備中です。原文の説明を表示しています。