skills-manager
搜索、安装和协调创建 Claude Code Agent Skills。当用户想要搜索技能、安装工具、创建自定义 Skill，或者说"find a skill"、"搜索技能"、"帮我做个 skill"、"create a skill"时触发。也适用于用户说"有没有做 X 的工具"、"我想扩展 Agent 能力"，或当前能力不足需要先查找可复用方案的场景。
git clone --depth 1 https://github.com/CherryHQ/cherry-studio /tmp/skills-manager && cp -r /tmp/skills-manager/resources/builtin-agents/cherry-assistant/.claude/skills/skills-manager ~/.claude/skills/skills-managerSKILL.md
# Skills Manager ## 搜索和安装 **优先复用**: `find-skills` 可用时先调用它完成搜索、质量检查和安装确认；只有不可用时才执行下面的 CLI 降级流程。 **运行时检测**: 优先 `npx skills`，备选 `$CHERRY_STUDIO_BUN_PATH x skills`，都没有则提示安装 Node.js **搜索**: 理解需求→提取关键词→`npx skills find [query]`→展示名称/功能/来源 **安装**: Skills 是第三方代码有完整权限，必须: 展示安全警告→提供源码链接→用户确认→`npx skills add <owner/repo@skill> -y`。位置: 项目级 `.claude/skills/` 或用户级 `~/.claude/skills/` **无结果**: 告知→先尝试直接完成→可复用流程再建议创建自定义 Skill ## 创建 Skills 1. 先确认现有 Skill 无法满足需求，并判断这是不是值得复用的流程；一次性任务直接完成，不强行创建 Skill。 2. 检查可用 Skill 列表。`skill-creator` 可用时必须调用它，并完整遵循其需求澄清、初始化、编辑、验证和迭代流程；不要绕过它直接手写 `SKILL.md`。 3. 只有 `skill-creator` 确实不可用时，才明确告诉用户正在降级，然后用 `npx skills init <skill-name>` 初始化，保持 `SKILL.md` 精简，并按需加入 `scripts/`、`references/`、`assets/`。 4. 创建后至少验证 frontmatter、触发描述和一个真实用例；验证失败就修正后重跑。 **参考**: https://skills.sh/ | `npx skills find/add/init`
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。
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.
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。
Use when Cherry Studio 用户希望报告、提交或整理 BUG、UI/UX 问题或功能建议，但未明确要求创建 GitHub Issue。
Cherry Studio first-party tool and bundled-shell routing for general agents. For straightforward local work in shell-capable sessions, run JS/TS with `bun <file>` and one-off JS tools with `bun x`; run Python with `uv run [--with <pkg>] python` and one-off Python CLIs with `uvx`; search with `rg`. Load this guide before changing project dependencies, deciding whether a tool should be ephemeral or reusable, reading or converting local Office/PDF files, coordinating or delegating across Agent Sessions, or using Cherry-owned web/browser, knowledge, persistent memory, schedules/notifications, IM channels, image generation, artifact reporting, managed CLI, skill, or MCP-server-registration capabilities—even if the user names no tool. Consult it before shell/file workarounds; live tool schemas are authoritative.
Analyze a codebase and recommend Claude Code automations (hooks, subagents, skills, plugins, MCP servers). Use when user asks for automation recommendations, wants to optimize their Claude Code setup, mentions improving Claude Code workflows, asks how to first set up Claude Code for a project, or wants to know what Claude Code features they should use.
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.
Runs Claude Code non-interactively for code analysis and implementation tasks. Use when the user asks to delegate repository work to Claude Code or compare its result with another coding agent.