Explore

Percho 桌面端 UI 插件（槽位组件替换）。用户说「改 XX 组件样式 / 把 XX 换成 YY 风格 / 给 XX 加个功能 / DIY 界面 / 定制界面 / 界面能改什么 / 换风格」且对象是聊天界面里的工具调用卡、子代理卡、任务列表面板时使用。按 SPEC 在 ~/.percho/ui-plugins/ 下 scaffold 插件，保存即热重载，最后引导用户在设置面板启用。

First-run setup — learn the bboxes of your three key input boxes (Spotlight search, chat input keyboard-down, chat input keyboard-up) plus the keyboard keys and Paste buttons. Run once when SYSTEM shows the first-run notice, before opening apps by search or sending messages. Screenshot each page, read the box coordinates off the returned elements, save them with the screen_layout.py CLI.

Use when the task involves scheduling future work — any "remind me at …", "every weekday …", "check again in 30 min", or closing a fired cron job. Also use to reschedule or list jobs. NOT for one-off in-session waits. NEVER edit jobs.md by hand (the cron parser is strict).

Shop on the 淘宝 (Taobao) app. Use whenever the user wants to buy, order, or price something on 淘宝 / Taobao / 淘宝闪购, or names the Taobao app.

Record and validate swift-snapshot-testing snapshots in the SwiftStreamingMarkdown package: regenerate reference PNGs, run snapshot tests, and diff failed snapshots.

Use when working with MobileClaw's phone-to-desktop Codex bridge, including bridge setup, Tailscale access, MobileClaw-originated tasks, and safe desktop configuration.

Project memory workflow for init/upgrade, profile-aware setup, end session, normal/full close, file organization, document archiving, language switching, versioned update logs, rule review, status, wiki sync, and context recovery. Use for /project-butler, setup/初始化, foundation setup, profile setup, end session/收工, normal close, full close, foundation repair, organize files/整理文件, change language/切换语言, continue/接着上次, continue full context/全面回顾, review claude, sync wiki, status. Maintains project memory files. Runs version freshness check before trigger routing.

Platform-neutral guidance for using Open Browser Use, the open-source Chrome automation stack for AI agents. Use when an agent needs to install, verify, troubleshoot, or operate Open Browser Use through its browser extension, native CLI, JavaScript SDK, Python SDK, Go SDK, or Browser Use style JSON-RPC methods; use for tasks involving real Chrome tabs, user tab claiming, CDP commands, downloads, file choosers, clipboard helpers, or session cleanup.

Compatibility entry for the moved Codex ReHome Skill. Read CalebYcj/codex-rehome-skill before performing migration work.

浙江联通售前咨询技能。当用户咨询套餐资费、流量/语音/宽带产品、办理流程、携号转网、合约规则、靓号、积分、政企业务、国际漫游等联通业务时使用。

SQL 版归因分析技能。当用户问为什么、指标异常、趋势下滑/增长、KPI 未达标、营收/订单/转化/成本波动时使用。

SOP 路由与执行技能。当任务涉及应急预案启动、网络割接、重大故障升级上报、操作规范遵从，或用户要求"按制度/按流程/按规范办"时使用。

问题诊断与归因分析技能。当用户报告指标异常、数据波动、KPI未达标、营收下滑、成本上升时使用。

资源容量分析技能。当用户询问算力资源、GPU 节点、机房、IDC、传输链路、带宽、资源利用率、容量水位、扩容需求、资源规划时使用。

生成数据分析报告。当用户要求"报告/可视化报告/趋势报告/分析报告/统计报告/月报/周报"等任何报告类输出时启用。完整流程：理解需求 → 检索表结构 → 多维度 SQL 查询 → 智能深度分析归因 → 动态风格 HTML 可视化报告（ECharts）→ 分隔符包裹后直接输出到对话。

续约与商机跟进扫描技能。当用户询问哪些客户合同快到期、续约提醒、商机跟进、停滞商机、跟进优先级，或要求生成跟进清单、做客户盘点时使用本技能。

运营指标分析技能。当用户询问网络运营指标、KPI、接通率、掉线率、时延、SLA 达标率、趋势对比、环比周报月报，或报告指标异常时使用。

市场情报与竞争分析技能。当用户需要行业动态、竞品分析、市场规模、政策变化、对标数据时使用。

市场情报每日简报技能。当用户需要市场简报/早报/日报/周报/行业动态汇总，或问"最近有什么值得关注的"时使用。产出 HTML 在线看板。

市场预警研判技能。当用户转述一条市场消息/竞品动作，问要不要紧、影响多大、要不要上报预警时使用。产出分级预警简卡。

保险经营分析技能。用户要求分析保费、赔付率、续保率、渠道贡献、产品结构、机构异常、经营月报或可视化看板时使用。

故障影响分析技能。当用户报告基站故障、网络中断、退服、大面积断网，或需要评估故障影响范围、安排装维人员时使用。

客户 360 画像技能。当用户询问某个客户的综合情况、合作状态、历史消费、已购产品、联系人、在途商机、合同状态，或要求做客户盘点、拜访前背景调查时使用本技能。

将不同领域的主题、资料或脚本制作成有旁白、画面、字幕、动效和成片检查的视频；适用于希望建立或复用个人视频生产流程的创作者。

竞品深度对标分析技能。当用户需要竞品分析/竞品对标/新品解读/价格战应对/竞品档案查询时使用。产出含 ECharts 对比图表的 HTML 在线看板。

经营全景分析技能。当用户需要了解整体经营状况、核心KPI、趋势对比、同比环比、利润分析时使用。

Validate current branch follows feature branch naming conventions

Detect Git remote URL for GitHub integration

Initialize a Git repository with an initial commit

Create a feature branch with sequential or timestamp numbering

Auto-commit changes after a Spec Kit command completes

Post-cycle retrospective for the os-clovy build pipeline: mine the finished session conversation, every PR review surface, and the user's corrections for evidence-backed lessons, then fold each lesson into the one skill or doc file that owns it (repo-review axes and CALIBRATION.md, repo-build-pr workflow, repo-delegate briefs, CONTEXT.md, AGENTS.md) as a small docs-only change. Use after a PR merges or a build/review loop closes, or when the user asks to run a retrospective, capture learnings, or improve the pipeline/loop from what just happened.

End-to-end implementation loop for the os-clovy repo: study the prompt, ask the clarifying questions that change what gets built, plan and architect on the session model while delegating bulk implementation and review to cheaper strong models, isolate work in git worktrees, validate with deterministic checks plus agent-driven live app walkthroughs, record and upload QA video evidence through os-platform, open a draft PR, run the automated Greptile and Codex review loop with judgment, and mark it ready. Use when the user invokes /repo-build-pr (or $repo-build-pr in Codex), or asks to build, implement, ship, or fix something in os-clovy from a feature prompt, bug report, screenshot, PR comment, or tracker task id, optionally with a cross-harness implementer directive like "/repo-build-pr JUN-200 with codex".

Stand up design-featuring work the Open Software way. Use when building, redesigning, or polishing any UI surface in an OS repo (os-clovy, os-scribe, os-marketing-page, os-platform), when a feature needs visual design decisions before or during implementation, when reviewing interface work for house taste, or when someone asks to "stand up the design", "design this feature", "make it feel like ours", "make it feel better", or wants the Open Software design sensibility applied. Orchestrates specialist design skills (impeccable, emil-design-eng, make-interfaces-feel-better, transitions-dev, mobbin) when installed, but is fully self-sufficient without them. Core commitments it enforces - compose from the repo's existing design-system blocks rather than inventing new ones, and quiet, subtle, elegant over loud.

Test, screenshot, and debug a Tauri or Vite frontend in a real browser using Playwright or Chrome DevTools MCP. Use when asked to run/screenshot a desktop-app or web UI, verify a frontend change visually, debug rendering without the native build, or when the native webview (macOS WKWebView) can't be driven by Playwright/CDP. Covers faking the Tauri IPC bridge so surfaces render with no backend.

Help the user create, install, or package zot themes, including theme-only extensions.

Help the user create a new zot extension (slash command, LLM tool, or guard) in any language.

Diagnose and minimally fix a failing test, then re-run.

Check a change for accidentally exposed credentials and private data.

Generate, critique, install, and validate AI-assisted Android string-resource translations for Hermes-Relay. Use when adding a locale, refreshing a locale after English changes, or applying canonical translation corrections.

Setup and run the Hermes-Relay server for Android chat support, terminal, bridge, and related relay features.

UdonSharp scripting skill for VRChat SDK 3.10.5 (active and verified target). Use when writing, reviewing, debugging, or migrating UdonSharp C# and UdonBehaviour code. Positive triggers include UdonSharp, NetworkCallable, NetworkCalling, CallingPlayer, Udon network authorization, synced runtime state, a local public helper, public-method audit, C# to Udon conversion, and UdonSharp Assembly Version Defines. VRCQualitySettings and VRCTween calls, PhysBone/Contact callbacks, world VRCPhysBoneCollider runtime access, persistence, collection, web, and other component APIs trigger this skill when the request is about Udon, C#, or runtime API access. Excludes scene setup, component setup, Build Panel work, layers, optimization, and upload; route those requests to unity-vrc-world-sdk-3.

VRChat skill renovator for knowledge fill, refresh, and quality improvement. Use this skill when updating VRChat skills to new SDK versions, filling missing knowledge, fixing outdated information, or improving skill quality. Targets unity-vrc-udon-sharp and unity-vrc-world-sdk-3 skills. Triggers on: update skills, SDK latest, knowledge fill, skill maintenance, information audit, catch-up, renovate, refresh, improve skills, SDK update.

Drive the open Unity Editor from the shell with the isuzu-unity-cli command: console errors, compiling and tests, scenes, GameObjects and components, prefabs, play mode, uGUI clicks, and C# run inside the Editor. SKILL.md gives a command for each common task, so no tool list is needed.

Manage agent-owned RLM heartbeats from the Python REPL. Use when the user asks the agent to start, create, schedule, or manage a heartbeat, unless they explicitly request the user's /heartbeat.

Search Notion and read/create/update pages and databases via Notion's official hosted MCP server. Tools are auto-discovered from the server at runtime.

Read and write Linear issues, projects, cycles, comments, and more via Linear's official MCP server. Tools are auto-discovered from the server at runtime.

Load an on-disk image (PNG, JPEG, GIF, WebP) into the model's context as a viewable attachment so the model can directly SEE it — for screenshots, diagrams, charts, photos, or scanned pages. Use this when you need to perceive an image's visual contents. Requires a vision-capable model; errors clearly otherwise.

Read-only roster and observation of an agent's parent, siblings, and direct children. Use to discover reachable agents and to inspect family status and bounded recent-message previews without mutating sessions.

Message an agent's parent, siblings, or direct children through the daemon. Discover reachable agents with agent_observe.list_agents, then send direct text without spoofing sender identity.

Continual harness refinement (prompt notes, memory, skills, subagents). Schedule `await refine.run(...)` / `await refine.status()`; apply happens at turn boundary.

Manage the persistent thread goal from the Python REPL. Use to read goal status and budget usage, to start a goal when the user explicitly asks for one, or to mark the active goal complete once its objective is fully achieved.

Check context usage and compact the conversation from the Python REPL. Use when context is filling up, so the session is summarized and context is freed for the next stretch of work.

Orchestrates multi-agent forensic investigations on public GitHub repositories, coordinating parallel evidence collection, hypothesis formation, verification, and report generation.

Provides adversarial code comprehension for security research, mapping architecture, tracing data flows, and hunting vulnerability variants to build ground-truth understanding before or alongside static analysis.

Answers common customer questions from a knowledge base and escalates to a human agent when unable to help. Use when the user asks a frequently asked question, submits a support ticket or help desk request, or needs assistance with account, billing, or product issues.

Consume the bundled native Element Plus Vue runtime for Kooboo AI Apps using exact exports, documented APIs and public theme tokens.

使用此 Skill 创建、检查、编辑和验证 XLSX 工作簿。用户提到 Excel、xlsx、工作簿、工作表、表格，或要求修改、汇总、检查 Excel 文件时使用。只支持基础 XLSX 操作，不处理宏、VBA、图表、数据透视表或外部工作簿引用。

Evaluates agent skills against Anthropic's best practices. Use when asked to review, evaluate, assess, or audit a skill for quality. Analyzes SKILL.md structure, naming conventions, description quality, content organization, and identifies anti-patterns. Produces actionable improvement recommendations.

Create, route, supervise, and communicate with Sutando workers.