发现 Skills

Extract reusable knowledge from the current session into a persistent skill. Use when you discover something non-obvious, create a workaround, or develop a multi-step workflow that future sessions would benefit from.

Comprehensive guide for implementing native Stream Deck plugins (C++, C#, etc.) that communicate with Stream Deck via WebSocket. Native plugins are launched as executable binaries with specific command-line arguments and communicate using the WebSocket protocol defined by the Stream Deck SDK.

Query and summarize Claude Code terminal session history.

Extracts questions from .xlsx/.xls/.docx/.pdf exam papers or question banks, evaluates item-writing quality across five dimensions (content validity, structural validity, difficulty control, discrimination potential, standardization), and generates an interactive HTML visualization report. Use when: (1) a user requests evaluation of an exam or question bank, (2) quality checks are needed after item writing, (3) comparing item-writing levels across multiple papers, (4) generating an item analysis report, (5) a user uploads Excel/Word/PDF files for analysis. Trigger phrases: evaluate exam, evaluate question bank, question quality, exam analysis, question-bank analysis, item-writing quality, item inspection, exam evaluation, question quality assessment, .xlsx, .docx, .pdf

A skill that analyzes a project and automatically generates an OpenAPI 3.x specification. It reads project overviews from CLAUDE.md or AGENT.md, auto-detects programming languages and frameworks, and extracts API endpoints. Use cases: (1) when you want to create API specifications for an existing project, (2) when you want to auto-generate API documentation, (3) when an OpenAPI/Swagger specification is required.

Ask Fortytwo high-complexity questions where the best answer is required — coding, architecture, hard reasoning, and more.

Use when the user asks to create or edit videos end-to-end (script → video, auto-cut/jumpcut, captions/subtitles, polishing for Shorts/Reels/TikTok). Current implemented backend: local FFmpeg (probe/render/jumpcut/burn-subtitles/polish). Planned/optional backends: Remotion (motion graphics templates), VectCutAPI (CapCut/剪映 timeline editing), and video-audio-mcp (MCP tool wrapper) when available. Produces a finished video artifact (MP4 by default) from assets + copy + a design/storyboard plan.

Use this agent when you need to create email campaigns, newsletters, or automated email sequences. Call this agent when setting up email marketing, creating welcome sequences, or developing email communication strategies.

Younity angle debate skill. Use after younity-researcher has produced a research report. Independently validates each recommended angle with fresh web research — affirming or refuting the researcher's reasoning. Produces a full debate report for user review. The user decides which angles to approve, refine, or reject. For any angles the user rejects, triggers younity-researcher to re-research replacements using the debator's counter-evidence as a starting point. Invoke when the user says "debate angles", "check the angles", or wants to pressure-test angles before writing copy.

Retrieving current, curated API documentation for external libraries and SDKs during development. Covers search strategies, token-aware fetching, annotation persistence, and provider fallback hierarchy. Use when writing code against an external API (Stripe, OpenAI, Anthropic Claude API, AWS, FastAPI, Supabase, etc.), debugging an integration issue, evaluating an SDK's capabilities, looking up current endpoint signatures or parameters, or when the agent may hallucinate API details due to training data staleness. Also activates for external API reference, SDK documentation, library docs lookup, current API signatures, and integration reference.

QuickBooks Online Integration — use when pulling financial reports (P&L, Balance Sheet, Cash Flow) from QuickBooks Online. Requires OAuth 2.0 credentials in .env. Works with the finance-ops skill for full CFO analysis.

<!-- AUTO-GENERATED by export-skills.py — DO NOT EDIT -->

<!-- AUTO-GENERATED by export-skills.py — DO NOT EDIT -->

Applies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.

Fast headless browser for QA testing and site dogfooding. Navigate any URL, interact with elements, verify page state, diff before/after actions, take annotated screenshots, check responsive layouts, test forms and uploads, handle dialogs, and assert element states. ~100ms per command. Use when you need to test a feature, verify a deployment, dogfood a user flow, or file a bug with evidence. gstack also includes development workflow skills. When you notice the user is at these stages, suggest the appropriate skill: - Brainstorming a new idea → suggest /office-hours - Reviewing a plan (strategy) → suggest /plan-ceo-review - Reviewing a plan (architecture) → suggest /plan-eng-review - Reviewing a plan (design) → suggest /plan-design-review - Auto-reviewing a plan (all reviews at once) → suggest /autoplan - Creating a design system → suggest /design-consultation - Debugging errors → suggest /investigate - Testing the app → suggest /qa - Code review before merge → suggest /review - Visual design audit → suggest /design-review - Ready to deploy / create PR → suggest /ship - Post-ship doc updates → suggest /document-release - Weekly retrospective → suggest /retro - Wanting a second opinion or adversarial code review → suggest /codex - Working with production or live systems → suggest /careful - Want to scope edits to one module/directory → suggest /freeze - Maximum safety mode (destructive warnings + edit restrictions) → suggest /guard - Removing edit restrictions → suggest /unfreeze - Upgrading gstack to latest version → suggest /gstack-upgrade If the user pushes back on skill suggestions ("stop suggesting things", "I don't need suggestions", "too aggressive"): 1. Stop suggesting for the rest of this session 2. Run: gstack-config set proactive false 3. Say: "Got it — I'll stop suggesting skills. Just tell me to be proactive again if you change your mind." If the user says "be proactive again" or "turn on suggestions": 1. Run: gstack-config set proactive true 2. Say: "Proactive suggestions are back on."

Library OS — the persistent digital library system. Use when adding a book, deepening an existing review, extracting insights from handwritten notes / highlights / photos, or when the user wants to build their own library on their site. Handles the full workflow from capture to publish.

Production-ready Go code review (QA + security + maintainability) for this project only.

Use when preparing project releases with version management, changelog updates, and documentation synchronization - automates semantic version calculation, file updates, and release preparation

Manages connected MCP sources for Tandem search. Detects available sources, guides users to connect new ones, handles source priority ordering, and manages rate limiting awareness.

Guide for modernizing and migrating MSBuild project files to SDK-style format. Only activate in MSBuild/.NET build context. USE FOR: converting legacy .csproj/.vbproj files with verbose XML to SDK-style, migrating packages.config to PackageReference, removing Properties/AssemblyInfo.cs in favor of SDK auto-generation, eliminating explicit <Compile Include=\"...\"> lists via implicit globbing, and consolidating shared settings into Directory.Build.props. Indicators of legacy projects include the ToolsVersion attribute, explicit <Import Project=\"$(MSBuildToolsPath)\"> of Microsoft.*.targets, packages.config, Properties/AssemblyInfo.cs, explicit <Compile> entries, or unusually large .csproj files (> ~50 lines for simple projects). DO NOT USE FOR: projects already in SDK-style format, non-.NET build systems (npm, Maven, CMake), or .NET Framework projects that cannot be converted to SDK-style. INVOKES: dotnet try-convert and upgrade-assistant tools.

Provides expertise on kintone API integration, authentication, data synchronization, and customization.

Google Gemini AI models for multimodal tasks. Use for multimodal AI.

フックを介してセッションを観察し、信頼度スコアリング付きのアトミックなインスティンクトを作成し、スキル/コマンド/エージェントに進化させるインスティンクトベースの学習システム。

Enforce a 500-line per-file limit. Emit a warning at 400 lines. Provide guidance on splitting large files into focused modules. Apply during code authoring and editing.