Custom Skills Hub

Ask which skill or flow fits your situation. A router over the skills in this repo.

Unified ASR (Speech Recognition) skill with pluggable providers (strategy pattern). Supports: SiliconFlow (TeleAI/TeleSpeechASR). Use for transcribing audio/video files to text. Trigger: "转录", "识别语音", "语音转文字", "ASR", "transcribe", "语音识别". Auto-extracts audio from video files via ffmpeg.

CLI skill for Bilibili (哔哩哔哩, B站) with token-efficient YAML output for AI agents to browse videos, users, search, trending, dynamics, favorites, and interactions from the terminal

Use boss-cli for ALL BOSS 直聘 operations — searching jobs, viewing recommendations, managing applications, chatting with recruiters, and batch greeting. Invoke whenever the user requests any job search or recruitment platform interaction on BOSS 直聘.

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

Butler skill — project awareness, daily report analysis, and knowledge base maintenance. Auto-scans session data, analyzes daily work, and maintains the wiki overnight.

Review the changes since a fixed point (commit, branch, tag, or merge-base) along two axes — Standards (does the code follow this repo's documented coding standards?) and Spec (does the code match what the originating issue/PRD asked for?). Runs both reviews in parallel sub-agents and reports them side by side. Use when the user wants to review a branch, a PR, work-in-progress changes, or asks to "review since X".

Shared vocabulary for designing deep modules. Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable or AI-navigable, or when another skill needs the deep-module vocabulary.

Analyze video content and generate platform-ready titles, descriptions, and tags for publishing. Pure prompt-driven, no scripts required.

Markdown prompt-engineering content repurposing skill with 7 sub-skills (LinkedIn/Twitter/Medium/Substack/Newsletter/GitHub Pages + orchestrator). Supports voice profile system, emoji control, and regex blocklist. Zero dependencies, zero modifications required.

Darwin Skill (达尔文.skill): autonomous skill optimizer inspired by Karpathy's autoresearch. Evaluates SKILL.md files using an 8-dimension rubric (structure + effectiveness), runs hill-climbing with git version control, validates improvements through test prompts, and generates visual result cards. Use when user mentions "优化skill", "skill评分", "自动优化", "auto optimize", "skill质量检查", "达尔文", "darwin", "帮我改改skill", "skill怎么样", "提升skill质量", "skill review", "skill打分".

Turn a loose idea into a sequenced map of investigation tickets, then drive them to resolution one at a time.

Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. Use when user says 'diagnose this' / 'debug this', reports a bug, says something is broken/throwing/failing, or describes a performance regression.

Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.

Build and sharpen a project's domain model. Use when the user wants to pin down domain terminology or a ubiquitous language, record an architectural decision, or when another skill needs to maintain the domain model.

Douyin (TikTok China) video upload CLI tool. Log in to Douyin, check cookies, upload videos or publish image posts. Uses `uv run douyin` — no custom scripts needed.

Use when user requests diagrams, flowcharts, architecture charts, or visualizations. Also use proactively when explaining systems with 3+ components, complex data flows, or relationships that benefit from visual representation. Generates .drawio XML files and exports to PNG/SVG/PDF locally using the native draw.io desktop CLI.

Edit and improve articles by restructuring sections, improving clarity, and tightening prose. Use when user wants to edit, revise, or improve an article draft.

Emil Kowalski 的设计工程哲学。涵盖 UI 打磨、组件设计、动画决策和让软件体验出众的隐形细节。

Export Feishu/Lark docs or entire Drive folders to local Markdown using the official Open Platform APIs. Use this skill whenever the user asks to export a Feishu/Lark document, docx link, wiki/doc link, Drive folder, folder tree, or batch of Feishu docs to md/Markdown, especially for personal backups of documents that are viewable but not copy/export enabled.

Guidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults.

Set up Claude Code hooks to block dangerous git commands (push, reset --hard, clean, branch -D, etc.) before they execute. Use when user wants to prevent destructive git operations, add git safety hooks, or block git push/reset in Claude Code.

Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions 'grill me'.

A relentless interview to sharpen a plan or design, which also creates docs (ADR's and glossary) as we go.

Interview the user relentlessly about a plan or design. Use when the user wants to stress-test a plan before building, or uses any 'grill' trigger phrases.

Design-focused skill for building horizontal swipe web PPTs as a single HTML file. Includes both editorial magazine and Swiss-style visual systems, suitable for talks, launch decks, analytical presentations, supporting images, covers, and screenshot redesign.

Compact the current conversation into a handoff document for another agent to pick up. Use when switching sessions, handing off work, or transitioning between agents.

Nuwa character skill distiller. Input a person's name, topic, or vague requirement; automatically performs deep research, extracts a thinking framework, and generates a runnable character Skill. Two entry points: (1) explicit name → direct distillation, (2) vague requirement → diagnosis → recommendation → distillation.

Unified image generation skill with pluggable providers (strategy pattern). Supports: Vertex AI Imagen (Google), dvcode (company tool). Use for any image generation need: article inline images, cover images, general purpose. Trigger: "生成图片", "配图", "封面图", "generate image", "create cover".

生产级前端界面设计与迭代工具。真实可运行的代码、坚定的设计选择、极致的工艺。覆盖网站、落地页、仪表盘、产品 UI、组件、表单和设计系统。

Implement a piece of work based on a PRD or set of issues.

Find deepening opportunities in a codebase, informed by the domain language in CONTEXT.md and the decisions in docs/adr/. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase more testable and AI-navigable.

Personal knowledge base skill. Ingest web pages, Bilibili, WeChat articles, Xiaohongshu posts and more into PostgreSQL; retrieve them via keyword or semantic search. Auto-generates AI summaries and vector embeddings; supports one-click URL ingestion and nightly harvesting.

Track and compare LLM API pricing across 147+ providers. Uses models.dev JSON API as primary source + manual scraping for uncatalogued Chinese providers. Triggers: 查模型价格, 价格对比, model pricing, 降价, 新模型, 供应商查询.

Grill me about specs for the workflows I want to build, within this workspace.

Long-term memory organization guide. Use when you need to clean up, organize, or restructure a MEMORY.md file, or decide what information should or should not be stored in long-term memory.

Migrate test files from `as` type assertions to @total-typescript/shoehorn. Use when user mentions shoehorn, wants to replace `as` in tests, or needs partial test data.

All-in-one WeChat Official Account operations skill. Covers the full workflow from trending topic research, editorial planning, article writing, image generation, cover design, formatting, to pushing drafts.

Search, create, and manage notes in the Obsidian vault with wikilinks and index notes. Use when user wants to find, create, or organize notes in Obsidian.

Use this skill any time a .docx file is involved -- as input, output, or both. This includes: creating Word documents, reports, letters, memos, or proposals; reading, parsing, or extracting text from any .docx file; editing, modifying, or updating existing documents; working with templates, tracked changes, comments, headers/footers, or tables of contents. Trigger whenever the user mentions 'Word doc', 'document', 'report', 'letter', 'memo', or references a .docx filename.

Use this skill to extract structured Markdown/JSON from PDFs and document images—tables with cell-level precision, formulas as LaTeX, figures, seals, charts, headers/footers, multi-column layout and correct reading order. Trigger terms: 文档解析, 版面分析, 版面还原, 表格提取, 公式识别, 多栏排版, 扫描件结构化, 发票, 财报, 复杂 PDF, PDF转Markdown, 图表, 阅读顺序; reading order, formula, LaTeX, layout parsing, structure extraction, PP-StructureV3, PaddleOCR-VL.

Use this skill whenever the user wants text extracted from images, photos, scans, screenshots, or scanned PDFs. Returns exact machine-readable strings with line-level text and optional bbox coordinates. Strong accuracy for CJK, small print, and handwritten text. Trigger terms: OCR, 文字识别, 图片转文字, 截图识字, 提取图中文字, 扫描识字, 识字, 纯文字, plain text extraction, 坐标, 检测框, bbox, bounding box, image to text, screenshot, photo scan, recognize text.

Build a throwaway prototype to flesh out a design before committing to it. Routes between two branches — a runnable terminal app for state/business-logic questions, or several radically different UI variations toggleable from one route. Use when the user wants to prototype, sanity-check a data model or state machine, mock up a UI, explore design options.

Provides React Native performance optimization guidelines for FPS, TTI, bundle size, memory leaks, re-renders, and animations. Applies to tasks involving Hermes optimization, JS thread blocking, bridge overhead, FlashList, native modules, or debugging jank and frame drops.

Use when you need to resolve an in-progress git merge/rebase conflict.

Two-axis code review — Standards (does code follow documented standards?) and Spec (does code match the originating issue/PRD?). Runs both reviews in parallel and reports side by side. Use when reviewing a branch, a PR, work-in-progress changes, or asking to 'review since X'.

Create exercise directory structures with sections, problems, solutions, and explainers that pass linting. Use when user wants to scaffold exercises, create exercise stubs, or set up a new course section.

Configure this repo for the engineering skills — set up its issue tracker, triage label vocabulary, and domain doc layout. Run once before first use of the other engineering skills.

Set up Husky pre-commit hooks with lint-staged (Prettier), type checking, and tests in the current repo. Use when user wants to add pre-commit hooks, set up Husky, configure lint-staged, or add commit-time formatting/typechecking/testing.

Full-pipeline AI short drama / short video production skill. End-to-end workflow from topic research, scriptwriting, and video generation to final output. Reuses tavily (research), xiaohongshu-cli / bilibili-cli (platform data), and libtv-skill (video generation).

All-in-one viral short video replication tool. Input a Douyin / Bilibili / YouTube link or local video path; it automatically handles: download & transcription → structured transcript → 3-dimension viral analysis → content rewrite → 3 voiceover scripts + 3 mini-drama scripts → Viral-5D score → PDF report export.

Intelligent stock analysis skill. Input a stock ticker (A-share/HK/US), auto-fetches real-time quotes + historical K-line data, calculates technical indicators (MA/MACD/RSI/volume/bias), generates a composite score (100-point scale) with buy/sell signals, searches latest news, and outputs an AI-powered decision dashboard. Python script with multi-source fallback (akshare → yfinance → Alpha Vantage), 6-dimension technical scoring.

Read-only storage analysis assistant for macOS / Windows (auto-detects OS). Scans entire disk usage, identifies space hogs, categorizes each item as 🟢 auto-cleanable / 🟡 needs human judgment / 🔴 clean with caution, and generates an interactive HTML report with collapsible sections and one-click copy commands. Can start a local server for browser-based one-click deletion (move to trash or permanent delete). Scan is fully read-only.

反模板化前端设计技能。适用于落地页、作品集和改版项目。自动推断设计方向，输出不千篇一律的界面。

Unified Tavily CLI skill — web search, URL extraction, and deep research via `tvly`. Use this skill when the user wants to search the web, extract content from URLs, conduct research, find articles, look up information, or needs current information from the internet. Routes to the appropriate sub-command based on user intent.

Test-driven development with red-green-refactor loop. Use when user wants to build features or fix bugs using TDD, mentions 'red-green-refactor', wants integration tests, or asks for test-first development.

Teach the user a new skill or concept, within this workspace.

Break a plan, spec, or PRD into independently-grabbable issues on the project issue tracker using tracer-bullet vertical slices. Use when user wants to convert a plan into issues, create implementation tickets, or break down work.

Turn the current conversation context into a PRD (Product Requirements Document). Use when user wants to create a PRD, write product requirements, or formalize a feature specification.

Move issues and external PRs through a state machine of triage roles — categorise, verify, grill if needed, and write agent-ready briefs.

Unified TTS (Text-to-Speech) skill with pluggable providers (strategy pattern). Supports: Edge TTS (free, no API key), Vertex AI Gemini TTS (Google). Use for generating voiceovers, narration, dubbing. Trigger: "配音", "生成语音", "TTS", "text-to-speech", "合成语音", "语音合成".

Use twitter-cli for ALL Twitter/X operations — reading tweets, posting, replying, quoting, liking, retweeting, following, searching, and user lookups. Invoke whenever the user requests any Twitter/X interaction.

Use this skill to read, analyze, and understand video files using Google Cloud Vertex AI's lightweight video model (gemini-3.1-flash-lite-preview). Use it whenever the user asks to summarize a video, extract text/actions from a video, describe what happens in an mp4/mov file, or ask questions about a specific video clip.

Video content analysis tool. Auto-extracts keyframes, transcribes speech (ASR), and generates structured video summaries with content tags. Outputs structured data for agent understanding and can pair with content-adapt for publishing copy.

All-in-one voiceover video editing skill. Full workflow covering transcription, filler-word detection, subtitle generation, and high-quality export. Trigger words: edit voiceover, process video, cut video, generate subtitles, export HD.

Search Weibo content, view trending topics, read user posts and comments. Uses the m.weibo.cn mobile API — no account or API key required. Trigger scenarios: search Weibo content/topics, check real-time hot searches, fetch user posts, view comments.

WeChat Read (微信读书) assistant — search books, manage bookshelf, view notes & highlights, browse reviews, reading statistics, and discover personalized recommendations. Calls WeChat Read APIs via Agent API Gateway.

Generate an interactive bash wizard that walks a human through a manual procedure — third-party setup, a one-off migration, an A→B state transition — opening URLs, capturing values, confirming each step, and writing .env files and GitHub Actions secrets.

Writing, exploit — assemble raw material into a journey of beats, grounding each term before a beat leans on it.

Writing, explore — mine raw fragments, no structure yet.

Reference for writing and editing skills well — the vocabulary and principles that make a skill predictable.

Writing, exploit — shape raw material into an article, paragraph by paragraph.

Use xiaohongshu-cli for ALL Xiaohongshu (Little Red Book, 小红书) operations — searching notes, reading content, browsing users, liking, collecting, commenting, following, and posting. Invoke whenever the user requests any Xiaohongshu interaction.