sql-cli
Token-efficient MySQL/PostgreSQL operations using mycli and native CLI tools (Windows/Mac/Linux compatible). Replaces Artisan Tinker for database queries with 87% token savings.
webapp-testing
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
wechat-article
Convert Markdown articles to WeChat Official Account compatible HTML format. Use when users need to convert Markdown files to WeChat articles, apply themed styling for WeChat formatting, generate inline CSS HTML compatible with WeChat editor, process technical articles with code blocks and tables for WeChat, or apply professional WeChat-style layouts with headers and footers.
claudeception
|
containerize
Containerise the current project - Dockerfile, .dockerignore and compose.yaml with current Docker practice.
execute-prp
Implement a feature from a PRP file (written by prp-generator) and drive it until every validation gate passes.
fifteen-factor-app
Fifteen-Factor App methodology - Twelve-Factor plus API First, Telemetry and Security - for cloud-native SaaS architecture. Use when designing a SaaS product or microservice, reviewing an architecture against twelve or fifteen factors, adding architecture constraints to a PRP or PRD, planning a migration of a legacy app to the cloud, or assessing architectural technical debt. Do NOT use for local-only scripts or internal tools that will never run as a service; do NOT use for writing the PRP itself - use prp-generator.
prp-generator
Writes a Product Requirement Plan (PRP) - a researched, self-contained implementation blueprint with executable validation gates, saved to PRPs/. Use when the user asks for a PRP or PRD, wants an implementation plan for a feature, says "plan out" or "spec out" a feature before coding, hands over a feature description file to turn into a plan, wants a one-pass implementation brief for another agent, or wants an existing plan researched and tightened. Do NOT use for client discovery, requirements gathering or scope definition - run a clarification pass first; do NOT use for implementing an existing PRP - use /execute-prp; do NOT use for cloud-native architecture review - use fifteen-factor-app.
document-parse
Structured parsing of complex documents into JSON and LLM-ready markdown with two local engines - LiteParse `lit` for PDFs and images (layout blocks, markdown, built-in OCR) and AILANG Parse `docparse` for office, email and e-book formats plus AI parsing of charts and images. Use when the user needs headings, tables, reading order or layout blocks out of a PDF, wants to read or parse a DOCX, PPTX, XLSX, ODT, EPUB, EML or MBOX file, needs track changes, comments or speaker notes, wants a chart, diagram, photo or handwritten page interpreted, or wants a file converted between formats ("turn this markdown into a DOCX"). Do NOT use for a quick plain-text read of a text-layer PDF - use pdf-extract; for making a scanned PDF searchable or bulk OCR - use ocr; for filling forms, merging, splitting or table export to CSV/Excel - use pdf-processing-pro; for authoring Word, Excel or PowerPoint files from scratch - use the document-skills plugin.
ocr
Adds a searchable text layer to scanned or image-only PDFs with OCRmyPDF (Tesseract) - local, free, deterministic, no AI tokens. Use when a PDF has no text layer, when the user says "OCR this", "make this searchable", "extract text from a scan" or "digitise this book", when pdf-extract returned blank output, for large scanned books or batches of scans, or when OCR must stay offline and cost-free. Do NOT use for PDFs that already have a text layer - use pdf-extract; for tables, headings or structured output from complex layouts, handwriting or charts - use document-parse; for forms, merging or splitting - use pdf-processing-pro.
pdf-extract
Fast, zero-AI text extraction from PDFs that have a text layer (digitally created PDFs from Word, Typst, WeasyPrint, wkhtmltopdf, LaTeX, etc). Uses pymupdf - instant and deterministic. Use when you need to quickly pull raw text from a known text-layer PDF, e.g. "extract text from this PDF", "read this PDF", "get the content of", "what does this PDF say", "quickly read this PDF". Do NOT use for scanned or image-only PDFs, or when this returns blank text - use ocr; for headings, tables, layout blocks or AI analysis of complex layouts - use document-parse; for forms, merging, splitting or table export to CSV/Excel - use pdf-processing-pro.
pdf-processing-pro
PDF toolkit of uv-runnable scripts for forms, tables, merging, splitting and validation. Use when filling a PDF form from data, reading a form's fields, extracting tables to CSV or Excel, merging or splitting PDFs, checking a PDF is valid, or batch-processing a folder of PDFs through those steps. Do NOT use for a quick read of a text-layer PDF - use pdf-extract; for making a scanned PDF searchable - use ocr; for headings, layout or AI parsing of complex documents - use document-parse; for Word, Excel or PowerPoint files - use document-parse to read them or the document-skills plugin to author them.
ffmpeg-cli
FFmpeg CLI reference for video and audio processing, format conversion, filtering, and media automation. Use when converting video formats, resizing or cropping video, trimming by time, replacing or extracting audio, mixing audio tracks, overlaying text or images, burning subtitles, creating GIFs, generating thumbnails, building slideshows, changing playback speed, encoding with H264/H265/VP9, setting CRF/bitrate, using GPU acceleration, creating storyboards, or running ffprobe. Covers filter_complex, stream selectors, -map, -c copy, seeking, scale, pad, crop, concat, drawtext, zoompan, xfade. Do NOT use for image-only processing (use ImageMagick), live-streaming server setup (use OBS/nginx-rtmp), or NLE-style timeline editing (use a video editor).
git-github
Git commits, GitHub Actions workflows, release tags and changelogs, with a live check that resolves every action version from GitHub. Use when writing or editing a workflow `uses:` line, bumping or auditing action versions, validating a workflow file, staging or writing a commit message, naming a branch, tagging a release, or editing CHANGELOG.md. Do NOT use for running a full release (bump, tag, push, GitHub release) - use /release; or for general git troubleshooting such as merge conflicts - handle directly.
release
Cut a release - bump the version in every version file, move the CHANGELOG, commit, tag, push and create the GitHub release, confirming before anything leaves the machine.
google-ads-scripts
Google Ads Scripts development - JavaScript run inside the Google Ads UI against the AdsApp API and GAQL. Use when writing a script to pause or enable keywords, adjusting bids or budgets from performance rules, pulling a GAQL report into Google Sheets, monitoring quality scores, alerting on spend or anomalies by email, or running a script across accounts from a manager (MCC) account. Do NOT use for the Google Ads API from Python, REST or gRPC - this covers Ads Scripts only. Do NOT use for Google Apps Script outside Google Ads - use google-apps-script. Do NOT use for Microsoft Ads, Meta Ads or other ad platforms.
bigquery
Query the GA4 BigQuery export (events_* tables) through a BigQuery MCP server and answer questions from raw event data. Use when asked to run or write SQL against GA4 export tables, unnest event_params or items, build sessions, landing pages or session-level source/medium from raw events, reconcile BigQuery numbers with the GA4 UI, estimate or cap query cost with a dry run, or find which dataset holds a property's export. Do NOT use for live GA4 reports via the Data API or for setting up the BigQuery link and export - use google-analytics; for BigQuery work unrelated to GA4 data, use Google's bigquery-data-analytics plugin directly.
google-analytics
Google Analytics 4 implementation and analysis reference. Use when querying live GA4 report data, setting up a GA4 property or data stream, installing gtag.js, designing event or ecommerce tracking, registering custom dimensions or audiences, sending server-side events via Measurement Protocol, linking GA4 to BigQuery and understanding the export schema, debugging with DebugView, or configuring Consent Mode and data retention. Do NOT use for general GTM container work such as triggers, variables, custom templates or server-side containers - use google-tagmanager; Google Ads scripting or ad reporting - use google-ads-scripts; writing or running SQL against GA4 export tables - use the bigquery skill; Universal Analytics or non-Google analytics tools.
google-apps-script
Google Apps Script development for automating Google Workspace - Sheets, Docs, Gmail, Drive, Calendar and Forms - with built-in services, triggers and the appsscript.json manifest. Use when writing or fixing a .gs file, automating a Google Sheet, sending or processing Gmail from a script, setting up time-based or on-edit triggers, writing a custom spreadsheet function, calling an external API with UrlFetchApp, or scoping OAuth in appsscript.json. Do NOT use for Google Ads scripts - use google-ads-scripts. Do NOT use for Node.js, Cloud Functions, Cloud Run or REST clients of the Workspace APIs - those use different APIs and quotas.
google-tagmanager
Google Tag Manager implementation reference for web and server containers. Use when installing a GTM container snippet, configuring tags, triggers or variables, designing dataLayer pushes, setting up Consent Mode in GTM, debugging with Preview mode or Tag Assistant, building a custom template, setting up a server-side (sGTM) container, or automating containers through the GTM API or the bundled Stape MCP server. Do NOT use for gtag.js installs that bypass GTM, GA4 reporting or BigQuery export - use google-analytics; Adobe Launch, Tealium or other tag managers.
humanise
Humanise text by removing AI writing patterns so it reads as human-written. Use when the user asks to humanise, de-AI or de-slop a draft, says it reads like a robot or like ChatGPT wrote it, or wants a press-release voice given a pulse. Applies 34 patterns from Wikipedia's "Signs of AI writing". Do NOT use for grammar-only proofreading or spell checking - edit normally; or for translating text - translate directly.
transcribe
Transcribe audio or video files to text locally with Whisper - mlx-whisper on macOS with Apple Silicon, openai-whisper elsewhere. Use when the user wants to transcribe a file, get a transcript of a recording, turn a voice note into text, transcribe a meeting, podcast or interview, produce subtitles (SRT/VTT) from audio, or transcribe audio just downloaded with youtube-search. Do NOT use for finding or downloading YouTube videos - use youtube-search; for converting, trimming or extracting audio - use ffmpeg-cli; for live dictation or real-time speech recognition.
youtube-search
Search YouTube with yt-dlp and return structured results (title, channel, views, duration, date, URL), then fetch captions or download audio for a chosen video. Use when the user asks to "search youtube", "find youtube videos about", "what's on youtube about", wants recent or most-viewed videos on a topic, wants videos from a specific channel, needs metadata for a YouTube URL, wants a YouTube video's captions or subtitles, or wants a video's audio downloaded to transcribe. Do NOT use for transcribing a local audio file - use transcribe; for converting or editing media files - use ffmpeg-cli; for playing videos or for non-YouTube platforms.
message
Create and edit rich text message drafts for Gmail, Outlook, and WhatsApp with live browser preview. Runs on Bun for near-instant cold start and opens the preview automatically. Use when writing emails, drafting emails, composing replies, sending messages, writing WhatsApp messages, or when user mentions Gmail, Outlook, WhatsApp, "email to", "reply to", "draft an email", "write an email", "send a message". Do NOT use for reading emails, managing contacts, or calendar invitations.
claude-model-upgrade
Find and upgrade stale Claude model IDs (Fable, Opus, Sonnet, Haiku) across skills, agents, prompts, configs and code after a new Claude release - read-only scan first, approval gate, then edits. Use when a new Claude model is out, the user asks to upgrade or bump model IDs, update sonnet/opus/haiku/fable references, find old or deprecated model names, or migrate Bedrock or Vertex model strings. Do NOT use for choosing which model to use or API pricing questions - use the claude-api skill or the models docs; or for changing Claude Code's own session model - use /model.
quarto-revealjs
Builds HTML slide decks with Quarto's Reveal.js format - new decks from bundled templates, SCSS themes, speaker notes, columns, fragments, code highlighting, Mermaid diagrams and PDF export. Use when creating a presentation or slide deck, making a meeting deck or technical talk in Quarto, editing a .qmd file with format revealjs, adding speaker notes or incremental reveals, customising a Reveal.js theme, highlighting code lines on slides, or exporting Quarto slides to PDF. Do NOT use for PowerPoint (.pptx) files - use Anthropic's document-skills pptx; do NOT use for Quarto documents, websites or books that are not slides.
react-best-practices
React and Next.js performance rules from Vercel Engineering, ranked by impact, with local overrides for React Compiler and Next.js 16 caching. Use when writing or refactoring React components, reviewing a React or Next.js codebase for performance, fixing async data-fetching waterfalls, shrinking a JavaScript bundle, hunting wasteful re-renders, choosing a Next.js server caching pattern, or fixing a hydration mismatch. Do NOT use for Vue, Svelte, Solid or Angular code - follow that framework's docs; React Native - use React Native docs; browser or end-to-end testing of a web app - use Anthropic's webapp-testing skill; Shopify themes - use shopify-developer.
shopify-developer
Shopify theme, Liquid, Hydrogen and app development reference, with live API lookups handed to Shopify's own tooling. Use when editing .liquid files, building OS 2.0 sections, blocks or JSON templates, wiring theme cart behaviour with the Ajax API, speeding up a slow Shopify storefront, debugging Liquid or theme editor errors, building or upgrading a Hydrogen storefront, scaffolding a Shopify app or extension, writing Shopify Functions, or picking a Shopify API version. Do NOT use for WooCommerce, Magento, BigCommerce or other platforms - answer from general knowledge; Do NOT use for exact GraphQL field or schema lookups - use the shopify-ai-toolkit plugin or @shopify/dev-mcp.
tampermonkey
Write and debug Tampermonkey userscripts - scripts that change or automate a website from inside the browser. Use when the user wants a userscript (Tampermonkey, Greasemonkey or Violentmonkey), wants JavaScript or CSS injected into a site they visit, wants page elements hidden or rearranged, wants a form auto-filled or page data scraped, needs to react to URL changes in a single-page app, needs cross-origin requests from a page script, has a userscript that does not run, or is setting one up in TypeScript. Do NOT use for Selenium, Puppeteer or Playwright automation - use a browser-automation tool; for building a browser extension (WebExtensions/Manifest V3) - write it as an extension; or for server-side scripts.
timezone-tools
Get current time in any timezone and convert times between timezones. Use when working with time, dates, timezones, scheduling across regions, "what time is it in X", "convert 3pm Sydney to London", DST checks, or when the user mentions specific cities/regions for time queries. Supports IANA timezone names. Do NOT use for date arithmetic (adding days/months), recurring event scheduling, business-day calculations, or full calendar/booking logic - those need a dedicated date library or scheduling tool.
typst
Typst language reference - markup, math and code modes, set/show rules, layout, tables, templates, and PDF/HTML/PNG/SVG export, current to the latest stable release. Use when writing or editing a .typ file, fixing a Typst compile error, building a Typst table, figure or equation, writing a Typst template, exporting Typst to PDF/A, PDF/UA or HTML, migrating a document to a newer Typst release, or asking what changed in Typst. Do NOT use for Word (.docx) files - use Anthropic's document-skills; for filling existing PDF forms - use the documents plugin's pdf-processing-pro; for extracting text from an existing PDF - use the documents plugin's pdf-extract.
pegasus-projects
|
resolve-pegasus-conflicts
Resolve merge conflicts when upgrading SaaS Pegasus. Use when the user has merge conflicts after running git merge during a Pegasus upgrade, or when they need help with the merge process. Also use when the user mentions 'merge pegasus".
upgrade-pegasus
Upgrade to the latest version of SaaS Pegasus. Use when the user mentions 'upgrade pegasus,' or 'pegasus upgrade.'
solana-analysis
Execute a single Solana MCP tool call over the /mcp HTTP endpoint using a TypeScript script. Use when a user needs transaction, instruction, or account analysis from the command line with explicit arguments.
json-canvas
优化的JSON Canvas可视化系统，专为学习笔记的知识图谱设计。提供思维导图、概念关系图、学习路径图等可视化功能，帮助构建直观的知识结构。
obsidian-bases
优化的Obsidian Bases数据库系统，专为学习笔记管理设计。提供强大的笔记查询、分类、统计和可视化功能，帮助构建高效的知识管理体系。
obsidian-markdown
优化的Obsidian Markdown语法，专注于学习笔记的高效编写和知识管理。包含wikilinks、嵌入、提示框、属性等Obsidian特有语法，提升笔记可读性和知识连接性。
fastapi
FastAPI - Modern Python web framework for building APIs with automatic validation, documentation, and async support. Use for API routes, dependency injection, Pydantic models, middleware, and authentication.
pgvector
pgvector - PostgreSQL extension for vector similarity search. Use for embedding storage, cosine similarity, IVFFlat indexes, and HNSW indexes.
postgresql-16
PostgreSQL 16 database documentation. Use for SQL queries, indexes, transactions, JSON operations, full-text search, and performance tuning.
pymupdf
PyMuPDF (fitz) - PDF manipulation library. Use for PDF text extraction, table detection, image extraction, and document parsing.
python-telegram-bot
python-telegram-bot library for building Telegram bots. Use for handlers, callbacks, inline keyboards, conversations, and bot commands.
tailwindcss
Tailwind CSS utility-first CSS framework. Use for responsive design, dark mode, utility classes, and component styling.
agent-sounds
Add audible notifications (chimes) to Claude Code and other coding agents. Provides a small script that plays configurable sounds on common lifecycle hooks like plan completion, task completion, errors, and long-running command completion. Defaults work out of the box on macOS (system sounds) with cross-platform fallbacks.
calendar-audit
Protect your deep work time. Calendar Audit scores every meeting on your calendar, calculates your deep work gap, and makes specific suggestions to reclaim focus time. Supports multiple calendar tools (screenshot, Google Calendar MCP, Apple Calendar, icalBuddy, gcalcli) and scoring frameworks (5-Dimension, Eisenhower, RACI, Value vs Effort, Custom). Value first — your first audit takes 2 minutes with just a screenshot. Just say "calendar-audit" to get going.
oya
Plan and stay on track. Fast. Practical. Intentional. Oya is a productivity tool for doers who think strategically. It combines your long-term vision with daily adaptation by bringing together ideas from Getting Things Done (GTD) and the 7 Habits of Highly Effective People in a simple and lightning fast workflow. Daily use is recommended. Just say "oya" to get going.
trip-planner
Plan trips with assumptions, not questions. Trip Planner extracts flight constraints from natural conversation, fills in the gaps with smart defaults, and returns structured flight recommendations with direct booking links. Assume first. Show your work. Book the flight. Just say "trip-planner" to get going.
backlog-grooming
Break down an epic into well-refined subtasks. Use when the user wants to groom a backlog, create subtasks from an epic, break down an epic into tickets, or refine tasks from user stories. Triggers on "groom", "backlog grooming", "break down epic", "create subtasks from epic", or any request to turn an epic's user stories into individual tickets.
commit-and-mr
Commit changes and create a Merge/Pull Request following project conventions. Use when the user asks to commit, push, create an MR/PR, or ship/land their work. Handles staging, committing (with conventional commit format), pushing, and MR/PR creation with proper descriptions. Single commit per MR/PR; amends if one already has a commit.
Page 196 of 1621 · 81048 results