Claude Skills
The open format is called Agent Skills and works in Claude Code, Codex, Cursor and other agents — most people know it as Claude Skills.
Every Agent Skill we could find on GitHub, deduplicated by content. 79 341 files from 1 736 authors, of which 61 700 are unique — the rest is the same skill repackaged into someone else's repository. For each one: what it weighs in tokens, whether it ships runnable scripts, and which MCP servers it needs.
Use when one Python service must send each agent's, tenant's, team's, or request's spans to its correct Arize space and project using application metadata. Covers dynamic OpenTelemetry routing for custom agent builders and multi-tenant applications, including register_with_routing, set_routing_context, multi-space tracing, and custom span routing.
Optimizes, improves, and debugs LLM prompts using production trace data, evaluations, and annotations. Extracts prompts from spans, gathers performance signal, and runs a data-driven optimization loop using the ax CLI. Use when the user mentions optimize prompt, improve prompt, make AI respond better, improve output quality, prompt engineering, prompt tuning, or system prompt improvement.
Generates deep links to the Arize UI for traces, spans, sessions, datasets, labeling queues, evaluators, and annotation configs. Produces clickable URLs for sharing Arize resources with team members. Use when the user wants to link to or open a trace, span, session, dataset, evaluator, or annotation config in the Arize UI.
Use when AI-written code needs structural cleanup or when implementing a change under a guard against AI coding slop — duplicated reimplementations of existing helpers, try-import fallbacks, broad exception swallowing, speculative abstraction layers, _v2 copies, dead "for future use" code, narrating comments — including requests to "deslop", "remove AI slop", "humanize code", review an AI-generated PR beyond tests passing, or prevent these patterns in the current change.
Kiểm tra chính tả, dấu câu, khoảng trắng, cấu trúc câu, tham chiếu đại từ và kết hợp từ trong văn bản tiếng Việt. Dùng khi người dùng yêu cầu soát lỗi hoặc proofread; không dùng để áp một phong cách, đổi nội dung, sửa code/URL/identifier, hay coi lựa chọn register và phương ngữ là lỗi ngữ pháp.
Làm sạch văn phong tiếng Việt dịch sát từ tiếng Anh, gồm trật tự câu, danh hóa, bị động, business cliché và cụm đúng ngữ pháp nhưng không tự nhiên. Dùng khi bản dịch hoặc bản nháp mang cấu trúc Anh-Việt; không dùng để thay đổi thuật ngữ chuẩn, đơn giản hóa văn bản pháp lý, hay tự dịch nội dung chưa có bản tiếng Việt.
Biên tập văn bản tiếng Việt máy móc, sáo rỗng hoặc đều giọng để câu chữ tự nhiên, rõ và đúng ngữ cảnh. Dùng khi người dùng yêu cầu làm văn bản bớt khuôn mẫu, chỉnh giọng hoặc bảo toàn giọng tác giả; không dùng để suy đoán tác giả là AI, lách detector, sửa thuần ngữ pháp hay biến văn bản chuyên môn thành văn nói.
Kiểm tra và áp dụng nhất quán style guide tiếng Việt cho xưng hô, thuật ngữ, viết hoa, số, ngày giờ, đơn vị, heading, bullet và code. Dùng khi chuẩn hóa một hoặc nhiều tài liệu; không dùng để thay thế grammar review, áp quy tắc mặc định lên style guide riêng, hay đổi tên sản phẩm và nội dung được trích nguyên văn.
Produce media assets using AnyCap: generate images, videos, music, speech, dialogue, and complete audio scenes from text or reference inputs, refine images through interactive visual annotation, and deliver finished assets. Covers the full production workflow from concept to delivery across all media types (image, video, music, audio). Use when creating images, videos, music, voice content, dialogue, complete audio scenes, or any visual/audio content -- including iterative refinement with human feedback. Also use for image-to-image transformation, video generation from images, audio generation from references, and annotation-driven precise edits. Trigger on: media production, asset generation, generate image/video/music/audio, create visual content, produce assets, iterative image editing, annotate and refine, creative workflow, content creation, or any task requiring AI-generated media output.
Turn user-provided data, notes, research, product facts, or structured inputs into AnyCap-style blog posts and then enrich those posts with first-party evidence blocks when needed. Use when a user wants a blog, article, tutorial, learn page, or SEO content draft written in the tone of the AnyCap website. Best fit when the input is data-first rather than prose-first: spreadsheets, JSON, bullets, notes, source URLs, benchmarks, product facts, or rough research. Trigger on: write blog from data, turn notes into article, blog drafting, article production, learn page drafting, SEO article in AnyCap tone, or requests to convert structured inputs into publish-ready content.
Collect structured visual feedback from humans using AnyCap's annotation tool, or create and iterate on diagrams using the interactive whiteboard (Excalidraw). Covers image annotation, URL/web page review with screen recording, video review, audio feedback, and collaborative diagramming with Mermaid input. Use when you need a human to point at things, mark regions, draw on screenshots, review a web page or UI, narrate feedback over a recording, provide any spatially-grounded visual input, create or iterate on architecture diagrams, flowcharts, or wireframes. Also use when you need to present work-in-progress to a human for approval or revision. Trigger on: get feedback, show to user, review UI, annotate, mark up, visual feedback, screen recording, user review, human-in-the-loop, approval flow, interactive review, whiteboard, diagram, draw, flowchart, wireframe, or architecture chart.
Guide for planning and auditing SEO for AI tool, SaaS, and product-led websites. Powered by AnyCap -- the capability runtime that equips AI agents with web search and web crawl through a single CLI. Use when Codex needs to define SEO ICPs, map search intent to page types, inspect live SERPs, write page briefs for tool/comparison/alternatives/pricing/tutorial pages, prioritize technical SEO foundations, plan citations or backlinks, or decide whether programmatic SEO is safe and worthwhile. Trigger on mentions of AI tool SEO, SaaS SEO, product-led SEO, search intent, page type mapping, vs pages, alternatives pages, pricing pages, directory submissions, backlink plans, citations, or pSEO.
Guide for conducting thorough, multi-source research and producing comprehensive, well-sourced reports. Powered by AnyCap -- the capability runtime that equips AI agents with web search (including AI Grounded citations), web crawl, image generation, cloud storage, and one-click web publishing through a single CLI. Use when the user asks for deep research, competitive analysis, market research, technical deep dive, literature review, technology comparison, or any task requiring multi-source information gathering and synthesis. Also use when users say \"investigate\", \"survey the landscape\", \"compare X vs Y\", \"state of the art\", \"write a report on\", \"look into\", \"find out about\", \"analyze the market\", or any inquiry that needs more than a single search. Trigger on mentions of research, analysis, investigation, comparison, report, survey, or deep dive.
Create meme-style images, funny meme drawings, captioned photos, and lightweight social visuals with AnyCap. Use when an agent needs to turn a joke, reaction, campaign hook, quote, or screenshot into a meme image, funny doodle-style meme drawing, captioned image, or short meme-video concept. Combines AnyCap image generation and editing with deterministic local text overlay, optional video generation, and Drive or Page delivery.
AnyCap CLI -- create media humans can see and hear (generate images, video, music, and audio), understand media humans share (analyze images, video, audio), access the web (search, crawl), and deliver results humans can use (Drive for shareable file links, Page for hosted web pages). Use whenever a task involves creating visual or audio content, analyzing media, searching or reading the web, sharing files with humans, or publishing anything as a web page -- even if the user doesn't mention AnyCap by name. Also use for AnyCap authentication (login, API key, credentials), configuration, and feedback. Trigger on: image/video/music/audio generation, media analysis, web search, web crawl, file sharing, page hosting, drive storage, delivering results to users, or any mention of AnyCap.
Use Gemini Omni Flash Preview through AnyCap for practical short-video editing workflows. Trigger when a user wants to edit or refine an existing video with natural-language instructions, especially product replacement, object removal or replacement, relighting, restyling, preserving a person/scene while changing one element, or using reference images with a source video. Covers model/schema discovery, prompt construction, AnyCap CLI command patterns, scenario-specific templates, video QA, and retry strategy for `gemini-omni-flash-preview` / `edit-video`.
World Cup match prediction and pre-match intel for football (soccer) fans: predict who will win and a likely scoreline, backed by structured, verifiable evidence -- player profiles, squad dossiers, current form, head-to-head records, and past-tournament history -- using AnyCap web search and crawl. Weighs FIFA ranking, recent form, head-to-head, and key players, plus any user-defined dimensions (home advantage, injuries, rest, weather, vibes), then makes a clearly-labeled for-fun call with a confidence level and a mandatory disclaimer. Built on the 'search less, not faster' strategy: narrow the search space first, retrieve only what matters, then verify against authoritative sources. Use when a fan wants a World Cup match prediction, predicted score, or who-will-win call; wants a player's profile, bio, form, or stats; wants a full World Cup squad or national-team lineup table; wants to size up two teams before a match; or wants head-to-head and past World Cup context. Trigger on: World Cup prediction, predict the score, who will win, match prediction, predicted scoreline, World Cup, World Cup squad, national team lineup, pre-match analysis, matchup, player form, footballer stats, player profile, head-to-head, or scout a team.
Work with the Toss Securities Open API for Korean and US stock market data, stock info, exchange rates, market calendars, account and holdings lookups, order history, buying power, sellable quantity, commissions, and delegated autonomous trading workflows that can create, modify, or cancel live orders. Use when an agent needs to run user-delegated buy/sell/order-management loops, call or build against developers.tossinvest.com, inspect the Toss OpenAPI schema, generate client code, or operate with TOSS_API_KEY/TOSS_SECRET_KEY credentials.
Git workflow best practices including commit messages, branching strategies, pull requests, and collaboration patterns. Use when working with Git version control.
API security best practices and common vulnerability prevention. Enforces security checks for authentication, input validation, SQL injection, XSS, and OWASP Top 10 vulnerabilities. Use when building or modifying APIs.
Create and manage Claude Code skills following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, keyword triggers, hook mechanisms (UserPromptSubmit), session tracking, and the 500-line rule.
Python development best practices including PEP 8 style guidelines, type hints, docstring conventions, and common patterns. Use when writing or modifying Python code.
Standardize remediation decisions, baseline-policy entries, and legal-review banners for accepted, deferred, or policy-blocked audit findings.
Audit the security domain - auth, authz, secrets, transport, sensitive data exposure, dependency CVEs. Run as part of /audit Phase E.
Audit the UI/UX and accessibility domain - error states, loading states, form validation, keyboard nav, ARIA, color contrast, mobile responsiveness. Run as part of /audit Phase E.
Audit the infra and DevOps domain - CI/CD, deployment, IaC, secrets management, environments, logging, observability. Run as part of /audit Phase E.
Walk the B1-B15 AI-to-AI blind-spot catalog systematically. Use during the audit pipeline AFTER hard-stops and Tambon hunt and BEFORE the 13 domain audits. Output: PRESENT / NOT PRESENT for each B-class with evidence. Findings flow into the appropriate domains.
Audit the database and data layer - schema design, query patterns, migrations, indexing, RLS, soft-delete, transactions. Run as part of /audit Phase E.
Audit the architecture and code quality domain - module boundaries, abstraction layers, code organization, naming, documentation. Run as part of /audit Phase E.
Audit the performance domain - query efficiency, caching, async / blocking calls, payload sizes, scaling assumptions. Run as part of /audit Phase E.
Audit what a normal production app at this scale would have but is absent - health checks, error tracking, feature flags, audit logs, on-call. Run as part of /audit Phase E.
Audit the legal/privacy/compliance signals - privacy policy alignment, data retention, GDPR/CCPA basics, cookie consent, export/delete capabilities. Run as part of /audit Phase E.
Audit the reliability domain - error handling, retries, idempotency, race conditions, partial failures, timeouts, graceful degradation. Run as part of /audit Phase E.
Audit the cost and billing risk domain - runaway loops, unbounded LLM/API calls, missing rate limits on paid services, infrastructure cost amplifiers. Run as part of /audit Phase E.
Audit the dev experience and maintainability - local setup, documentation, test quality, CI feedback time, debuggability. Run as part of /audit Phase E.
Audit the gap between demo-quality and production-quality - hardcoded test data, dev-only configurations shipped, missing prod hardening. Run as part of /audit Phase E.
Generate a detailed AI fix prompt for a specific audit finding. Use when the user asks "/audit-fix F-X.Y" or "give me the fix prompt for finding N." Produces a complete prompt the user can hand to a fresh AI session (Claude Code, Cursor, ChatGPT) to remediate the finding with verification and rollback steps.
Audit code integrity and coherence - duplication, hallucinated references, Tambon signatures, spec drift, Frankenstein patterns, mystery code. THIS DOMAIN incorporates findings from the Tambon hunt and most of the blind-spots walk. Run as part of /audit Phase E.
Multi-LLM debate orchestration (Claude vs Codex vs Gemini) with rounds + synthesis.
Run the methodology setup phase of a technical audit - fingerprint the stack, inventory the repository, and produce a system map. Use this BEFORE running any audit domains. Output: detected stack, repo summary, and high-level architecture in 3-5 paragraphs. No findings yet.
Walk the H1-H9 hard-stops catalog systematically against an audit scope. Use as the FIRST audit phase after audit-method completes. Runs detection commands for each hard stop class and reports FOUND / NOT FOUND with evidence. If any hard stop is found, the audit verdict locks to HARD STOP: DO NOT LAUNCH.
Use after deployments to verify health endpoints, run API smoke tests, and confirm release integrity with CLI checks.
Conduct a full technical due diligence audit of a vibe-coded codebase. Use when the user types `/audit`, asks for a "full audit," asks for a "due diligence review," asks "is this codebase ready to ship-", or asks for a comprehensive code review of an entire repository (not a single PR). Coordinates the 13 domain skills, the hard-stop hunt, the Tambon LLM-signature hunt, and the blind-spot walk. Produces a single dual-layer report (founder view + technical evidence).
Hunt the three Tambon LLM-failure-mode signatures (Hallucinated Object, Wrong Attribute, Silly Mistake) across an audit scope. Use during the audit pipeline AFTER hard-stops and BEFORE blind-spots. Output: counts per signature, top occurrences, density per 1000 LoC. Findings flow into Domain 13 (Code Integrity).
Orchestrate the full audit improvement cycle: audit, roadmap, remediation, rerun, delta comparison, and skill patching for missed findings until convergence or a terminal blocked state.
Use when debugging a bug or incident and you need a strict end-to-end investigation loop from reproduction to verified fix.
Use when implementation is complete and needs validation - tests REAL app features with Playwright, checks console errors, validates CRUD operations, generates persistent HTML reports to /test/ directory
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).
Write a structured session handoff file for clean context transfer. Use when ending a session, before /clear, when context is heavy, when handing off work to another pane, or when asked to "handoff", "checkpoint", "session-transfer", "save state", "prepare for clear".
Hand-tune a project's `.claude/` directory by cherry-picking ECC (everything-claude-code) rules/skills/agents and generating project-specific rules based on the project's real intelligence layers (GitNexus, MemoryMaster, graphify, Obsidian wiki, existing CLAUDE.md). Run preflight intelligence builds for layers that don't exist yet. Run in any project to curate or re-curate its `.claude/` directory. Use when asked to "curate this project", "cherry-pick rules for this repo", "/project-curate", or when a project's `.claude/` is bloated with mass-installed noise or missing stack-specific rules.
Analyze codebase and generate proper CLAUDE.md + AGENTS.md with MemoryMaster integration, real commands, and architecture boundaries. Run in any project to set up or refresh agent instructions. Use when entering a new project, when asked to "set up instructions", "generate CLAUDE.md", "configure this project for agents", or when project instructions are missing/outdated.
Use this skill when adding authentication, handling user input, working with secrets, creating API endpoints, or implementing payment/sensitive features. Provides comprehensive security checklist and patterns.
Audit the current project for stale or broken orchestration setup. Verifies monitoring.md, AGENTS.md/CLAUDE.md, graphify freshness, dead script references, placeholder values still in YAML frontmatter. Read-only - never mutates. Use after /monitoring-setup, after pulling someone else's setup, or when something feels off.
Integrate MercadoPago OAuth payments into web applications. Use when adding MercadoPago payment processing, OAuth account connection, or checkout functionality. Covers Supabase Edge Functions, React/frontend setup, and database schema for multi-tenant SaaS platforms.
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.
Use when an implemented web app, UI feature, workflow, or frontend change needs evidence-based validation in a real browser. Runs interactive discovery, browser execution, screenshot analysis, console/error checks, critical-flow testing, stateful resume, and HTML reporting. Supports CLI-agnostic adapters including agent-browser, playwright-cli, Playwright MCP, Claude/Codex Chrome MCP tools, and browser-harness.
Codify the workflow the user just successfully completed into a reusable skill. Use after a multi-step task finished green (tests pass / build succeeds / user confirmed "done") and the same pattern is likely to recur. Walks the recent conversation, drafts a SKILL.md, asks the user to approve, then writes it to disk. Inspired by gstack/skillify (Garry Tan) - one-shot deterministic codification, not autonomous trace mining.
Use this skill when writing new features, fixing bugs, or refactoring code. Enforces test-driven development with 80%+ coverage including unit, integration, and E2E tests.
Use when diagnosing Supabase issues like RLS, auth, empty queries, webhooks, and performance problems in production apps.
Converts scattered meeting data (recordings, handwritten notes, diagrams, digital notes, supplementary documents) into a structured MS Word deliverable with summary, decisions, action items, open questions, and follow-up message. Use when the user mentions meeting notes, meeting summary, meeting minutes, action items from meeting, meeting documentation, or needs to consolidate meeting materials.
Claude Skills — questions
Answers built from the skills we actually parsed.