Cluster a GitHub issue backlog by root cause into a small set of plan-master issues, redirect children with a standardized comment, and bundle architectural-fix PRs that close clusters atomically. Use when an issue tracker has accumulated dozens of reports that share underlying defects, when asked to triage / consolidate / cluster / dedupe issues, when asked to build a plan series or roadmap from open issues, or when routing a new incoming bug into an existing plan.

Run local LLM workloads with Ollama, Open WebUI, and GPU-aware tuning for private development environments. Use when setting up private inference, local AI dev environments, or air-gapped LLM deployments.

When the user wants open source strategy, OSS commercialization, or open source growth. Also use when the user mentions "open source strategy," "OSS strategy," "open source commercialization," "open source to paid," "open core," "COSS," "commercial open source," "GitHub stars strategy," "DevHunt," "open source marketing," "open source growth," "Llama," "Dify," "Cursor," "open source business model," or "developer tools directory." For GitHub tactics, use github.

Query OpenAI API documentation with accurate, up-to-date information. Use this skill proactively when the conversation involves: - OpenAI API usage (Responses API, Chat Completions, models, pricing) - OpenAI SDK (Python, Node.js, Go, Java, C#) - Function calling, structured outputs, tool use - Agents SDK, Agent Builder, ChatKit - Realtime API, WebRTC, WebSocket - Fine-tuning, embeddings, moderation - Image/video/audio generation, speech-to-text, text-to-speech - OpenAI model selection (GPT-4.1, GPT-5, o3, gpt-oss)

Expert-level OpenAPI/Swagger specification for API design, documentation, and code generation

Manage Red Hat OpenShift clusters and deployments. Configure projects, routes, builds, and deploy applications using OpenShift-specific features. Use when working with OpenShift Container Platform or OKD for enterprise Kubernetes.

Generate comprehensive OpenSpec specifications directly from the current project state. Use when the user wants to create or populate main specs by analyzing existing code, documentation, AGENTS.md, GitHub issues, and pull requests — without going through the change/proposal workflow. Ideal for bootstrapping specs on a project that already has working code but no specs yet, or for refreshing specs to match the current implementation.

Capture an architectural decision record — document a decision that was made during work, why it was made, and its consequences.

Help users transform organizations toward modern product practices. Use when someone is trying to shift from feature teams to product teams, introduce empowered teams, modernize legacy processes, or drive cultural change in how product is built.

A skill with an orphan reference file. Do NOT use for real work.

AI image outpainting API — extend an image's borders to fill a target aspect ratio using OpenAI's image edit endpoint. Use this skill when the user says "add outpainting", "extend image", "ai image resize", or "outpaint".

Use the Agent class for multi-step tool loops, conversation history, and reusable LLM agents. Use when building agents with skills, structured output, or stateful conversations.

Audit an existing eval suite for trustworthiness. Use when inheriting evals, suspecting evals miss real failures, or after significant pipeline changes.

Validate LLM judges against human labels using TPR/TNR metrics and train/dev/test splits. Use after writing a judge prompt to verify it agrees with human judgment.

Run cold and warm outreach campaigns to find and engage potential customers or partners. Use when building a prospecting pipeline, writing cold emails or LinkedIn messages, identifying and qualifying leads, planning an outreach strategy, or scaling lead generation as a solopreneur. Covers lead identification, qualification frameworks, cold email writing, LinkedIn outreach, multi-touch sequences, and tracking. Trigger on "cold outreach", "prospecting", "find customers", "cold email", "LinkedIn outreach", "lead generation", "outreach strategy", "build a pipeline", "find clients".

This skill should be used when the user asks to "run OWASP analysis", "check OWASP Top 10", "OWASP scan", or invokes /appsec:owasp. Dispatches 10 category subagents (A01-A10) in parallel for comprehensive OWASP Top 10 coverage.

Implement secure coding practices following OWASP Top 10. Use when preventing security vulnerabilities, implementing authentication, securing APIs, or conducting security reviews. Triggers on OWASP, security, XSS, SQL injection, CSRF, authentication security, secure coding, vulnerability.

Package npm/TypeScript/Bun CLI tools for Nix. Use when creating Nix derivations for JavaScript/TypeScript tools from npm registry or GitHub sources, handling pre-built packages or source builds with dependency management.

When the user wants to optimize, improve, or increase conversions on any marketing page — including homepage, landing pages, pricing pages, feature pages, or blog posts. Also use when the user says "CRO," "conversion rate optimization," "this page isn't converting," "improve conversions," or "why isn't this page working." For signup/registration flows, see signup-flow-cro. For post-signup activation, see onboarding-cro. For forms outside of signup, see form-cro. For popups/modals, see popup-cro.

PancakeSwap — Multi-chain DEX (BNB Chain, ETH, ARB, Base)

Scores a candidate paper against a keyword watchlist and a relevance-criteria document, returning KEEP/DROP/REVIEW with a one-line rationale and a 0-100 relevance score. Combines keyword-match strength, criteria-fit, and a historical-context check (was this paper or its preprint already covered in a recent digest). Domain-neutral - usable for any literature-scan workflow. Use after fetching candidate papers from bioRxiv, medRxiv, or PubMed and before clustering or synthesis. Trigger keywords - paper relevance, filter papers, keep or drop, score papers, relevance rationale.

Decompose a task into independent concerns and execute them through broadly parallel, specialized agent groups. Use when a request involves multiple independent sub-tasks, research across separate domains, or work that can be parallelized across files or modules.

Proactively analyzes Parquet file operations and suggests optimization improvements for compression, encoding, row group sizing, and statistics. Activates when users are reading or writing Parquet files or discussing Parquet performance.

This skill should be used when the user asks to "run PASTA analysis", "PASTA threat model", "risk-centric threat analysis", or invokes /appsec:pasta. Dispatches 7 stages SEQUENTIALLY -- each stage's output feeds the next. This is the ONLY framework that runs sequentially.

Use when tasks involve reading, creating, or reviewing PDF files where rendering and layout matter; prefer visual checks by rendering pages (Poppler) and use Python tools such as `reportlab`, `pdfplumber`, and `pypdf` for generation and extraction.

Use when the user needs PDF generation, manipulation, form filling, table extraction, OCR, merging, splitting, watermarking, or metadata handling. Trigger conditions: generate PDF reports, extract text or tables from PDFs, fill PDF forms programmatically, merge or split PDF files, add watermarks, OCR scanned documents, read or write PDF metadata, convert HTML to PDF.

Define performance budgets for a product, page, flow, or service. Sets targets for latency (p50/p95/p99), throughput, payload sizes (JS/CSS/images), CPU/memory, cold-start, and Core Web Vitals (LCP/INP/CLS). Produces per-component allocation, measurement plan, budget-violation action policy, and degradation strategy.

Web Vitals, Lighthouse CI, bundle optimization, CDN, caching, and load testing

Cross-model performance optimization and scaling configurations for autonomous agents

Use when optimizing performance or reviewing security. Covers code-splitting, React Compiler patterns, asset optimization, a11y testing, and security hardening for React apps.

Source royalty-free images and videos from Pexels API for design, placeholders, or content. Supports search, curated/popular content, collections, multiple resolutions, and ALWAYS creates detailed sidecar metadata files.

Create implementation plans for features and changes. Use when user says "plan how to implement user authentication", "how should I add dark mode to the app", "what's the best way to refactor the database layer", "create a plan for migrating to the new API", or "I need to implement [feature] - help me plan it out".

Plan work before coding: do repo research, analyze options/risks, and ask clarifying questions before proposing an implementation plan. Use when the user asks for a plan, design/approach, scope breakdown, or implementation steps.

Create implementation plan files with dependency analysis and parallel execution support. Breaks down complex features into multiple self-contained plans with proper dependency tracking. Use when user wants to create plans, break down features, or plan implementation.

Show overview of all plans and their current execution status from the file PROGRESS.md located in the plans/ folder. Displays progress bar, completed plans, in-progress plans, failed plans, and suggested next actions. Use when user asks about plan status, progress, what's done, what's left, or wants to see plan overview.

Translates an approved design specification into a sequenced list of TDD implementation tasks, each representing 2–5 minutes of work. Use after the researching skill produces an approved design.md, before writing any implementation code.

Typed library of social media platform image/video dimensions and safe zones. Use this skill when the user says "add platform specs", "social media dimensions", "platform presets", or "aspect ratio presets".

Cognitive load management for players: perception/attention/memory framework, Gestalt principles for game UI, signal-noise discipline on the HUD, onboarding ramp design, and tooltip/affordance patterns. Use when designing UI or HUD layouts, writing tutorials or first-time-user flows, evaluating whether players can read the screen at a glance, debugging 'confusing' feedback, deciding what information goes where (HUD, pause, codex, none), or when playtesters say 'I didn't know I could do that' or 'I missed that this happened.' For deeper accessibility work see accessibility-design.

Persistent browser interaction through a normal Node.js Playwright script for fast iterative web UI debugging.

Master quick plotting and interactive visualization with hvPlot. Use this skill when creating basic plots (line, scatter, bar, histogram, box), visualizing pandas DataFrames with minimal code, adding interactivity and hover tools, composing multiple plots in layouts, or generating publication-quality visualizations rapidly.

Complete guide to Claude Code plugin architecture and development

Lint a project (a Linear Project / GitHub Milestone) against its registered project-type template — checks the type marker is present, the lead is assigned (Linear; skipped on GitHub Milestones), a target date exists, every required section from the manifest is substantive (non-placeholder), milestones are defined, and child issues are scoped. Optional --apply posts findings as a comment (Linear) or appends a versioned marked block to the milestone description (GitHub). Operates on a project, not an individual issue — issue review lives in pm-review. Trigger phrases — "review this project", "review project CHK", "is this project ready", "audit the project", "lint the project", "/pm-project-review".

Use this skill when the user says "sync the team", "who's on the team", "show the roster", "set my skills", "/pm-team", or asks about team members, member skills, or assignee recommendations. Manages the team roster — Linear-synced identity (gitignored cache) paired with a curated checked-in team.md of handle + skills + owns + aliases — merged at read time by handle. The roster is the data source for pm-organize assignee recommendations and pm-notes speaker matching.

Analyzes a .NET 8 codebase and produces analysis findings for upgrading to .NET 10. Uses pm7y-ralph-planner to generate TASKS.md with validation requirements and learnings tracking for autonomous execution. Use when planning a .NET 10 upgrade.

Personal productivity app with Ideas/Tasks, Journal, Habits, Package tracking, Lists, and more via MCP

Kills processes that are listening on specific network ports. Detects the operating system (Windows, macOS, or Linux) and uses the appropriate native commands to find PIDs by port and terminate them. Use this skill whenever the user wants to free up a port, kill something running on a port, stop a server on a port, or mentions that a port is "already in use" or "blocked". Also triggers when the user says things like "something is running on port 3000" or "I can't start my server because port 8080 is taken". Supports multiple ports at once.

Add, remove, and manage page-level and visual-level filters on Power BI PBIR reports using pbi-cli. Invoke this skill whenever the user mentions "filter", "TopN filter", "top 10", "bottom 5", "relative date filter", "last 30 days", "categorical filter", "include values", "exclude values", "clear filters", "slicer filter", "page filter", "visual filter", or wants to restrict which data appears on a page or in a specific visual.

Reference document describing PR lifecycle: pre-flight gates (4 conditions), verification gate (5 conditions), body structure (7 sections), reviewer-suggestion algorithm (CODEOWNERS → file expertise → recent activity → workload balance), and finding-ledger merge prerequisite. Reference only (policy document; consumed by `/flow:pr` and `/flow:merge`).

List PRs awaiting your review, checkout the branch, run linter, and generate a structured code review document.

Use when asked to review a PR with Codex and save, document, post, or update the review results. This skill owns PR comment/cache state, invokes codex-review-pass for the six-subagent review bundle, and publishes one canonical review comment through .pr-review-cache/pr-#.json.

Fetch unresolved PR comments (both code-level and PR-level), validate issues, and fix them. Also checks CI status and fixes failing tests, lint errors, and build issues. Use when reviewing and addressing GitHub PR feedback. Filters out resolved comments, keeps only the last Claude review comment per thread (matching either `claude[bot]` from the upstream Action or the `<!-- claude-local-review -->` sentinel from the local-review agent), validates issues, posts review report as a PR comment, then fixes validated issues.

Extract Intent Layer content from GitHub PR descriptions and review comments. Mines merged PRs for pitfalls, contracts, architecture decisions, and anti-patterns that should be documented in AGENTS.md files.

Generate a Product Requirements Document (PRD) for SDK Bridge autonomous execution. Use when the user explicitly asks to create a PRD, write requirements, generate user stories for sdk-bridge, or scope a feature into a PRD. Also trigger on 'create prd', 'write prd for', 'spec out for sdk-bridge', 'generate requirements document'. Do NOT trigger on general planning, brainstorming, or task decomposition — those belong to other skills.

Implement PRDs/specs with a mandatory precheck review before coding. Use when a user asks to implement a PRD/feature spec/requirements doc or says "implement PRD/spec". Perform a preflight review, raise questions on scope/consistency/risks, then implement after confirmation.

Gathers existing MCP work items, note schemas, and gate requirements to establish a definition floor before planning begins. Queries get_context() for active, blocked, and stalled items, reads .taskorchestrator/config.yaml for schema-defined note requirements per phase, and structures the plan so each task maps to one MCP work item with correct dependency ordering. Use when entering plan mode, starting implementation planning, creating a task breakdown, or beginning project setup for any non-trivial implementation task. Triggered automatically by the plan-mode hook.

When the user wants to create a press coverage page, "As Seen In" section, or media mentions aggregation. Also use when the user mentions "press coverage," "media mentions," "as seen in," "as featured in," "in the news," "press mentions," "media coverage page," or "trusted by publications." For pitching journalists and press releases, use public-relations.

Score a generated CLI against the Steinberger bar, compare two CLIs side-by-side

Prisma ORM for type-safe database access - schema design, migrations, queries, relations, and connection management When user works with Prisma, database schemas, migrations, Prisma Client queries, or mentions prisma commands

This skill should be used when the user asks about "prisma model", "schema.prisma", "prisma relations", "@@index", "prisma attributes", "@id", "@unique", "prisma enum", "prisma types", "add field", "add column", "add model", "new model", "add table", "update model", "change model", "modify schema", "rename field", "foreign key", "prisma index", or mentions schema design, model definitions, or any structural change to a Prisma schema file.

Track user metrics and provide data-driven insights for product decisions. Use when measuring product health, analyzing user behavior, conducting cohort analysis, or optimizing key metrics. Covers acquisition, engagement, retention, revenue metrics, and data-driven decision making.

Generate professional studio or lifestyle backgrounds for product photography that match product aesthetics, brand positioning, and target audience. Use when creating product shoots, fashion photography, or e-commerce visuals requiring contextual backgrounds.

This skill transforms Claude into a product launch specialist following Jeff Walker's Product Launch Formula (PLF) methodology and Frank Kern's Mass Control framework. Use it for seed launches, internal launches, JV launches, and evergreen launches. Covers Pre-Launch Content (PLC), the Sideways Sales Letter, launch sequences, JV recruitment, and post-launch optimization.

Assist with core product management activities including writing PRDs, analyzing features, synthesizing user research, planning roadmaps, and communicating product decisions. Use when you need help with PM documentation, analysis, or planning workflows that integrate with your codebase.

Expert framework for assessing and achieving product-market fit. Combines PMF measurement methodologies, Sean Ellis survey, retention analysis, segment-specific PMF, and post-PMF scaling strategy. Use when determining if product has PMF, measuring user engagement, running PMF surveys, or deciding whether to scale or keep iterating.

When the user wants to create or update their product marketing context document. Also use when the user mentions 'product context,' 'marketing context,' 'set up context,' 'positioning,' or wants to avoid repeating foundational information across marketing tasks. Creates `.claude/product-marketing-context.md` that other marketing skills reference.

Create SEO-optimized pages at scale using programmatic/template-based approaches. Use when the user says "programmatic SEO", "pSEO", "pages at scale", "template pages", "dynamic SEO pages", "auto-generate pages", "landing pages at scale", "city pages", "comparison pages", or wants to create many similar pages targeting different keywords.

Teaches progressive hydration for prioritized client-side interactivity. Use when server-rendered pages have non-critical sections whose JavaScript can be deferred to improve Time to Interactive.

Generate project documentation from codebase analysis — ARCHITECTURE.md, API_ENDPOINTS.md, DATABASE_SCHEMA.md. Reads source code, schema files, routes, and config to produce accurate, structured docs. Use when starting a project, onboarding contributors, or when docs are missing or stale. Triggers: 'generate docs', 'document architecture', 'create api docs', 'document schema', 'project documentation', 'write architecture doc'.

Comprehensive setup steps to help the user create complete project structures in a VS Code workspace; this tool is designed for full project initialization and scaffolding, not for creating individual files. When to use this tool: user wants to create a new complete project from scratch; setting up entire project frameworks (TypeScript projects, React apps, Node.js servers, etc.); initializing Model Context Protocol (MCP) servers with full structure; creating VS Code extensions with proper scaffolding; setting up Next.js, Vite, or other framework-based projects; user asks for "new project", "create a workspace", "set up a [framework] project"; need to establish a complete development environment with dependencies, config files, and folder structure. When NOT to use this tool: creating single files or small code snippets; adding individual files to existing projects; making modifications to existing codebases; user asks to "create a file" or "add a component"; simple code examples or demonstrations; debugging or fixing existing code. This tool provides complete project setup including: folder structure creation; package.json and dependency management; configuration files (tsconfig, eslint, etc.); initial boilerplate code; development environment setup; build and run instructions. Use other file creation tools for individual files within existing projects.

Use when setting up project-specific skills via /cc:setup-project or when user requests custom skills for their codebase - analyzes project to create specialized skills that capture architecture knowledge, coding conventions, testing patterns, and deployment workflows

Scan rough markdown, session exports, or pasted prose and isolate each core intent into a structured prompt atom — classified as law, value, directive, constraint, question, or branch-vector — then rewrite it into machine-readable YAML frontmatter per the RAW → CLEANED/FORMALIZED → ELEVATED doctrine. Triggers on "prompt atom extraction pass", "atomize this", "formalize these prompts", or converting loose operator text into registry-grade atoms.

Design effective prompts for LLM agents with structured input/output formats, chain-of-thought reasoning, few-shot examples, and system prompt architecture. Covers Claude-specific patterns and multi-turn conversation design. Triggers on prompt design, LLM interaction patterns, or system prompt architecture requests.

Meta's 86M prompt injection and jailbreak detector. Filters malicious prompts and third-party data for LLM apps. 99%+ TPR, <1% FPR. Fast (<2ms GPU). Multilingual (8 languages). Deploy with HuggingFace or batch processing for RAG security.

Meta's 86M prompt injection and jailbreak detector. Filters malicious prompts and third-party data for LLM apps. 99%+ TPR, <1% FPR. Fast (<2ms GPU). Multilingual (8 languages). Deploy with HuggingFace or batch processing for RAG security.

Retrospective terseness audit for iteratively-edited prompts and skills. Applies mechanical wording cuts and proposes structural cuts with falsification gates. Invoke when auditing, tightening, compressing, or de-bloating a prompt.

Transforme un prompt en prompt exécutable compatible avec le Task Management System (TaskCreate/TaskUpdate/TaskList)

How Ethereum evolves — EIP lifecycle, fork process, where decisions happen, and how to track upcoming changes. Use when your human asks about upcoming features, when building for future protocol capabilities, or when they want to propose a change. Also use when YOU need to know if a feature exists yet or when it's coming.

Background knowledge for droid-control workflows -- not invoked directly. Capture ground-truth byte sequences from real terminal emulators.

Appwrite Python SDK skill. Use when building server-side Python applications with Appwrite, including Django, Flask, and FastAPI integrations. Covers user management, database/table CRUD, file storage, and functions via API keys.

Python design patterns including KISS, Separation of Concerns, Single Responsibility, and composition over inheritance. Use when making architecture decisions, refactoring code structure, or evaluating when abstractions are appropriate.

Static type checking with Mypy for Python code quality. Use when writing or reviewing Python code to ensure type safety, catch bugs early, and maintain code quality through proper type annotations and Mypy configuration.

Leading deep learning framework. Provides Tensors and Dynamic Computational Graphs with strong GPU acceleration. Widely used for research, neural networks, and differentiable programming.

QA verification skill that tests like a human QA would — navigating the app in a browser, looking at what's on screen, and flagging anything that looks wrong. Use when verifying implemented features match task specifications, testing user flows, or validating acceptance criteria from docs/TASKS.md. Triggers on "verify this PR", "QA this feature", "visual test", "check the UI", "screenshot test", "verify task X". Works against localhost, preview/staging deployments, and production URLs. Auto-selects Google accounts; pauses for user assistance only on credential-based login flows.

Maintain a markdown knowledge base at ~/git/knowledge-base/. Use when the user asks to interact with 'kb' or 'knowledge base' - adding content, deleting obsolete information, searching, or maintaining organization. Triggers on phrases like "add this to the kb", "delete from kb", "update kb", "maintain kb", "search kb", or any knowledge base operation.

Expert in code quality, formatting, linting, and quality gates workflow. Use when user needs to setup quality tools, fix linting errors, configure Biome/Prettier, setup pre-commit hooks, or run quality checks. Examples - "setup code quality", "fix lint errors", "configure Biome", "setup Husky", "run quality checks", "format code", "type check errors".

Converts a PRD markdown file into GitHub Issues (parent + sub-issues) for ralph-github-start-loop to execute. Use when user wants to push PRD stories to GitHub Issues.

Embrace vibe coding for rapid UI exploration. Covers when to iterate vs refine, ephemeral app patterns, and the art of fast, disposable prototyping with AI-assisted development.

Build Raycast extensions with React and TypeScript. Use when the user asks to create a Raycast extension, command, or tool.

React Native patterns for mobile app development with Expo and bare workflow. Trigger: When building mobile apps, working with React Native components, using Expo, React Navigation, or NativeWind.

Use when implementing React 19 features, optimizing components, or choosing between Actions vs TanStack Query for mutations. Covers React Compiler optimization, Server Actions, Forms, and new hooks.

Implements React Server Components (RSC) patterns with Next.js App Router including data fetching, streaming, selective hydration, and server/client composition. Use when users request "server components", "RSC", "Next.js app router", "server-side rendering", or "streaming".

Use React/React Native code as the source of truth and implement the equivalent feature in iOS/Swift. Understands the feature behavior, components, state management, and logic from React, then creates idiomatic iOS code that matches the target codebase's existing patterns. Use when porting features from React/React Native to native iOS or building native alternatives to web components.

Implement real-time features with WebSockets, Server-Sent Events, and long polling. Covers connection management, room-based messaging, presence tracking, and scaling strategies. Triggers on WebSocket implementation, real-time communication, or live update feature requests.

Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear, technically questionable, or comes from multiple reviewers

Formatted digital recipe card generation with ingredient scaling, nutritional information, and organized collections. Use when creating recipe cards, meal plans, grocery lists, or recipe collections.

Execute decomposed fullstack tasks with layer-aware agent routing