Instruction files agents can follow.

63 skills

• testingclaude-code · codex · cursor

  Browser testing with Chrome DevTools MCP

  Tests in real browsers via Chrome DevTools MCP.

  @addyosmani · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• ciclaude-code · codex · cursor

  CI/CD and automation: set up build and deploy pipelines

  Automates CI/CD pipeline setup.

  @addyosmani · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• reviewclaude-code · codex · cursor

  Multi-axis code review before merge

  Conducts multi-axis code review.

  @addyosmani · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• planningclaude-code · codex · cursor

  Constraint-driven development: stop agents quietly lowering the quality bar

  Establishes a project's quality bar as a written contract and stops agents quietly lowering it.

  @addyosmani · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• agentic-engineeringclaude-code · codex · cursor

  Context engineering: set up agent context for quality output

  Optimizes agent context setup.

  @addyosmani · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• debuggingclaude-code · codex · cursor

  Debugging and error recovery: systematic root cause

  Guides systematic root-cause debugging.

  @addyosmani · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• reviewclaude-code · codex · cursor

  Doubt-driven development: fresh-context adversarial review of decisions

  Subjects every non-trivial decision to a fresh-context adversarial review before it stands.

  @addyosmani · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• gitclaude-code · codex · cursor

  Git workflow and versioning: commits, branches, releases

  Structures git workflow practices.

  @addyosmani · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• agentic-engineeringclaude-code · codex · cursor

  Incremental implementation: thin, verifiable slices

  Delivers changes incrementally in thin, verifiable slices.

  @addyosmani · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• planningclaude-code · codex · cursor

  Planning and task breakdown: spec to ordered tasks

  Breaks work into ordered tasks.

  @addyosmani · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• securityclaude-code · codex · cursor

  Security and hardening: audit inputs, auth and secrets

  Hardens code against vulnerabilities.

  @addyosmani · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• agentic-engineeringclaude-code · codex · cursor

  Source-driven development: ground decisions in official docs

  Grounds every implementation decision in official documentation.

  @addyosmani · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• planningclaude-code · codex · cursor

  Spec-driven development: write the spec before code

  Creates specs before coding.

  @addyosmani · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• testingclaude-code · codex · cursor

  Test-driven development: red-green-refactor loop

  Drives development with tests using the red-green-refactor loop.

  @addyosmani · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• agentic-engineeringclaude-code · codex · cursor

  Agent harness construction: action spaces, tools, observations

  Design and optimize AI agent action spaces, tool definitions, and observation formatting for higher completion rates.

  @affaan-m · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• agentic-engineeringclaude-code · codex · cursor

  Agentic engineering: eval-first, decomposed, cost-aware execution

  Operate as an agentic engineer using eval-first execution, decomposition, and cost-aware model routing.

  @affaan-m · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• testingclaude-code · codex · cursor

  AI regression testing for agent-written code

  Regression testing strategies for AI-assisted development.

  @affaan-m · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• agentic-engineeringclaude-code · codex

  Context budget: check live context use in Codex or Claude Code

  Routes live context checks to the native Codex or Claude Code meter without scanning files, and performs a bounded static component audit only when explicitl…

  @affaan-m · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• mobileclaude-code · codex

  Android skills: the android-cli entry skill from the official Android pack

  Provides instructions for installing and using the `android` CLI.

  @android · Apache-2.0 · Verified Oct 17, 2026

  Evidence not reviewed→
• backendclaude-code · codex · cursor

  Claude API reference: models, params, streaming, tool use

  |-

  @anthropics · Apache-2.0 · Verified Oct 17, 2026

  Evidence not reviewed→
• mcpclaude-code · codex · cursor

  MCP builder: guide to building high-quality MCP servers

  Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools.

  @anthropics · Apache-2.0 · Verified Oct 17, 2026

  Evidence not reviewed→
• skills-authoringclaude-code · codex · cursor

  Skill creator: write, improve and eval new skills

  Create new skills, modify and improve existing skills, and measure skill performance.

  @anthropics · Apache-2.0 · Verified Oct 17, 2026

  Evidence not reviewed→
• agentic-engineeringclaude-code · codex · cursor

  Caveman: short-output skill that cuts agent tokens

  >

  @TheHappyGe · Apache-2.0 · Verified Oct 16, 2026

  Evidence not reviewed→
• backendclaude-code · codex · cursor

  Cloudflare Agents SDK: build, debug and review agents apps

  Build, debug, or review Cloudflare Agents SDK applications using the agents package.

  @cloudflare · Apache-2.0 · Verified Oct 17, 2026

  Evidence not reviewed→
• backendclaude-code · codex · cursor

  Cloudflare Durable Objects: stateful coordination code

  Build, debug, or review Cloudflare Durable Objects code for persistent state and coordination.

  @cloudflare · Apache-2.0 · Verified Oct 17, 2026

  Evidence not reviewed→
• frontendclaude-code · codex · cursor

  Web performance audit: Core Web Vitals and Lighthouse

  Audit, diagnose, or optimize website loading and interaction performance, Core Web Vitals, and Lighthouse performance scores.

  @cloudflare · Apache-2.0 · Verified Oct 17, 2026

  Evidence not reviewed→
• backendclaude-code · codex · cursor

  Cloudflare Workers best practices for production

  Cloudflare Workers best practices for production applications.

  @cloudflare · Apache-2.0 · Verified Oct 17, 2026

  Evidence not reviewed→
• backendclaude-code · codex · cursor

  Wrangler: run and troubleshoot Cloudflare Wrangler CLI

  Run or troubleshoot Wrangler CLI commands in projects that have a wrangler.jsonc, wrangler.json, or wrangler.toml file, including local development, Previews…

  @cloudflare · Apache-2.0 · Verified Oct 17, 2026

  Evidence not reviewed→
• ciclaude-code · codex · cursor

  EAS workflows: write Expo CI/CD workflow YAML

  Helps understand and write EAS workflow YAML files for Expo projects.

  @expo · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• mobileclaude-code · codex · cursor

  Expo SDK upgrade and dependency fixes

  Guidelines for upgrading Expo SDK versions and fixing dependency issues

  @expo · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• gitclaude-code · codex · cursor

  Commit: conventional commit messages (Sentry style)

  Use for every request to commit changes or draft a commit message.

  @getsentry · Apache-2.0 · Verified Oct 17, 2026

  Evidence not reviewed→
• reviewclaude-code · codex · cursor

  Find bugs: bugs, vulns and quality issues in branch changes

  Find bugs, security vulnerabilities, and code quality issues in local branch changes.

  @getsentry · Apache-2.0 · Verified Oct 17, 2026

  Evidence not reviewed→
• securityclaude-code · codex · cursor

  GitHub Actions security review: find workflow exploitation bugs

  GitHub Actions security review for workflow exploitation vulnerabilities.

  @getsentry · Apache-2.0 · Verified Oct 17, 2026

  Evidence not reviewed→
• ciclaude-code · codex · cursor

  Iterate on a PR until CI passes and review feedback is addressed

  Iterate on a PR until actionable CI passes and high/medium review feedback is addressed.

  @getsentry · Apache-2.0 · Verified Oct 17, 2026

  Evidence not reviewed→
• gitclaude-code · codex · cursor

  PR writer: reviewer-facing PR titles and descriptions

  Create or refresh reviewer-facing PR titles and descriptions.

  @getsentry · Apache-2.0 · Verified Oct 17, 2026

  Evidence not reviewed→
• securityclaude-code · codex · cursor

  Skill scanner: scan agent skills for security issues

  Scan agent skills for security issues.

  @getsentry · Apache-2.0 · Verified Oct 17, 2026

  Evidence not reviewed→
• skills-authoringclaude-code · codex · cursor

  Skill writer: create and improve skills per the Agent Skills spec

  Create, synthesize, and iteratively improve agent skills following the Agent Skills specification.

  @getsentry · Apache-2.0 · Verified Oct 17, 2026

  Evidence not reviewed→
• ciclaude-code · codex · cursor

  gh-aw agentic workflows: design, create, debug and upgrade

  Route gh-aw design, creation, diagnosis, patching, active debugging, and upgrade requests to the right strategies.

  @github · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• securityclaude-code · codex · cursor

  Checkout credential review: safe git/gh ops in agentic workflows

  Review code that performs git or gh operations against repository checkouts in gh-aw, checking that the right credentials are available at the right time and…

  @github · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• ciclaude-code · codex · cursor

  Debug gh-aw workflow failures from logs and audits

  Diagnose gh-aw failures using logs and audits; follow the shared strategy for patches and permitted active debug loops.

  @github · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• ciclaude-code · codex · cursor

  Optimize agentic workflow token usage (gh-aw)

  Analyze and reduce token consumption in agentic workflows — guardrail-specific entry points, measurement, and optimization techniques.

  @github · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• ciclaude-code · codex · cursor

  Review agentic workflow changes for correctness and security

  Review agentic workflow changes for correctness, security posture, and optimization opportunities with compile, validation, and audit evidence.

  @github · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• dataclaude-code · codex · cursor

  Hugging Face Hub CLI: models, datasets and Spaces

  Hugging Face Hub CLI (`hf`) for downloading, uploading, and managing models, datasets, spaces, buckets, repos, papers, jobs, and more on the Hugging Face Hub.

  @huggingface · Apache-2.0 · Verified Oct 17, 2026

  Evidence not reviewed→
• dataclaude-code · codex · cursor

  Hugging Face model evals with inspect-ai and lighteval

  Run evaluations for Hugging Face Hub models using inspect-ai and lighteval on local hardware.

  @huggingface · Apache-2.0 · Verified Oct 17, 2026

  Evidence not reviewed→
• reviewclaude-code · codex · cursor

  Code review (Matt Pocock): review changes since a fixed point on two axes

  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 sta…

  @mattpocock · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• debuggingclaude-code · codex · cursor

  Diagnosing bugs: loop for hard bugs and perf regressions

  Diagnosis loop for hard bugs and performance regressions.

  @mattpocock · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• planningclaude-code · codex · cursor

  Grill with docs: interview to sharpen a plan and write ADRs

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

  @mattpocock · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• designclaude-code · codex · cursor

  Improve codebase architecture: find deep-module opportunities

  Scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.

  @mattpocock · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• gitclaude-code · codex · cursor

  PR body writer (Matt Pocock)

  Use when writing a PR body.

  @mattpocock · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• testingclaude-code · codex · cursor

  TDD (Matt Pocock): build features and fix bugs test-first

  Test-driven development.

  @mattpocock · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• planningclaude-code · codex · cursor

  To tickets: turn a plan into tracer-bullet tickets

  Break a plan, spec, or the current conversation into a set of tracer-bullet tickets, each declaring its blocking edges, published to the configured tracker (…

  @mattpocock · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• planningclaude-code · codex · cursor

  Triage: move issues and PRs through a triage state machine

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

  @mattpocock · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• planningclaude-code · codex · cursor

  Wayfinder: plan work bigger than one agent session

  Plan a huge chunk of work (more than one agent session can hold) as a shared map of decision tickets on your issue tracker, and resolve them one at a time un…

  @mattpocock · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• agentic-engineeringclaude-code · codex · cursor

  Superpowers: plan, test, and review skills framework

  You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in s…

  @xppt8a · Verified Oct 16, 2026

  Evidence not reviewed→
• testingclaude-code · codex · cursor

  Playwright skill: browser-test your app from the agent

  Complete browser automation with Playwright.

  @suraj_sharma14 · Verified Oct 16, 2026

  Evidence not reviewed→
• ciclaude-code · codex · cursor

  pstack for Claude Code, Codex, and more

  Watch an open PR — fix failing CI, handle the straightforward review comments, and drive it to a mergeable state.

  @LLMpsycho · Verified Oct 16, 2026

  Evidence not reviewed→
• backendclaude-code · codex · cursor

  Stripe integration best practices

  >-

  @stripe · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• backendclaude-code · codex · cursor

  Upgrade Stripe API versions, SDKs and webhooks

  >-

  @stripe · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• backendclaude-code · codex · cursor

  Supabase: database, auth, storage and edge functions tasks

  Use when doing ANY task involving Supabase.

  @supabase · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• dataclaude-code · codex · cursor

  Postgres best practices (Supabase)

  Postgres best practices maintained by Supabase, for Postgres running anywhere.

  @supabase · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• designclaude-code · codex · cursor

  SVG diagram skill for architecture and flow charts

  SVG diagramming conventions blending a technical look with CJK-friendly typography.

  @Cedric_3L · Verified Oct 16, 2026

  Evidence not reviewed→
• frontendclaude-code · codex · cursor

  React composition patterns that scale (Vercel)

  React composition patterns that scale.

  @vercel-labs · MIT · Verified Oct 17, 2026

  Evidence not reviewed→
• frontendclaude-code · codex · cursor

  React and Next.js performance best practices (Vercel)

  React and Next.js performance optimization guidelines from Vercel Engineering.

  @vercel-labs · MIT · Verified Oct 17, 2026

  Evidence not reviewed→