Coding skills for Claude
Skills for the everyday parts of programming: reviewing a diff, writing unit tests, decoding error messages, writing READMEs, commit messages and pull request descriptions, and explaining SQL, regular expressions and unfamiliar code in plain language.
10 Skillshelf originals, each with a tested worked example. 156 open-source community skills. Every file is shown before you download.
- Commit MessageWrites Conventional Commits messages from your diff, including the why.
- Bug ReportTurns “it's broken” into a bug report a developer can act on.
- Code ExplainerExplains code line by line for people learning to program.
- Code ReviewReviews a diff for bugs, risks and readability, ranked from most to least serious.
- Error ExplainerExplains an error message or stack trace in plain words, then shows the fix.
- PR DescriptionWrites a pull request description from your diff: what changed, why, and how to test.
- README WriterWrites a clear README from your project's real files: what it does, setup and usage.
- Regex ExplainerExplains what a regex matches, piece by piece, and tests it on real examples.
- SQL HelperWrites SQL from a plain-English question and explains each part of the query.
- Unit Test WriterWrites unit tests for your function, covering normal cases, edge cases and errors.
Community skills
From open-source collections, shared under their original licenses. They don't have a tested Skillshelf example yet, so read their files before installing. See credits.
- Accessibility AuditWCAG 2.2 Accessibility Audit and Remediation Skill
- Adversarial Code ReviewerTough, critical review of a diff or PR before merge, through hostile reviewer…
- Agent DesignerDesign multi-agent systems (supervisor, swarm, pipeline) and tool schemas; find…
- Agent HarnessThe loop — not your optimism — decides when work is done.
- Agent LauncherEvery session starts with a goal — one sentence for one CMA.
- Agent MemoryFor when an agent keeps relearning the same correction, CLAUDE.md has grown…
- Agent Regression Evals"Pass rate went from 82% to 79%" is not a finding until you know whether three points…
- Agent Run ForensicsAgents are confidently wrong about their own history.
- Agent Workflow DesignerDesign production-grade multi-agent workflows with clear pattern choice, handoff…
- AgentHubRun several agents on one task in parallel and keep the best result: faster code…
- AI SecurityAI and LLM security assessment skill for detecting prompt injection, jailbreak…
- Analyze Test Coverage GapsFind what is not tested: map untested pages and flows and coverage gaps.
- API Design ReviewerThe API Design Reviewer skill provides comprehensive analysis and review of API…
- API Test Suite BuilderOutputs ready-to-run test files for Vitest+Supertest (Node) or Pytest+httpx (Python).
- AR Loop (Autoresearch Agent)Start a recurring experiment loop that runs at a user-selected interval.
- AR Resume (Autoresearch Agent)Resume a paused or context-limited experiment.
- AR Run (Autoresearch Agent)Run exactly ONE experiment iteration: review history, decide a change, edit, commit…
- AR Status (Autoresearch Agent)Show autoresearch results and progress.
- Autoresearch AgentMake a file measurably better by trial: faster code, smaller bundle, higher pass rate…
- AWS Solution ArchitectDesign scalable, cost-effective AWS architectures for startups with…
- Azure Cloud ArchitectDesign scalable, cost-effective Azure architectures for startups and enterprises with…
- BeHumanReply like a real person, direct and warm, without lists or hedging.
- Board (Agenthub)Interface for the AgentHub message board.
- Book-to-Skill ConverterTurn written knowledge into an agent skill by extracting structure, not summaries.
- Boost.Asio / standalone AsioWrite async C++ networking code that compiles on the user's Boost, not the newest one.
- Browser AutomationAutomate a browser: scrape websites, fill forms, pull structured data from web pages.
- BrowserStack IntegrationRun Playwright tests on BrowserStack's cloud grid for cross-browser and cross-device…
- Caveman ModeTerse mode: drop filler, articles and pleasantries, keep technical accuracy, about…
- Changelog GeneratorIt separates commit parsing, semantic bump logic, and changelog rendering so teams…
- Chaos EngineeringDesign experiments that surface real weaknesses in production systems — without…
- CI/CD Pipeline BuilderGenerate a CI/CD workflow for a repo from its detected stack: lint, test, build, deploy.
- Claude CoachA coaching layer that runs alongside normal conversations.
- Claude Code Team RolloutOne engineer's Claude Code setup is a preference.
- Cloud SecurityCloud security posture assessment skill for detecting IAM privilege escalation…
- Code ReviewerAutomated code review tools for analyzing pull requests, detecting code quality…
- Code TourCreate CodeTour files — persona-targeted, step-by-step walkthroughs of a codebase…
- Codebase OnboardingExplain an unfamiliar repo: architecture brief and onboarding docs.
- Collab ProofSurfaces AI collaboration evidence the developer didn't consciously record.
- Commit CraftTurn a mix of changes into clean, reviewable commits with good messages, blocking…
- Database DesignerDesign a database schema: tables, relationships, indexes, normalization, SQL vs…
- Database Schema DesignerDesign relational database schemas from requirements and generate migrations…
- Deep LearningStudying or applying Goodfellow, Bengio and Courville's Deep Learning, or asking if…
- Demo VideoNot a slideshow maker.
- Dependency AuditorAudit project dependencies for CVEs and licenses and plan safe upgrades.
- Docker DevelopmentOpinionated Docker workflow that turns bloated Dockerfiles into production-grade…
- Email Template BuilderBuild complete transactional email systems: React Email templates, provider…
- Embedded / IoT MentorAct as an experienced embedded-systems and IoT mentor.
- Engineering Advanced Skills (POWERFUL Tier)Pick the right advanced engineering skill: agents, RAG, MCP, CI/CD, databases…
- Engineering Team SkillsFind which engineering skill fits a task: architecture, frontend, backend, QA…
- Env & Secrets ManagerManage environment-variable hygiene and secrets safety across local development and…
- Epic Design SkillWith this skill you work as a world-class epic design expert.
- Eval (Agenthub)Rank finished AgentHub agents by metric or judge.
- Extract (Self Improving Agent)Transforms a recurring pattern or debugging solution into a standalone, portable…
- Feature Flags ArchitectEnd-to-end discipline for feature flags: classify them, ship them, ramp them, and…
- Fix Failing or Flaky TestsDiagnose and fix a failing or flaky Playwright test.
- Focused FixMaps every file and dependency of the area before changing anything.
- Full Page ScreenshotCapture a full-page screenshot of any web page via Chrome DevTools Protocol.
- GCP Cloud ArchitectDesign, deploy and cut the cost of Google Cloud setups: Cloud Run, GKE, Functions…
- Generate Playwright TestsGenerate production-ready Playwright tests from a user story, URL, component name, or…
- Git Worktree ManagerIt standardizes branch isolation, port allocation, environment sync, and cleanup so…
- Google Workspace CLIExpert guidance and automation for Google Workspace administration using the…
- Grill MeInterview the user relentlessly about a plan or design until reaching shared…
- Grill with DocsInterview me relentlessly about every aspect of this plan until we reach a shared…
- HandoffCompact the current conversation into a handoff document for another agent to pick…
- Helm Chart BuilderOpinionated Helm workflow that turns ad-hoc Kubernetes manifests into maintainable…
- HivemindClaude Code = brain (plans, reviews, merges).
- Hub Init (Agenthub)Initialize an AgentHub collaboration session.
- Hub Run (Agenthub)Run the full AgentHub lifecycle in one command: initialize, capture baseline, spawn…
- Hub Status (Agenthub)Display the current state of an AgentHub session: agent branches, commit counts…
- Human GateMachine verification answers "do the checks pass?" — engineering/agent-harness does that.
- i18n LocalizationA release with a new language usually breaks in the same places: keys added in…
- Incident Commander SkillIncident response framework for availability/reliability incidents (outages…
- Incident ResponseHandle a suspected security breach: triage, SEV1-SEV4 severity, escalation and…
- Initialize Playwright ProjectSet up a production-ready Playwright testing environment.
- Interview System DesignerComprehensive interview loop planning and calibration support for role-based hiring…
- Karpathy CoderCheck a diff for overbuilt code: hidden assumptions, needless abstraction…
- Kubernetes OperatorBuild operators that reconcile correctly.
- LLM Cost OptimizerCuts LLM costs by 40–80% without degrading user-facing quality -- using model…
- LLM WikiInspired by Andrej Karpathy's LLM Wiki pattern (gist).
- Loop LibraryDiscover, find, compare, audit, repair, adapt, and design repeatable AI-agent loops…
- MCP OAuth TriageAn OAuth-backed MCP connector rarely fails at the exchange.
- MCP Server BuilderIt focuses on fast scaffolding, schema quality, validation, and safe evolution.
- Memory EngineeringAdding memory to an agent or when a memory store keeps growing (long context vs RAG…
- Memory Review (Self Improving Agent)Review what the agent has learned in auto-memory: what to promote to permanent rules…
- Memory Status (Self Improving Agent)Health check of agent auto-memory: line counts, topic files, capacity, stale entries…
- Merge (Agenthub)Merge the best agent's branch into the base branch, archive losing branches via git…
- Microsoft 365 Tenant ManagerExpert guidance and automation for Microsoft 365 Global Administrators managing…
- Migrate to PlaywrightInteractive migration from Cypress or Selenium to Playwright with file-by-file…
- Migration ArchitectPlan a migration with zero or minimal downtime: phased plan, compatibility check…
- MinimalistWrite the least code: no over-engineering, fewer dependencies.
- Monorepo NavigatorNavigate, manage, and optimize monorepos.
- Named-Persona Adversarial ReviewCites the source for each persona's critique.
- Observability Designer (POWERFUL)Observability Designer creates production-ready dashboards, alert configurations, and…
- Performance ProfilerIdentifies CPU, memory, and I/O bottlenecks; generates flamegraphs; analyzes bundle…
- Phase 1No API key needed in this phase — the output is a plan.
- Phase 2Turn the build sheet into runnable artifacts, then let the founder launch with their…
- Phase 3This is the plugin's loop: CMA's outcome primitive self-grades the agent's work in an…
- Phase 4A scheduled deployment fires a fresh session on a cron cadence — the agent runs…
- Playwright ProProduction-grade Playwright testing toolkit for AI coding agents.
- PR Review ExpertFile-by-file pull request review for bugs, security and quality, with a merge verdict.
- Profile from CSVSystematicallies assess dataset health, surface hidden issues that corrupt downstream…
- Promote (Self Improving Agent)Moves a proven pattern from Claude's auto-memory into the project's rule system…
- Prompt GovernanceTreats prompts as first-class infrastructure -- versioned, tested, evaluated, and…
- RAG ArchitectDesign a RAG pipeline: chunk size, embedding model, vector database and retrieval…
- Red TeamRed team engagement planning and attack path analysis skill for authorized offensive…
- Red-Green-ProofTurns a suspected bug into a proven one.
- Remember (Self Improving Agent)Save something important to auto-memory with a timestamp and context.
- Review Playwright TestsSystematically review Playwright test files for anti-patterns, missed best practices…
- Runbook GeneratorGenerate operational runbooks quickly from a service name, then customize for…
- Sample Text ProcessorReference BASIC-tier skill used as a fixture by skill-tester.
- Secrets Vault ManagerProduction secret infrastructure management for teams running HashiCorp Vault…
- Security Guidance HookOnce installed, it runs automatically before every Edit, Write, or MultiEdit…
- Security Penetration TestingAuthorized penetration testing of an app or API: vulnerability scans, OWASP Top 10…
- Self-Eval: Honest Work EvaluationGrade finished work honestly on two axes with a devil's-advocate pass against inflation.
- Self-Improving AgentCurate what the agent has learned into durable project knowledge.
- Senior ArchitectSystem architecture decisions with a written rationale: monolith vs microservices…
- Senior Backend EngineerBuild and fix backend services: REST or GraphQL APIs, auth, microservices, PostgreSQL…
- Senior Computer Vision EngineerComputer vision models: object detection, segmentation, training a custom model, and…
- Senior Data EngineerData pipelines: Airflow DAGs, Spark, dbt, Kafka, ETL/ELT, data quality checks.
- Senior Data ScientistDesign experiments and analyze tabular data: A/B tests, causal inference, feature…
- Senior DevopsSet up CI/CD pipelines, infrastructure as code, containers and deployments on AWS…
- Senior FrontendBuild and review React, Next.js, TypeScript or Tailwind frontends: components…
- Senior FullstackFullstack development skill with project scaffolding and code quality analysis tools.
- Senior ML EngineerProduction ML engineering patterns for model deployment, MLOps infrastructure, and…
- Senior Prompt EngineerImprove prompts and measure LLM quality: fix inconsistent prompts, build eval sets…
- Senior QA EngineerWrite and improve tests for React or Next.js apps: Jest, React Testing Library, MSW…
- Senior SecOps EngineerFind and fix application security problems: security reviews, CVE remediation…
- Senior Security EngineerCheck if a design or code is secure: STRIDE threat model, DREAD risk scores, leaked…
- Setup (Autoresearch Agent)Set up a new autoresearch experiment with all required configuration.
- Ship GatePre-production audit that scans a codebase and reports pass/fail/manual across 8…
- Skill RetroRun from the repo being graded.
- Skill Security AuditorScan and audit AI agent skills for security risks before installation.
- Skill TesterMeta-skill that validates, tests, and scores skills in this repository.
- Skillopt SleepSkillOpt-Sleep gives the user's agent a sleep cycle.
- SLO ArchitectDefine SLOs that mean something.
- Smart Test ReportingSummarize Playwright results.
- Snowflake DevelopmentSnowflake SQL, data pipelines, Cortex AI, and Snowpark Python development.
- Spawn (Agenthub)Launch the competing agents in isolated worktrees.
- Spec-Driven WorkflowSpec-driven workflow enforces a single, non-negotiable rule: write the specification…
- Spinning Up in Deep RLLearn or apply deep reinforcement learning with OpenAI Spinning Up: policy gradients…
- SQL Database AssistantWrite, fix and speed up SQL: slow queries, EXPLAIN plans, indexes, ORM migrations…
- Statistical AnalystIs a difference real?
- Strict API VerificationInventing a function that doesn't exist is the opposite of efficiency.
- Stripe Integration ExpertImplement production-grade Stripe integrations: subscriptions with trials and…
- TC TrackerTrack every code change with structured JSON records, an enforced state machine, and…
- TDD GuideTest-first development and test generation for Jest, Pytest, JUnit, Vitest or Mocha…
- Tech Debt TrackerFind, rank and track tech debt in a codebase.
- Technology Stack EvaluatorCompare frameworks or tech stacks before committing: side-by-side scoring, total cost…
- Terraform PatternsOpinionated Terraform workflow that turns sprawling HCL into well-structured, secure…
- TestRail IntegrationBidirectional sync between Playwright tests and TestRail test management.
- Threat DetectionThreat detection skill for proactive discovery of attacker activity through…
- Universal Scraping ArchitectBuild a scraper or extractor for sites, PDFs or paginated APIs and validate the…
- Workflow BuilderOnly leaf agent() calls spend tokens, so the main session stays clean and the whole…
- Wrap-upThe explicit close-out.
- Writing SkillsCreate new agent skills with proper structure, progressive disclosure, and bundled…
- Zero-Hallucination CoderA disciplined, senior engineering partner.