Skills
97,501 packaged capabilities and playbooks for agents. Install any of them with one command.
Testing & quality
Apply to multi-step work (sweeps, migrations, runs of similar edits) and to how you stack commits and PRs. Break work into small units that each end in a verifiable state, check each before the next, and order delivery so the sequence proves itself to a reviewer.
Use when running project-wide lint across a Datex Studio codebase (or a scope subset) — checks 5 categories of cross-component issues that per-file validation misses: (1) description-rule conformance across all component files, (2) schema-code alignment between declared `inParams`/`outParams` and the embedded TypeScript that consumes them, (3) cross-component `objectType` reference resolution, (4) OData query pre-flight against connection schemas, (5) result-shape sync between a datasource's query options and its declared `outParams`. Triggers: "validate the project", "run cross-component validation checks", "lint the codebase", "find broken type references", "find OData pre-flight failures", "check the project before merge". For single-file audits use `component-validator` instead.
A strategic advisor for QA, quality, testing, and automation processes. Acts as a senior QA lead. First interviews the user and scans the project to build rich context, then identifies the highest-impact gaps and delivers a prioritized step-by-step roadmap, routing execution to specialized skills. Trigger this skill when the user wants to set up QA from scratch, improve an existing process, set up test automation, decide where to start with testing, or run a QA maturity review — e.g. "set up QA for this project", "improve our QA/testing/automation process", "where should we start with testing", "we have no tests, help", "review our quality process". Also trigger this skill when the user describes problems like "low/poor quality", "missing/a lot of bugs", "no testing".
Generates a structured QA test plan from business-requirements.yaml and technical-requirements.yaml. Produces test suites keyed to functional requirements and user personas, covering positive, negative, edge, security, and performance cases. Outputs qa-test-plan.yaml ready for use in delivery timeline QA phases.
Analyzes test coverage data (Istanbul, c8, lcov) to identify uncovered lines, branches, and functions with risk-weighted gap detection. Use when analyzing coverage reports, identifying coverage gaps, comparing coverage between branches, or prioritizing which untested code to cover first.
React and Next.js performance optimization from Vercel Engineering. Use when building React components, optimizing performance, eliminating waterfalls, reducing bundle size, reviewing code for performance issues, or implementing server/client-side optimizations.
Use when reviewing/judging a finished technical-design document (a TDD / engineering design doc / design RFC) for one feature, deciding whether an engineer can implement it without re-deriving the design. An acceptance gate, not authoring. Judges a single-sourced 11-condition implementability bar: every decision traces bidirectionally to a requirement; scoped to one feature; approach and decomposition implementable, diagram synced to narration; interfaces an api-spec or data-model owns are referenced, not duplicated; one real alternative with a decision criterion; failure modes carry handling; observability signals named; testing covers failures and contract conformance; rollout and rollback with measurable triggers; nothing fabricated; an amend reviewed delta-scoped. Emits exactly `VERDICT: approve|revise` plus findings, and does not false-revise a thin-but-complete TDD. Not for authoring, the upstream PRD or feature-spec, the api-spec or data-model, or generic design docs and RFCs.
Reviews and tunes SAST tool configurations against OWASP ASVS 4.0.3 and CWE Top 25. Auto-invoked when reviewing Semgrep rules, CodeQL queries, SAST CI integration, or false positive triage workflows. Produces a SAST maturity assessment covering rule authoring, severity tuning, custom rule development, and CI integration patterns.
Review Sentry React and TypeScript changes for bug patterns drawn from real production issues. Use when reviewing a frontend diff or PR, checking Warden findings, auditing the current branch, reviewing production-error patterns, or looking for common regressions in `static/`.
Evaluate and benchmark Salesforce skill quality. Compares AI-generated code with vs without skill context, scores against a Salesforce-specific rubric (security, governor limits, bulkification, patterns, completeness), and produces a comparison report. Use to run benchmarks, verify skill value, or check Apex code quality. Activate on mentions of "evaluate skills", "benchmark", "skill quality", "run eval", or "compare with/without skills".
Build a side-by-side before/after HTML page with numbered annotations to show a visual change: UI, chart, diagram, rendered document, image output, or terminal output. Use when the user asks for a before/after, a visual diff, or to see a plan or change visually. PR screenshots belong to before-and-after.
Pre-flight checks, reliability acceptance testing, and format review tooling for Skill creation. Use this skill when a user creates, substantially reworks, or reviews a Skill and needs to identify unstable instruction following, output drift, validation-modality mismatches, or missed constraints in older Skills, or to check Harness contracts, candidate binding evidence, fault injection, directory structure, business flows, and security risks. Do not use for: replacing domain validators, code review, application functional testing, or general programming tasks.
Red-green-refactor discipline for writing code driven by tests. Covers the TDD cycle, test quality, mocking boundaries, and design for testability. Use when: implementing features, fixing bugs, refactoring existing code, designing module interfaces, or working within vertical slices from a tracer-bullets workflow.
Use to diagnose and fix incorrect gradients in differentiable Warp programs. Anything trained, optimized, calibrated, or fit through Warp kernels depends on wp.Tape gradients, so treat any misbehavior of such a workflow as a gradient problem until proven otherwise — use this when training diverges or NaNs, won't train at all, stalls or plateaus above the expected loss, converges to a wrong or biased answer, is worse than a reference implementation, works at small scale but fails at production scale, or fails a QA/validation recheck. Also for explicit symptoms — exploding, NaN/inf, zero, or subtly wrong gradients, suspected wp.Tape/backward issues, gradcheck failures — but users usually describe only the surface symptom ("the sim explodes", "the fit gets dragged toward outliers") without mentioning gradients: make that leap. Not for forward-only Warp work, build/install problems, or autograd issues in other frameworks without Warp.
Comprehensive website content and UX audit. Use this skill whenever someone asks to review, audit, check, or improve a website, landing page, or web product. Triggers include: "audit my website", "review my site", "check my content", "is my website ready to launch", "what's wrong with my site", "content review", "UX review", "website feedback", "check for broken links", "check for lorem ipsum", "review my copy", or any request that involves evaluating an existing website's content quality, UX structure, copy, or readiness for launch. Also trigger when someone shares a URL and asks for feedback, improvements, or a teardown. This skill combines content editing, UX strategy, creative direction, and technical QA into a single structured audit.
EXPERIMENTAL. Use when looking for meaningfully duplicated logic in a codebase, especially duplicate behavior hidden behind different names, different syntax, different control flow, or independently evolved implementations. Not for style issues, not for syntactic clone detection, and not for fixing what it finds.
Run, debug, screenshot, and interact with a Flutter app on a running Android emulator via adb and the qemu console — boot/shut down an AVD, launch `flutter run` in the background, capture compact 360px-wide JPEG screenshots, dump the Android accessibility tree to find addressable widgets, tap by label or coordinates, long-press, swipe, and pinch-zoom (multi-touch). Use this skill whenever the user wants to launch, test, QA, take screenshots of, reproduce a bug in, verify a layout on, or otherwise interact with a Flutter app on Android — even when they don't say "emulator", "adb", or "screenshot" explicitly (e.g. "see what the home screen looks like", "try the new button", "check the layout on Pixel"). Auto-detects the project's applicationId, AVD, and `fvm flutter` vs `flutter`; falls back to ANDROID_EMU_* env vars.
Run isolated eval and grading calls using CC 2.1.81 --bare mode. Constructs claude -p --bare invocations for skill evaluation, trigger testing, and LLM grading without plugin/hook interference. Use when running eval pipelines, grading skill outputs, benchmarking prompt quality, or testing trigger accuracy in isolation.
Debug the signals pipeline locally end-to-end. Covers emitting test signals from fixtures, monitoring Temporal workflows via the REST API, reading sandbox agent logs from object storage, inspecting Docker sandbox containers, and diagnosing common failures (stale ClickHouse embeddings, agentsh network denials, inactivity timeouts). Use when a signal isn't reaching the inbox, a signal-report-summary workflow fails, or a sandbox task run times out.
Run after making Docyrus API changes to catch bugs, performance issues, and code quality problems. Use when implementing or modifying code that uses Docyrus collection hooks (.list, .get, .create, .update, .delete), direct RestApiClient calls, query payloads with filters/calculations/formulas/childQueries/pivots, or TanStack Query integration with Docyrus data sources. Triggers on tasks involving Docyrus API logic, data fetching, mutations, or query payload construction.
CodeStable is an AI coding workflow for serious software engineering that keeps humans in the loop. Rather than centering on Agent orchestration, it organizes requirements, architecture, features, issues, and historical decisions to make development driven by Codex and Claude controllable, traceable, and continuously evolvable.
Generate emulate seed configs for stateful API emulation. Wraps Vercel's emulate tool for GitHub, Vercel, Google OAuth, Slack, Apple Auth, Microsoft Entra, AWS, Okta, Clerk, Resend, Stripe, and MongoDB Atlas APIs — full state machines, not mocks. Use when setting up test environments, CI pipelines, integration tests, or offline development.
Invoke when the user wants to plan, execute, or analyze a structured playtest session with behavioral observation. Covers protocol design, observer guides, and data synthesis. Triggers on: "playtest", "player feedback", "usability test", "observation session", "playtest analysis". Do NOT invoke for QA bug testing (use game-qa-lead) or balance tuning (use game-balance-check). Part of the AlterLab GameForge collection.
Use when a user asks to debug or fix failing GitHub PR checks that run in GitHub Actions. Use `gh` to inspect checks and logs, summarize the failure context, draft a fix plan, and implement only after explicit approval. Treat external providers (for example, Buildkite) as out of scope and report only the details URL.
Audit a repo against Cloudflare/isitagentready signals for AI agent readiness: robots.txt, llms.txt, Markdown negotiation, MCP/API catalogs, OAuth discovery, WebMCP, and browser-agent usability. Includes explicitly scoped readiness audits; not generic SEO or standalone standards documentation questions.
Python linting with Ruff - an extremely fast linter written in Rust. Use when: (1) Standardizing code quality, (2) Fixing style warnings, (3) Enforcing rules in CI, (4) Replacing flake8/isort/pyupgrade/autoflake, (5) Configuring lint rules and suppressions.
(NS) Normative visual appearance guides (card/form/shell/overlay). MUST use for descrição normativa, guia visual, aparência, visual QA checklist, "how this looks", *-visual.md / *-painel-visual.md / *-appearance.md, or shell vs content / card vs form / carcaça vs formulário. Do NOT use for business rules, ERDs, layout math, API/auth, e2e, or ns-proto-creator.
Analyze HM Desktop PRD documents: extract requirement information, verify completeness, check chapter order (requirement source → requirement background → requirement value analysis → competitive analysis → requirement description), check KEP definitions, detect requirement conflicts, and generate a structured analysis report. Use for user requests to: (1) analyze or review PRD documents, (2) extract a KEP list from requirements, (3) check PRD completeness or consistency, (4) map requirements to module architecture, (5) verify PRD format compliance, (6) verify the completeness of the competitive analysis section. Keywords: PRD analysis, requirement extraction, KEP verification, completeness check, chapter order validation, competitive analysis check, analyze PRD, requirement extraction, completeness check, chapter order validation
Autonomous build-phase orchestrator with 8-stage per-slice pipeline (decompose, slice readiness review, implement via TDD pairs, three-stage code review, address feedback, mutation testing, push/CI, merge/flag). Manages slice queue (walking skeleton first, dependency ordering), dispatches TDD pairs with capability detection, enforces 5 binary quality gates with 3-cycle rework budgets and human escalation, maintains audit trail, and respects strict controller boundaries (never writes code, delegates all creative work). Supports three autonomy levels (conservative/standard/full). Use when running factory mode, orchestrating build phases, managing slice queues, or coordinating TDD pairs with review gates. Triggers on: "run the pipeline", "factory mode", "build phase", "slice queue", "quality gates", "autonomous build". NOT for: individual TDD (use tdd), standalone code reviews (use code-review), CI-only workflows (use ci-integration).
Plan, implement, and debug frontend tests: unit/integration/E2E/visual/a11y. Use for Playwright MCP browser automation, Vitest/Jest/RTL, flaky test triage, CI stabilization, and canvas/WebGL games (Phaser) needing deterministic input plus screenshot/state assertions. Trigger: "test", "E2E", "flaky", "visual regression", "Playwright", "game testing".
Design or critique an A/B test or controlled product experiment with statistical rigor — pre-registered falsifiable hypothesis, sample-size/power up front, a full business cycle, one primary metric plus guardrails, segment + long-term reads, and a pre-set iterate/pivot/persevere decision. Catches the defaults Claude misses: peeking/early stopping, tiny samples, no guardrails, vanity metrics, \"it won\" with no learning loop. Trigger when asked to set up or review an A/B test, design an experiment, pick experiment metrics, decide sample size/significance/duration, or interpret/act on test results.
Use this skill when emitting inline review comments, whether posted to a GitHub pull request or written to a local file under caller-declared local-file output. Apply when formatting comments following Bitwarden engineering standards with severity emojis, clear explanations, and actionable suggestions. Use after findings are classified and ready to emit. DO NOT USE when posting summary comments.
Prepares a reproducibility artifact (code/data) for submission and badging. Use when a researcher says "artifact evaluation", "artifact appendix", "reproducibility", "badge", "Artifacts Available/Evaluated/Functional/Reusable", "Results Reproduced/Replicated", "ACM badging", "USENIX/OSDI/SOSP AE", "SIGMOD ARI", "NeurIPS code/checklist", "ACL repro checklist", "Zenodo DOI", "Software Heritage", "anonymize my code/repo", or "package my code". Builds the artifact README + appendix, the dependency/run instructions, an anonymized repo for double-blind, and archival-DOI (Zenodo version vs concept) / Software Heritage SWHID guidance; resolves the ACM badge taxonomy and the Reproduced/Replicated era swap; and lints the artifact directory against the ML Code Completeness checklist with bundled stdlib-only scripts. Outputs an artifact-readiness checklist + packaging plan. Re-verifies the venue current artifact rules live. Advisory only; never submits.
Analyze a pull request's context - title, description, comments, linked issues, attached schemas/images — to understand the original intent, verify scope against the linked ticket, surface ambiguities and edge cases, and produce a structured requirements summary with testable acceptance criteria. Use this skill when the user wants to understand WHAT a PR is supposed to do (not just what code changed), check whether the PR matches its linked ticket/task or generate acceptance criteria from PR/ticket context rather than from code.
Runs autonomous red-green-refactor loops to fix failing tests, reach coverage targets, and satisfy quality gates. Use when tests need to pass, coverage thresholds must be met, quality gates require compliance, or flaky tests need stabilization.
Implement code against an existing Quint specification. Uses Research → Plan → Implement workflow (ACE-FCA style) grounded by the spec as the source of truth. Use when the user wants to refactor code, add a new feature, or close a gap between implementation and spec — with the Quint spec as the formal constraint that all changes must satisfy.
Reviews Red Hat Developer Hub pull request code on GitHub in rhdh, rhdh-operator, rhdh-plugins, or community-plugins. Use for a GitHub PR URL or number, "review this PR", analysis-only review, or posting inline comments. For label and merge-readiness triage of the overlay PR backlog, use /rhdh-overlay.
Follow-up Signals scout for recently merged pull requests. Works out whether each one has deployed, then checks the project's telemetry for whether it did what it claimed, had the impact it expected, or caused a side effect that needs a look.
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.