Skills
97,501 packaged capabilities and playbooks for agents. Install any of them with one command.
Testing & quality
Check a design token set against the Subatomic three-tier architecture (definitions → usage → components) and hand back evidence-tagged findings plus a prioritized work order, or establish a tier ladder from a production codebase that has no tokens yet. Use when the user pastes their token JSON and asks "is this right?", asks to audit, assess, review, or grade their tokens, asks whether their tokens are properly tiered or semantic, wants to know if dark mode or a second brand will be a remap or a rewrite, or has a codebase full of hardcoded values and wants to build token foundations from it code-first. Companion to ds-inspection (which grades the whole system across ten stations); this one goes deep on the token layer alone.
Lint a Figma node tree for WCAG 2.2 accessibility AND design-system quality — contrast, target size, focus indicators, color-only signaling, hardcoded colors, missing text styles, detached components, and more. Use when the user wants a design-side audit BEFORE handoff: triggers 'lint this frame', 'check accessibility', 'WCAG audit', 'is my design accessible', 'find contrast issues', 'check color contrast', 'find hardcoded colors', 'are my components using tokens', 'design QA', 'a11y check on this page', 'audit my design for issues'. Walks the tree from a node (or the whole page) and returns findings with severity (critical/warning/info) and WCAG level (A/AA/AAA/best-practice). The native Figma MCP has NO design-side linting — this is unique to this collection. For the deep per-component a11y scorecard use figma-audit-accessibility; for CODE-side (HTML) checks use figma-scan-code-accessibility
Run preflight and postflight checklists against UX flows before you design and before you ship. Covers onboarding, forms, pricing, dashboards, and empty states. Returns structured findings with severity and smell linkage. Requires uxuiprinciples.com API Access (pro tier).
Apply whenever you write or review a command that decides whether work is done: a test invocation, an acceptance check, a CI gate, a shell assertion, a grep-based verify step, or an upgrade, deployment, migration, or install whose success you will report. Names the failure signal, grounds paths, and catches the shapes that pass green while measuring nothing.
Alibaba Cloud PTS (Performance Testing Service) scenario-based skill for creating and managing stress testing scenarios. Supports both PTS native HTTP/HTTPS stress testing and JMeter-based stress testing. Triggers: "PTS", "压测", "性能测试", "stress testing", "performance testing", "JMeter", "load testing", "创建压测场景"
Check a docs-content page against Elastic content type guidelines (overview, how-to, tutorial, troubleshooting, changelog), or classify a proposed page idea against the content types before drafting. Use when the user asks to check content type compliance, validate page structure, review a doc against content type standards, or decide which content type a planned page should use
Review Express.js security audit patterns for middleware and routes. Use for auditing Helmet.js, CORS, body-parser limits, and auth middleware. Use proactively when reviewing Express.js apps. Examples: - user: "Secure my Express app" → add Helmet.js and disable x-powered-by - user: "Check Express CORS config" → verify origin allowlists and credentials - user: "Review Express auth middleware" → check route order and coverage - user: "Scan for Express path traversal" → verify path normalization and validation - user: "Audit Express session config" → check secure, httpOnly, and sameSite flags
Use this skill when auditing SEO, analyzing Core Web Vitals, verifying meta tags, checking structured data, or troubleshooting analytics implementations. Covers Google Search Console, Google Analytics 4, Cloudflare Web Analytics, and technical SEO best practices.
Tools for creating, auditing, and maintaining Claude Code skills. Includes /create-skill for scaffolding, /review-skill for quality checks, and /audit commands for bulk verification. Use when building new skills, maintaining skill quality, or forking the claude-skills repo.
Run an autonomous AI penetration test with Strix against a codebase, repository, URL, domain, or IP — either self-hosted with the open-source CLI or via the managed app.strix.ai cloud API — and read the validated findings (Markdown, JSON, CSV, SARIF, PoCs). Use when the user asks to pentest, security-scan, or find vulnerabilities in an app, API, website, or repo with Strix.
Flaky test fixes: detect nondeterministic tests empirically (repeat/shuffle/parallel runs), diagnose the root cause, fix it — never retry/skip/sleep — and verify across many randomized runs. Triggers on “absolute deflake,” “fix flaky tests,” “CI is flaky,” “this test fails randomly/intermittently.”
Fit a Datadog LLM-Obs evaluator to human labels. Takes an annotation queue, works out where in the trace the labelled property actually lives, drafts an LLM-judge that predicts the human label, scores that judge against the already-labelled rows with a metric agreed with the user, then hill-climbs it — inspect the errors, make one focused change, re-score, keep it only if it beats the best — for a bounded number of iterations, and finally publishes the winner to Datadog as a DISABLED evaluator (not a Datadog draft — a real evaluator with `enabled: false`). Use when the user says "build an eval from my annotations", "build an evaluator from the annotation queue", "turn my annotations into an evaluator", "learn an evaluator from my labels", "fit a judge to the annotation queue", "auto-label", "auto labelling", "automate this annotation queue", "scale up my human labels", or wants the rest of a queue graded the way the humans graded the first rows. Needs an annotation queue with at least two classes present in the human labels (e.g. one true and one false for a boolean).
World-class character and art style consistency for AI-generated images and videos - ensures visual coherence across series, maintains character identity, and provides rigorous QA before deliveryUse when "character consistency, art style, same character, consistent character, visual continuity, series, turnaround sheet, character sheet, reference image, character bible, style guide, anime character, consistent look, face consistency, outfit consistency, lora training, ip-adapter, flux kontext, visual qa, art quality, generation review, style drift, character drift, character-consistency, art-style, visual-qa, ai-art, image-generation, video-generation, anime, illustration, lora, ip-adapter, flux, midjourney, stable-diffusion" mentioned.
Use this skill to run an end-to-end Jetpack Compose performance audit when the symptom is broad ("the app feels sluggish", "scroll is rough everywhere", "we're starting a perf sprint", "what should we fix first?"). Orchestrates the four-phase Measure → Diagnose → Fix → Verify loop by sequencing the 25 focused skills (release-mode setup, R8, Baseline Profiles, Compose Compiler reports, stability inference, Layout Inspector, `@TraceRecomposition`, stabilization, strong skipping, phase-deferral, derivedStateOf, lazy layouts, lazy prefetch, Modifier.Node, modifier ordering, flow collection, effects, CI gates, hot-reload) and produces a written audit report with Before/After Macrobenchmark numbers. Use when the developer wants a perf sprint kickoff, a pre-release perf gate, onboarding to a perf-troubled codebase, or a written deliverable. Use when the user mentions "audit", "perf review", "perf sprint", "where do I start", or has no specific symptom yet.
Use when writing, changing or reviewing any Bagisto PHP or Blade — the conventions this codebase holds to, covering Laravel idiom, code style, comments and docblocks, database access, Blade, security and localization. Trigger phrases include "standards", "conventions", "code style", "docblock", "comments", "repository pattern", "blade", "component", "binding", "event", "migration", "security", "XSS", "authorization", "escaping", "is this safe", "best practice".
Applies behavior-driven development principles including Gherkin scenarios and test-driven development. This skill should be used when the user asks to implement features, fix bugs, or when writing executable specifications and tests before writing production code.
Verify Kibana Lens chart UI settings against the source code. Use when documenting chart types, reviewing chart settings accuracy, or checking UI labels and rendering order for Lens visualizations (bar, line, area, pie, treemap, mosaic, waffle, tag cloud, heat map, gauge, metric, region map, table).
Create, audit, repair, and optimize Agent Skills. Use when the user wants to create a Skill, check Skill quality, restore the original version or fix a specific error, compare old and new results, or update the upstream methods this Skill draws on. Triggers: “create skill,” “write a skill,” “change skill,” “optimize skill,” “score skill,” “skill checkup,” “review skill.”
Evaluate Omni AI accuracy using Omni's built-in eval system — define a prompt set, run a judged eval against a model (or branch), and read the accuracy-judge verdicts. Use this skill whenever someone wants to evaluate Omni AI, benchmark Blobby, run regression tests, compare AI output across branches or model-context changes, measure AI quality, run A/B tests on model changes, assess the impact of an ai_context or modeling change, or any variant of “run evals”, “test Blobby”, “benchmark query generation”, “compare AI results”, “regression test”, “how accurate is the AI”, or “measure the impact of my changes”.
Read a feature branch's merge-readiness from its actual git data — measured metrics, not vibes: collect raw git output for the branch vs base, compute branch-health metrics (size class, test ratio, focus, drift, sessions, hotspots), scan the diff for pre-PR artifacts (secrets, `.only`, conflict markers, debug lines, TODOs), then render a GREEN / YELLOW / RED verdict with per-signal recommendations — runs as a forked analysis workflow with its own reasoning budget, isolated from the main flow. Every metric is derived from concrete git data, never estimated; the verdict follows the documented signal thresholds and never reports GREEN while a blocking finding stands; read-only apart from the optional JSON snapshot, never touching code or git state. Use when the user asks for branch health, pre-PR analysis, or `/pr-retro`.
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.
Create an application-specific production parity skill by inspecting an app's docs, source, tests, CI, deployment, infrastructure, config, auth, and environment setup, then asking targeted harness questions only for source-unanswerable decisions. Use when local, CI, PR, preview, staging, or other non-production environments may drift from production behavior; when production-only auth, config, identity-provider groups, feature flags, infrastructure, backing services, or policy differences caused bugs; or when a team wants a reusable skill that detects, documents, tests, and helps fix parity drift for one specific application.
Validate Three.js WebGPU/TSL implementations against falsifiable claims. Use for visual or mechanism correctness, temporal behavior, target performance or GPU attribution, resource ownership, and lifecycle stability.
Lint and typecheck debt paydown: clear pre-existing repo-wide lint/type violations and suppressions (@ts-ignore, # type: ignore) one rule per wave, fixing causes not symptoms. Runs on green main. For diff-scoped quality use absolute-simplify. Triggers on “absolute debt”, “fix our lint warnings”, “clear the type errors”, “get to strict mode”, “burn down suppressions”.
Audit websites for Google AdSense application readiness and ad-serving compliance. Use when checking whether a site is likely to satisfy AdSense eligibility, site ownership, content quality, navigation, crawler access, ads.txt, privacy disclosure, Google Publisher Policies, AdSense Program policies, or when the user asks if a site can apply for AdSense, pass AdSense review, show ads, or fix AdSense rejection/site-not-ready issues.
Capture a user-reported defect as a durable GitHub issue written in the project's own domain language. Explores the codebase in parallel for context but never leaks file paths or line numbers into the issue. Use when the user reports a bug conversationally, runs a QA pass, or says "file an issue", "log this as a bug", "capture this".
Analyze a software codebase for algorithmic complexity and performance hotspots, then propose or implement safe optimizations without breaking behavior. Use when the user asks to scan many files, find inefficient loops, nested iteration, repeated scans, costly rendering/recomputation, N+1 queries, avoidable O(n^2) or O(n) operations, or reduce complexity such as O(n^2) to O(n log n) / O(n), while preserving tests, APIs, outputs, and maintainability.
Audit designs, color palettes, and UI components for WCAG 2.1 accessibility compliance, providing specific fixes for contrast, touch targets, focus indicators, and screen reader support. NOT for color harmony aesthetics or design system creation.
Use when creating a new CodeceptJS 4 test, extending an existing Scenario, or porting a manual test plan to code. Builds tests against the live browser via the CodeceptJS MCP server — never from imagined locators or flows. Trigger on any request to create, write, add, draft, or scaffold a CodeceptJS test, login flow, end-to-end check, or "test from scratch".
Build WCAG 2.2 AA compliant interfaces with semantic HTML, ARIA, keyboard navigation, focus management, color contrast, and screen reader support. Covers forms, dialogs, tabs, live regions, skip links, alt text, and data tables. Use when implementing accessible UIs, auditing WCAG compliance, fixing screen reader issues, keyboard navigation, focus traps, or troubleshooting "focus outline missing", "aria-label required", "insufficient contrast", "missing alt text", "heading hierarchy"
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.