Skill library
Muster bundles a library of skills in plugin/builtins/. They stay out of Claude Code's skill listing, so they cost nothing until a task uses them. The planner binds them to tasks by id (the skills attribute in PLAN.md), and the builder reads each bound SKILL.md by path before it works.
Use /muster:skills (or the muster skills CLI) to browse and opt in:
| Verb | What it does |
|---|---|
list [domain] | Builtins grouped by catalog role, with upstream and license. |
show <id> | Description, source, license, catalog roles, file path, and whether this project enabled or pinned it. |
search <query> | Builtins first, then the ui-skills.com registry. The registry is fetched only on search, with a 5-second timeout, and cached for 24 hours in the user cache directory. Offline, you get builtin results and a note. |
enable <id> | Copy a builtin into this project's .claude/skills/<id>/, so Claude Code offers it natively in this project only. The copy keeps its license and upstream, and it never overwrites an existing directory. |
disable <id> | Remove the files enable or add wrote, as listed in .claude/skills/.muster-enabled.json, and only while they are unmodified. |
add <publisher>/<skill> | Install a registry skill into .claude/skills/<skill>/. It checks the publisher repository's license through the GitHub API first and refuses repositories with no license, NonCommercial, NoDerivatives or proprietary terms. |
pin <id> [--role <role>] / unpin <id> | Project pins in .muster/config (see Configuration). Build-args binds a pinned skill to every matching task. |
usage | Which skills the runs in .muster/runs/ bound, applied, missed or never used. |
When no builtin matches an outcome strongly, preflight may suggest up to three registry skills from the cached registry. It never fetches during preflight. Each suggestion is marked as not installed.
Builtins
Generated from catalog/ and each builtin's upstream: frontmatter (vendor/manifest.yaml is the source of truth, see Credits). Builtins are grouped by their first catalog role. A builtin with no catalog entry is listed as unrouted: workflows read it by path, but the planner never binds it.
183 builtins: 164 in the catalog (preflight proposes them by role), 19 unrouted (read by path only).
docs-research (1)
| Skill | What it does | Source | License |
|---|---|---|---|
addy-source-driven-development | Ground framework-specific code in current official documentation: detect versions, fetch the docs, implement the documented pattern and cite each source. | addyosmani/agent-skills (adapted) | MIT |
brainstorm (2)
| Skill | What it does | Source | License |
|---|---|---|---|
intent-philosopher | Expansive design thinking: open exploration of a problem before solving it. | ghaida/intent (adapted) | CC0-1.0 |
sp-brainstorm | You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in s... | obra/superpowers (adapted) | MIT |
plan (6)
| Skill | What it does | Source | License |
|---|---|---|---|
intent-blueprint | Service design: service blueprints, ecosystem maps and the process and data dependencies behind an experience. | ghaida/intent (adapted) | CC0-1.0 |
intent-intent | Intent entry point: sets UX project context and principles and routes to the intent-* skills; carries UX foundation references. | ghaida/intent (adapted) | CC0-1.0 |
intent-specify | Design-to-engineering specs: component, interaction and state specifications for handoff. | ghaida/intent (adapted) | CC0-1.0 |
intent-strategize | UX strategy: problem framing, opportunity sizing, hypotheses and project scope. | ghaida/intent (adapted) | CC0-1.0 |
mp-domain-modeling | Sharpen a project's domain language as you design: challenge terms, keep GLOSSARY.md current, and record hard-to-reverse decisions as ADRs. | mattpocock/skills (adapted) | MIT |
sp-parallel | Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | obra/superpowers (adapted) | MIT |
implement (14)
| Skill | What it does | Source | License |
|---|---|---|---|
go-error-handling | Idiomatic Go error handling: creation, %w wrapping, errors.Is/As, sentinel and custom types, the single-handling rule and structured slog logging. | samber/cc-skills-golang (vendored) | MIT |
go-safety | Defensive Go against accidental bugs: nil panics, typed-nil interfaces, append aliasing, integer truncation, float comparison, defer in loops and safe zero v... | samber/cc-skills-golang (vendored) | MIT |
supabase-postgres-best-practices | Supabase's Postgres rules for schema design, indexes, RLS, connection pooling, locking, migrations and query diagnosis, for Postgres running anywhere. | supabase/agent-skills (vendored) | MIT |
wsh-api-design-principles | Master REST and GraphQL API design principles to build intuitive, scalable, and maintainable APIs that delight developers. Use when designing new APIs, revie... | wshobson/agents (vendored) | MIT |
wsh-architecture-decision-records | Write and maintain Architecture Decision Records (ADRs) following best practices for technical decision documentation. Use when documenting significant techn... | wshobson/agents (vendored) | MIT |
wsh-architecture-patterns | Implement proven backend architecture patterns including Clean Architecture, Hexagonal Architecture, and Domain-Driven Design. Use this skill when designing ... | wshobson/agents (vendored) | MIT |
wsh-async-python-patterns | Master Python asyncio, concurrent programming, and async/await patterns for high-performance applications. Use when building async APIs, concurrent systems, ... | wshobson/agents (vendored) | MIT |
wsh-debugging-strategies | Master systematic debugging techniques, profiling tools, and root cause analysis to efficiently track down bugs across any codebase or technology stack. Use ... | wshobson/agents (vendored) | MIT |
wsh-go-concurrency-patterns | Master Go concurrency with goroutines, channels, sync primitives, and context. Use when building concurrent Go applications, implementing worker pools, or de... | wshobson/agents (adapted) | MIT |
wsh-microservices-patterns | Design microservices architectures with service boundaries, event-driven communication, and resilience patterns. Use when building distributed systems, decom... | wshobson/agents (vendored) | MIT |
wsh-nodejs-backend-patterns | Build production-ready Node.js backend services with Express/Fastify, implementing middleware patterns, error handling, authentication, database integration,... | wshobson/agents (adapted) | MIT |
wsh-openapi-spec-generation | Generate and maintain OpenAPI 3.1 specifications from code, design-first specs, and validation patterns. Use when creating API documentation, generating SDKs... | wshobson/agents (adapted) | MIT |
wsh-python-design-patterns | Python design patterns including KISS, Separation of Concerns, Single Responsibility, and composition over inheritance. Use this skill when designing a new s... | wshobson/agents (adapted) | MIT |
wsh-typescript-advanced-types | Master TypeScript's advanced type system including generics, conditional types, mapped types, template literals, and utility types for building type-safe app... | wshobson/agents (vendored) | MIT |
code-review (4)
| Skill | What it does | Source | License |
|---|---|---|---|
sp-review | Use when completing tasks, implementing major features, or before merging to verify work meets requirements | obra/superpowers (adapted) | MIT |
sp-review-recv | Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technic... | obra/superpowers (vendored) | MIT |
wsh-code-review-excellence | Master effective code review practices to provide constructive feedback, catch bugs early, and foster knowledge sharing while maintaining team morale. Use wh... | wshobson/agents (vendored) | MIT |
wsh-multi-reviewer-patterns | Coordinate parallel code reviews across multiple quality dimensions with finding deduplication, severity calibration, and consolidated reporting. Use this sk... | wshobson/agents (vendored) | MIT |
security-review (12)
| Skill | What it does | Source | License |
|---|---|---|---|
go-security | Go security review: injection, cryptography, path traversal, SSRF, cookies, secrets, PII in logs, threat modeling, gosec and govulncheck. | samber/cc-skills-golang (vendored) | MIT |
openai-security-threat-model | Repository-grounded AppSec threat model: trust boundaries, assets, attacker capabilities, ranked abuse paths and mitigations, written as Markdown. | openai/skills (adapted) | Apache-2.0 |
sentry-gha-security-review | Exploitation-focused GitHub Actions workflow review: pwn requests, expression injection, credential escalation, unpinned actions and CI prompt injection. | getsentry/skills (adapted) | Apache-2.0 |
sentry-security-review | High-confidence security code review: trace attacker-controlled input before flagging injection, XSS, auth, crypto, SSRF or deserialization findings. | getsentry/skills (adapted) | CC-BY-SA-4.0 |
tob-fp-check | Verify suspected security findings to a TRUE or FALSE POSITIVE verdict with data-flow, exploitability and PoC evidence; not for hunting bugs. | trailofbits/skills (adapted) | CC-BY-SA-4.0 |
tob-supply-chain-risk-auditor | Script-measured dependency supply-chain audit for npm, PyPI and Go: advisories, abandoned upstreams, publisher concentration and install scripts. | trailofbits/skills (adapted) | CC-BY-SA-4.0 |
wsh-attack-tree-construction | Build comprehensive attack trees to visualize threat paths. Use when mapping attack scenarios, identifying defense gaps, or communicating security risks to s... | wshobson/agents (adapted) | MIT |
wsh-sast-configuration | Configure Static Application Security Testing (SAST) tools for automated vulnerability detection in application code. Use when setting up security scanning, ... | wshobson/agents (vendored) | MIT |
wsh-security-requirement-extraction | Derive security requirements from threat models and business context. Use when translating threats into actionable requirements, creating security user stori... | wshobson/agents (adapted) | MIT |
wsh-signed-audit-trails-recipe | Step-by-step cookbook for setting up cryptographically signed audit trails on Claude Code tool calls. Use when explaining, evaluating, or demonstrating the p... | wshobson/agents (adapted) | MIT |
wsh-stride-analysis-patterns | Apply STRIDE methodology to systematically identify threats. Use when analyzing system security, conducting threat modeling sessions, or creating security do... | wshobson/agents (adapted) | MIT |
wsh-threat-mitigation-mapping | Map identified threats to appropriate security controls and mitigations. Use when prioritizing security investments, creating remediation plans, or validatin... | wshobson/agents (vendored) | MIT |
test-author (7)
| Skill | What it does | Source | License |
|---|---|---|---|
go-testing | Production Go tests: table-driven and parallel tests, testify, fuzzing, fixtures, goleak, coverage, integration tests and flaky-test diagnosis. | samber/cc-skills-golang (vendored) | MIT |
sp-tdd | Use when implementing any feature or bugfix, before writing implementation code | obra/superpowers (vendored) | MIT |
sp-verify | Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output... | obra/superpowers (vendored) | MIT |
tob-property-based-testing | Write, review and triage property-based tests (Hypothesis, fast-check, proptest, Echidna): roundtrip, invariant and oracle properties. | trailofbits/skills (adapted) | CC-BY-SA-4.0 |
wsh-e2e-testing-patterns | Master end-to-end testing with Playwright and Cypress to build reliable test suites that catch bugs, improve confidence, and enable fast deployment. Use when... | wshobson/agents (vendored) | MIT |
wsh-javascript-testing-patterns | Implement comprehensive testing strategies using Jest, Vitest, and Testing Library for unit tests, integration tests, and end-to-end testing with mocking, fi... | wshobson/agents (vendored) | MIT |
wsh-python-testing-patterns | Implement comprehensive testing strategies with pytest, fixtures, mocking, and test-driven development. Use when writing Python tests, setting up test suites... | wshobson/agents (vendored) | MIT |
refactor (6)
| Skill | What it does | Source | License |
|---|---|---|---|
mp-codebase-design | Deep-module design vocabulary (module, interface, depth, seam, adapter, leverage, locality) for designing or reshaping a module's interface. | mattpocock/skills (adapted) | MIT |
mp-improve-codebase-architecture | Opt-in architecture review: find deepening opportunities in recently changed code, report them as HTML, then grill the chosen one. | mattpocock/skills (adapted) | MIT |
splab-finding-duplicate-functions | Find semantic duplicate functions (same intent, different names or code) by extracting a function catalog, clustering it by domain, then judging each cluster. | obra/superpowers-lab (adapted) | MIT |
wsh-dependency-upgrade | Manage major dependency version upgrades with compatibility analysis, staged rollout, and comprehensive testing. Use when upgrading framework versions, updat... | wshobson/agents (vendored) | MIT |
wsh-python-anti-patterns | Use this skill when reviewing Python code for common anti-patterns to avoid. Use as a checklist when reviewing code, before finalizing implementations, or wh... | wshobson/agents (vendored) | MIT |
wsh-react-modernization | Upgrade React applications to latest versions, migrate from class components to hooks, and adopt concurrent features. Use when modernizing React codebases, m... | wshobson/agents (vendored) | MIT |
frontend (75)
| Skill | What it does | Source | License |
|---|---|---|---|
addy-browser-testing-with-devtools | Verify UI in a real browser through the Chrome DevTools MCP: DOM, console errors, network requests, performance traces and screenshots. | addyosmani/agent-skills (vendored) | MIT |
addy-frontend-ui-engineering | Frontend engineering discipline: production-quality components, state, responsive layout and WCAG-ready markup without the generic AI look. | addyosmani/agent-skills (adapted) | MIT |
anthropic-canvas-design | Create original static visual art (posters, PNG or PDF canvases) from a written design philosophy; not for product UI. | anthropics/skills (adapted) | Apache-2.0 |
anthropic-frontend-design | Supporting visual-direction pass (Anthropic): commit to a distinctive, subject-grounded aesthetic for palette, type and hero instead of templated defaults. | anthropics/skills (adapted) | Apache-2.0 |
elaya-figma-designer | Design inside Figma through the Figma MCP with real variables, text styles, auto layout, components and token rules. | elayadesign/ai-design-skills (vendored) | MIT |
elaya-landing-page-design | Landing page system: intake, section structure, conversion copy, SEO and strict visual rules for high-converting marketing pages. | elayadesign/ai-design-skills (vendored) | MIT |
emil-animate | Build a web animation from scratch: whether to animate, purpose, tool, properties, easing, duration, interruption and exit. | emilkowalski/skills (vendored) | MIT |
emil-animate-expo | Build React Native and Expo animations with Reanimated, Gesture Handler, Expo Router and haptics, tuned to feel right on device. | emilkowalski/skills (vendored) | MIT |
emil-animation-performance | Keep web animations in the frame budget: composite-only properties, CSS vs WAAPI vs rAF, will-change, and diagnosing jank. | emilkowalski/skills (vendored) | MIT |
emil-animation-vocabulary | Glossary that names a vaguely described motion effect (pop in, rubber-banding, stagger) so it can be prompted precisely. | emilkowalski/skills (vendored) | MIT |
emil-apple-design | Apple-style fluid interfaces for the web: springs, gestures, interruptible transitions, materials, depth and optical typography. | emilkowalski/skills (vendored) | MIT |
emil-break-ui | Stress-test a UI with worst-case data (long names, empty, huge counts, non-Latin text) behind a demo toggle and report every break. | emilkowalski/skills (vendored) | MIT |
emil-choreograph | Choreograph a multi-element animation: staggered lists, sequencing, overlap, crossfades, and the hero or list entrance. | emilkowalski/skills (vendored) | MIT |
emil-design-eng | Emil Kowalski's design-engineering craft: the invisible details of UI polish, component feel and when motion earns its place. | emilkowalski/skills (vendored) | MIT |
emil-find-animation-opportunities | Read-only scan for places that should animate, and the many that should not, proposing motion with exact values. | emilkowalski/skills (vendored) | MIT |
emil-improve-animations | Read-only motion audit of a codebase: prioritized animation findings plus self-contained fix plans for other agents to execute. | emilkowalski/skills (vendored) | MIT |
emil-mobile-native | Make a mobile web app feel native: tap highlight, sticky hover, 100vh, input zoom, safe areas, overscroll and touch details. | emilkowalski/skills (vendored) | MIT |
emil-pick-ui-library | Pick an opinionated frontend library for a task (charts, OTP, command menu, toasts, drag and drop); runs only when explicitly invoked. | emilkowalski/skills (vendored) | MIT |
emil-prototype | Prototype several genuinely different versions of a UI piece behind a live picker; runs only when explicitly invoked. | emilkowalski/skills (vendored) | MIT |
emil-remove-animations | Subtract motion an interface does not need (keyboard actions, high-frequency surfaces, decoration) when it feels slow or over-animated. | emilkowalski/skills (vendored) | MIT |
emil-review-animations | Strict review of animation and motion code against Emil Kowalski's craft bar: easing, duration, springs, interruption, reduced motion. | emilkowalski/skills (vendored) | MIT |
gsap-core | Official GSAP core API: gsap.to, from and fromTo, easing, stagger, defaults and matchMedia for responsive or reduced-motion animation. | greensock/gsap-skills (vendored) | MIT |
gsap-frameworks | Official GSAP in Vue, Nuxt, Svelte and SvelteKit: lifecycle hooks, scoped selectors and cleanup on unmount. | greensock/gsap-skills (vendored) | MIT |
gsap-performance | Official GSAP performance: transforms over layout properties, batching, will-change and avoiding jank for smooth 60fps. | greensock/gsap-skills (vendored) | MIT |
gsap-plugins | Official GSAP plugins: Flip, Draggable, ScrollSmoother, SplitText, SVG drawing, CustomEase and plugin registration. | greensock/gsap-skills (vendored) | MIT |
gsap-react | Official GSAP in React and Next.js: the useGSAP hook, refs, gsap.context() and cleanup. | greensock/gsap-skills (vendored) | MIT |
gsap-scrolltrigger | Official GSAP ScrollTrigger: scroll-linked animation, pinning, scrub, parallax and trigger configuration. | greensock/gsap-skills (vendored) | MIT |
gsap-timeline | Official GSAP timelines: gsap.timeline(), the position parameter, nesting and playback control for sequenced animation. | greensock/gsap-skills (vendored) | MIT |
gsap-utils | Official gsap.utils helpers: clamp, mapRange, normalize, interpolate, random, snap, toArray, wrap and pipe. | greensock/gsap-skills (vendored) | MIT |
hallmark | Hallmark anti-slop page builder: dealt themes, a hard design floor and sloplint checks for new websites; existing designs only when Hallmark is named. | nutlope/hallmark (adapted) | MIT |
impeccable-adapt | Impeccable's adapt reference (vendored): Adapt a design for different devices, screen sizes and contexts. | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-animate | Impeccable's animate reference (vendored): Add purposeful animation and motion. | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-audit | Impeccable's audit reference (vendored): Technical quality checks: accessibility, performance, theming, responsive, anti-patterns. | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-colorize | Impeccable's colorize reference (vendored): Add strategic color to monochromatic or flat interfaces. | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-core | Impeccable's core design skill (vendored): principles, surface modes and the command index for designing, critiquing, auditing and polishing frontends. Every... | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-critique | Impeccable's critique reference (vendored): UX design review with heuristic scoring, personas and cognitive-load checks. | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-harden | Impeccable's harden reference (vendored): Production readiness: errors, i18n, overflow and edge cases. | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-layout | Impeccable's layout reference (vendored): Fix spacing, rhythm and visual hierarchy. | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-polish | Impeccable's polish reference (vendored): Final quality pass before shipping. | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-typeset | Impeccable's typeset reference (vendored): Improve typography hierarchy, fonts and readability. | pbakaus/impeccable (adapted) | Apache-2.0 |
intent-evaluate | Heuristic UX evaluation: Nielsen heuristics, cognitive walkthroughs and dark-pattern detection, findings ranked by severity. | ghaida/intent (adapted) | CC0-1.0 |
intent-fortify | Edge-case hardening: state inventories, error recovery, empty, loading and first-run states, stress tests. | ghaida/intent (adapted) | CC0-1.0 |
intent-include | Inclusive design depth: screen reader, keyboard, cognitive and motor accessibility and accessibility test methods. | ghaida/intent (adapted) | CC0-1.0 |
intent-journey | Experience flows: task flows, multi-step journeys, onboarding, checkout, settings and cross-platform adaptation. | ghaida/intent (adapted) | CC0-1.0 |
intent-localize | Cultural adaptation: localization beyond translation, density, formats, trust signals and market assumptions. | ghaida/intent (adapted) | CC0-1.0 |
intent-organize | Information architecture: navigation, taxonomy, labeling, search and browse, card sorting and tree tests. | ghaida/intent (adapted) | CC0-1.0 |
intent-transpose | Cross-platform translation of a design between web, mobile, TV and other surfaces. | ghaida/intent (adapted) | CC0-1.0 |
intent-wireframe | Wireframes: low-to-high fidelity HTML wireframe ladders with a viewer and a styleguide. | ghaida/intent (adapted) | CC0-1.0 |
jk-better-accessibility | Accessibility audit and fixes against WCAG 2.2: keyboard focus, ARIA, accessible names, form labels and errors, screen reader announcements, motion and zoom. | jakubkrehel/skills (vendored) | MIT |
jk-better-colors | Colour systems: generate OKLCH palettes, name semantic tokens, convert formats and measure contrast. | jakubkrehel/skills (vendored) | MIT |
jk-better-interface | Whole-interface review that runs the better-* domain skills and returns one ranked report across accessibility, layout, copy, type, colour and polish. | jakubkrehel/skills (vendored) | MIT |
jk-better-layout | Layout structure: grouping, alignment, reading order, responsive behaviour and room for translated or mirrored (RTL) text. | jakubkrehel/skills (vendored) | MIT |
jk-better-typography | Text rendering: type scale, spacing, font features, wrapping, truncation and punctuation. | jakubkrehel/skills (vendored) | MIT |
jk-better-ui | UI polish with exact values: border radius, optical alignment, shadows, icon states and micro-animation. | jakubkrehel/skills (vendored) | MIT |
jk-better-writing | UX writing for interface copy: rewrite labels, button text, error messages, empty state copy and confirmations in the product's voice. | jakubkrehel/skills (vendored) | MIT |
jk-break | Render one component under every reachable scenario on a temporary page to stress test it; runs only when explicitly invoked. | jakubkrehel/skills (vendored) | MIT |
jk-build-design | Build UI from a Figma file or a design image so it matches, using the project's existing tokens and components. | jakubkrehel/skills (vendored) | MIT |
jk-interface-review | Change-scoped interface review of a branch, pull request or uncommitted diff: UI problems it introduced or regressed. | jakubkrehel/skills (vendored) | MIT |
jk-state-machine | Render every state of a component on a throwaway page with mock data and a keyboard switcher; runs only when explicitly invoked. | jakubkrehel/skills (vendored) | MIT |
jk-variant | Build several variants of the component in progress on its real page and iterate to pick one; runs only when explicitly invoked. | jakubkrehel/skills (vendored) | MIT |
pfd | Perception-First Design: a five-layer, perception-grounded method to evaluate, derive or analyze design, copy and communication. | skovalik/perception-first-design (adapted) | CC-BY-SA-4.0 |
taste-high-end-visual-design | Opt-in classic taste style, use only when named: a calm premium look with soft contrast, generous whitespace and spring motion. | Leonxlnx/taste-skill-classics (vendored) | MIT |
taste-industrial-brutalist-ui | Opt-in classic taste style, use only when named: industrial brutalism with Swiss type, rigid grids and extreme scale contrast. | Leonxlnx/taste-skill-classics (vendored) | MIT |
taste-minimalist-ui | Opt-in classic taste style, use only when named: editorial minimalist product UI with a warm monochrome palette and flat bento grids. | Leonxlnx/taste-skill-classics (vendored) | MIT |
taste-stitch-design-taste | Opt-in classic, use only when named: write a Google Stitch export file with strict type, colour and layout rules; a project's canonical DESIGN.md wins. | Leonxlnx/taste-skill-classics (vendored) | MIT |
typeui | TypeUI MCP context (project design skills, brand kits, UI prompts, layout variations); use only when the TypeUI MCP server is connected. | bergside/typeui (adapted) | MIT |
uupm-design-system | Three-layer design tokens, component specs and token validation, plus BM25 slide decision data; output proposes DESIGN.md changes. | nextlevelbuilder/ui-ux-pro-max-skill (adapted) | MIT |
uupm-ui-styling | shadcn/ui, Tailwind and canvas styling recipes with scripts for adding components and generating a Tailwind config. | nextlevelbuilder/ui-ux-pro-max-skill (adapted) | MIT |
uupm-ui-ux-pro-max | Searchable UI/UX data and muster's default UX review skill: priority table, pre-delivery checklist, styles, palettes, fonts, UX guidelines and stack rules. | nextlevelbuilder/ui-ux-pro-max-skill (adapted) | MIT |
wsh-design-system-patterns | Build scalable design systems with design tokens, theming infrastructure, and component architecture patterns. Use when creating design tokens, implementing ... | wshobson/agents (vendored) | MIT |
wsh-nextjs-app-router-patterns | Master Next.js 14+ App Router with Server Components, streaming, parallel routes, and advanced data fetching. Use when building Next.js applications, impleme... | wshobson/agents (vendored) | MIT |
wsh-react-state-management | Master modern React state management with Redux Toolkit, Zustand, Jotai, and React Query. Use when setting up global state, managing server state, or choosin... | wshobson/agents (adapted) | MIT |
wsh-responsive-design | Implement modern responsive layouts using container queries, fluid typography, CSS Grid, and mobile-first breakpoint strategies. Use when building adaptive i... | wshobson/agents (adapted) | MIT |
wsh-tailwind-design-system | Build scalable design systems with Tailwind CSS v4, design tokens, component libraries, and responsive patterns. Use when creating component libraries, imple... | wshobson/agents (vendored) | MIT |
wsh-wcag-audit-patterns | Conduct WCAG 2.2 accessibility audits with automated testing, manual verification, and remediation guidance. Use when auditing websites for accessibility, fi... | wshobson/agents (vendored) | MIT |
debug (1)
| Skill | What it does | Source | License |
|---|---|---|---|
sp-debug | Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes | obra/superpowers (adapted) | MIT |
author (13)
| Skill | What it does | Source | License |
|---|---|---|---|
book-run | Book runs: canon ledger, continuity debts, outline methods, POV and tense discipline, nonfiction structure. | danjdewhurst/story-skills (inspired-by), Calliope-Editor/writing-skills (inspired-by), epicsagas/Velith (inspired-by), clutcher/book-writing-skills (inspired-by), How to Write a Novel Using the Snowflake Method (inspired-by), NousResearch/autonovel (inspired-by), Save the Cat! (inspired-by), The Story Grid (inspired-by), Techniques of the Selling Writer (scene and sequel) (inspired-by), Chapter Blog nonfiction outline; ProWritingAid non-fiction structure (inspired-by) | Apache-2.0 |
impeccable-clarify | Impeccable's clarify reference (vendored): Improve UX copy, labels, instructions and error messages. | pbakaus/impeccable (adapted) | Apache-2.0 |
intent-articulate | UX writing and content design: labels, errors, empty states, onboarding copy, voice and tone, content models. | ghaida/intent (adapted) | CC0-1.0 |
intent-present | Design presentations: reviews and research readouts as reveal.js, PowerPoint, Google Slides or Figma decks. | ghaida/intent (adapted) | CC0-1.0 |
intent-storytelling | Narrative for design work: story structure for research findings, proposals and pitches. | ghaida/intent (adapted) | CC0-1.0 |
mkt-copy-editing | Edit existing copy without rewriting it: seven focused sweeps, point edits for single lines, AI-tell removal and a content-refresh checklist. | coreyhaines31/marketingskills (inspired-by) | MIT |
muster-author | Built-in authoring provider — draft persuasive, audience-first copy using proven frameworks. Used by content/doc pipelines for the author role. | WynterJones/CoppieGPT (inspired-by), boraoztunc/skills (inspired-by), AIDA (Attention, Interest, Desire, Action) (inspired-by), The QUEST copywriting formula (inspired-by), How to Write Copy That Sells (the PASTOR framework) (inspired-by), Breakthrough Advertising (awareness and sophistication) (inspired-by), PAS (Problem-Agitate-Solution) and BAB (Before-After-Bridge) copywriting formulas (inspired-by), Search Quality Rater Guidelines (E-E-A-T) (inspired-by), Competing Against Luck (Jobs to Be Done) (inspired-by) | Apache-2.0 |
simple-english | Plain, ASD-STE100-style controlled English for docs, runbooks, procedures, error messages and release notes, with a Strict mode and a rule-catalog CHECK mode. | AminBlg/SimpleEnglish (vendored) | MIT |
uupm-brand | Brand voice, visual identity, messaging frameworks and asset consistency checks, with brand-to-token sync scripts. | nextlevelbuilder/ui-ux-pro-max-skill (adapted) | MIT |
uupm-slides | Strategic HTML presentations: deck strategies, layout patterns, copywriting formulas and Chart.js templates. | nextlevelbuilder/ui-ux-pro-max-skill (adapted) | MIT |
wsh-changelog-automation | Automate changelog generation from commits, PRs, and releases following Keep a Changelog format. Use when setting up release workflows, generating release no... | wshobson/agents (adapted) | MIT |
wsh-incident-runbook-templates | Create structured incident response runbooks with step-by-step procedures, escalation paths, and recovery actions. Use this skill when building a service out... | wshobson/agents (adapted) | MIT |
wsh-postmortem-writing | Write effective blameless postmortems with root cause analysis, timelines, and action items. Use when conducting incident reviews, writing postmortem documen... | wshobson/agents (adapted) | MIT |
research (7)
| Skill | What it does | Source | License |
|---|---|---|---|
intent-investigate | User research: interview guides, surveys, usability test plans, diary studies and synthesis. | ghaida/intent (adapted) | CC0-1.0 |
intent-measure | UX measurement: goal-signal-metric chains, A/B test design, funnels and counter-metrics. | ghaida/intent (adapted) | CC0-1.0 |
kdense-literature-review | Systematic, scoping or narrative literature reviews: recorded search strategies, coverage logs, screening counts, citation verification and thematic synthesis. | K-Dense-AI/scientific-agent-skills (adapted) | MIT |
kdense-market-research-reports | Evidence-traceable market research: source and claims ledgers, scenario-based TAM/SAM/SOM sizing, forecast sensitivity and an audit-ready release gate. | K-Dense-AI/scientific-agent-skills (vendored) | MIT |
kwp-knowledge-synthesis | Merge results from many sources into one deduplicated, attributed answer, with freshness and authority confidence and conflicts surfaced. | anthropics/knowledge-work-plugins (adapted) | Apache-2.0 |
kwp-validate-data | QA an analysis before it ships: methodology, calculation spot-checks, common data pitfalls, chart honesty and a ready, caveats or revise verdict. | anthropics/knowledge-work-plugins (adapted) | Apache-2.0 |
muster-research | Built-in research provider — gather and synthesize evidence with sources. Used by domain pipelines for the research role. | assafelovic/gpt-researcher (inspired-by), mbzuai-nlp/qraft (inspired-by), Unlock AI: Open Engine and Open Skills (inspired-by), LLM-Cite (inspired-by) | Apache-2.0 |
score (4)
| Skill | What it does | Source | License |
|---|---|---|---|
evals-eval-audit | Audit an LLM eval pipeline for missing error analysis, unvalidated judges and vanity metrics, then rank the fixes. | ai-evals-course/evals-skills (adapted) | Apache-2.0 |
evals-validate-evaluator | Calibrate an LLM judge against human labels: train/dev/test splits, TPR and TNR, Rogan-Gladen bias correction and a bootstrap interval. | ai-evals-course/evals-skills (adapted) | Apache-2.0 |
evals-write-judge-prompt | Write a binary Pass/Fail LLM-as-judge prompt for one failure mode: criterion, definitions, few-shot critiques and structured output. | ai-evals-course/evals-skills (adapted) | Apache-2.0 |
muster-scorer | Built-in scoring provider — evidence-cited rubric scoring with the floor principle. Used by domain pipelines for the score role. | ai-evals-course/evals-skills (inspired-by), felipelobomotta-blip/book-genesis-studio (inspired-by), CSHaitao/Awesome-LLMs-as-Judges (inspired-by), llm-as-a-judge/Awesome-LLM-as-a-judge (inspired-by), microsoft/LLM-Rubric (inspired-by) | Apache-2.0 |
architecture-review (1)
| Skill | What it does | Source | License |
|---|---|---|---|
wsh-slo-implementation | Define and implement Service Level Indicators (SLIs) and Service Level Objectives (SLOs) with error budgets and alerting. Use when establishing reliability t... | wshobson/agents (vendored) | MIT |
performance (1)
| Skill | What it does | Source | License |
|---|---|---|---|
wsh-performance-engineer | Expert performance engineer specializing in modern observability, application optimization, and scalable system performance. Masters OpenTelemetry, distribut... | wshobson/agents (vendored) | MIT |
seo (1)
| Skill | What it does | Source | License |
|---|---|---|---|
wsh-seo-structure-architect | Analyzes and optimizes content structure including header hierarchy, suggests schema markup, and internal linking opportunities. Creates search-friendly cont... | wshobson/agents (vendored) | MIT |
humanize (1)
| Skill | What it does | Source | License |
|---|---|---|---|
muster-humanizer | Built-in humanizer — rewrite human-facing text to remove AI tells while preserving meaning. Mandatory final filter on anything a human will read. | blader/humanizer (adapted), rudra496/StealthHumanizer (adapted), conorbronsdon/avoid-ai-writing (adapted), jalaalrd/anti-ai-slop-writing (inspired-by), aaaronmiller/humanize-writing (adapted), linexjlin/GPTs (inspired-by), hardikpandya/stop-slop (adapted), petergyang/no-ai-slop (adapted), Unlock AI: Open Engine and Open Skills (inspired-by), Wikipedia:Signs of AI writing (inspired-by) | MIT |
prompt-quality (4)
| Skill | What it does | Source | License |
|---|---|---|---|
evals-generate-synthetic-data | Generate diverse synthetic eval inputs from failure-targeted dimension tuples when real traces are sparse. | ai-evals-course/evals-skills (adapted) | Apache-2.0 |
mp-writing-for-agents | Write documents agents consume (skills, AGENTS.md, CLAUDE.md): context pointers, information hierarchy, completion criteria, leading words and pruning. | mattpocock/skills (vendored) | MIT |
muster-prompt-smith | Built-in prompt-quality provider — lint, eval, and optimize prompts an application generates to build agents/agentic workflows (and prompts found in a codeba... | ai-evals-course/evals-skills (inspired-by), hermes-labs-ai/lintlang (mechanism-port), promptfoo/promptfoo (inspired-by), shobrook/promptimal (inspired-by), gepa-ai/gepa (inspired-by), Prompting best practices (Claude Docs) (inspired-by), Reduce hallucinations (Claude Docs, Strengthen guardrails) (inspired-by), Define success criteria and build evaluations (Claude Docs) (inspired-by), The new rules of context engineering for Claude 5 generation models (inspired-by) | Apache-2.0 |
sp-writing-skills | Use when creating new skills, editing existing skills, or verifying skills work before deployment | obra/superpowers (adapted) | MIT |
image (3)
| Skill | What it does | Source | License |
|---|---|---|---|
muster-image | Built-in image-prompt authoring provider — reads the brand profile and drafts self-contained, brand-constrained image-generation prompts. Used by content/doc... | Unlock AI: Open Engine and Open Skills (inspired-by) | Apache-2.0 |
uupm-banner-design | Banner art direction and exact sizes for social, ads, web heroes and print. | nextlevelbuilder/ui-ux-pro-max-skill (adapted) | MIT |
uupm-design | Brand-asset production: logos, corporate identity mockups, icons, social images and banners, routing to the sibling uupm builtins. | nextlevelbuilder/ui-ux-pro-max-skill (adapted) | MIT |
video (1)
| Skill | What it does | Source | License |
|---|---|---|---|
muster-video | Built-in video-content authoring provider — tightens scripts (radio-edit pass), drafts timestamped b-roll shot lists, and outlines edit decisions. Text artif... | Muster | Apache-2.0 |
unrouted (19)
| Skill | What it does | Source | License |
|---|---|---|---|
cws-creative-writing-craft | Fiction execution craft: production modes, prose immersion, scene construction, style analysis and genre contracts. | haowjy/creative-writing-skills (vendored) | Apache-2.0 |
cws-reader-sim | Read a draft as a stated first-time reader and report the felt experience moment by moment, not a critique. | haowjy/creative-writing-skills (vendored) | Apache-2.0 |
cws-story-review | Fiction review by level: editorial, developmental, line, copyedit, proofread, craft critique and continuity. | haowjy/creative-writing-skills (vendored) | Apache-2.0 |
cws-writing-principles | What fiction readers want (reader reward channels) and the specific ways model prose damages them. | haowjy/creative-writing-skills (vendored) | Apache-2.0 |
impeccable-bolder | Impeccable's bolder reference (vendored): Amplify safe or bland designs without losing usability. | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-craft-floor | Impeccable's craft-floor reference (vendored): Quality floor, absolute bans and verification checks to read immediately before any UI edit. | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-delight | Impeccable's delight reference (vendored): Add personality and memorable touches that are earned. | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-distill | Impeccable's distill reference (vendored): Strip a design to its essence and remove complexity. | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-document | Impeccable's document reference (vendored): Generate DESIGN.md from existing project code. | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-extract | Impeccable's extract reference (vendored): Pull reusable tokens and components into the design system. | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-onboard | Impeccable's onboard reference (vendored): Design first-run flows, empty states and activation. | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-operate | Impeccable's operate reference (vendored): Extended guidance for Operate surfaces (app UI, dashboards, tools) and Read surfaces (docs). | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-optimize | Impeccable's optimize reference (vendored): Diagnose and fix UI performance. | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-overdrive | Impeccable's overdrive reference (vendored): Push past conventional limits with ambitious interaction design. | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-quieter | Impeccable's quieter reference (vendored): Tone down aggressive or overstimulating designs. | pbakaus/impeccable (adapted) | Apache-2.0 |
impeccable-shape | Impeccable's shape reference (vendored): Plan UX and UI before code: discovery interview and a confirmed design brief. | pbakaus/impeccable (adapted) | Apache-2.0 |
muster-pm | PM artifact kit for AI-native teams — INTENT and SPEC templates, the EARS requirement patterns and AI-eval guidance, all checked by muster spec-lint; read ... | gotalab/cc-sdd (vendored), ghaida/intent (adapted), mattpocock/skills (adapted), phuryn/pm-skills (adapted), github/spec-kit (inspired-by), Fission-AI/OpenSpec (inspired-by), Wirasm/prp (inspired-by), product-on-purpose/pm-skills (adapted), phuryn/product-engineering (adapted), open-gsd/gsd-core (inspired-by), Define success criteria and build evaluations (Claude Docs) (inspired-by), Demystifying evals for AI agents (inspired-by), Evaluation best practices (OpenAI API docs) (inspired-by), Your AI Product Needs Evals (inspired-by), AI Evals: Everything You Need to Know (inspired-by) | Apache-2.0 |
muster-roadmap | Built-in roadmap prioritizer — outcome-framed initiatives, cited evidence, RICE factors scored by muster prioritize, and a Now/Next/Later roadmap. Read by ... | RICE: simple prioritization for product managers (inspired-by), The Now-Next-Later roadmap (inspired-by) | Apache-2.0 |
sepia | De-AI pass for fiction and professional prose: StoryScope narrative-architecture repair, per-venue rules, four operations. | Nanako0129/sepia (vendored) | MIT |