Skills
97,501 packaged capabilities and playbooks for agents. Install any of them with one command.
Testing & quality
Deep performance optimization skill for Triton operators on Ascend NPU, dedicated to achieving the Triton operator performance improvements requested by users. Core techniques include but are not limited to Unified Buffer (UB) capacity planning, multi-token parallel processing, MTE/Vector pipeline parallelism, and mask optimization. Always trigger this skill when the user mentions the following: performance optimization of Vector-type Triton operators on Ascend NPU.
Designs new features, extensions, or modifications to Uncle Bob's Acceptance Pipeline Specification — new mutation strategies, Gherkin syntax support, report formats, pipeline stages, IR fields, or handler patterns. Trigger when someone asks "how would I add X to the acceptance pipeline" or discusses spec-level changes to the parser, generator, runtime, mutator, or reporter components — even if they don't explicitly say "feature design." Works in tandem with the acceptance-pipeline-catalog skill, which provides the baseline spec reference.
Audits or remediates Flutter widgets against WCAG 2.2 conformance levels A, AA, or AAA across iOS, Android, Web, macOS, Windows, and Linux, covering Semantics labels and screen reader output under VoiceOver and TalkBack, touch target sizes, dragging alternatives, focus order and keyboard navigation, color contrast, text scaling, animation gating on disableAnimations, and form autofillHints, including the test suite that locks every fix in. Use when building, auditing, or reviewing Flutter widgets for accessibility on one or several of those platforms.
Help the user pick, write, or apply an AgentRC policy. Policies customise readiness scoring by disabling irrelevant checks, overriding impact/level, setting pass-rate thresholds, or chaining org baselines with team overrides. Use when the user asks about strict mode, AI-only scoring, custom weights, CI gating, or wants org-wide standardisation.
Use when writing or changing a Bagisto Pest test — feature or unit tests, assertions, datasets, mocking, architecture tests, or registering a new package's suite. Trigger phrases include "pest", "test", "unit test", "feature test", "assertion", "dataset", "mock", "testsuite", "TDD", "coverage".
Create Behavior-Driven Development (BDD) feature files using Gherkin syntax. Write clear, executable specifications that describe system behavior from the user's perspective. Use for requirements documentation, acceptance criteria, and living documentation.
Use this skill to tune Jetpack Compose lazy-layout prefetch with LazyLayoutCacheWindow (Compose Foundation 1.9+, @ExperimentalFoundationApi) and pausable composition in prefetch (Compose Foundation 1.10+, default on). Covers configurable Dp-based ahead/behind cache windows plumbed through rememberLazyListState(cacheWindow = ...), NestedPrefetchScope for items containing inner lazy layouts (HorizontalPager inside a LazyColumn row), version requirements, and the trade-off between memory pressure and idle-frame work. Use when the developer mentions dropped frames at high scroll velocity, prefetch window, ahead/behind extents, LazyLayoutCacheWindow, NestedPrefetchScope, pausable composition for prefetch, or wants composition retained for items briefly scrolled past. Item-level fixes (keys, contentType) live in a sibling skill.
Reverse-engineering a Sketch file (or Figma export with similar shape) into pixel-perfect React + CSS — the iteration mental model, tree reconstruction, layout inference algorithms, geometry math, visual-regression diffing, and the style/typography/path conversions that make "improvement without regression" enforceable. Trigger even if the user doesn't explicitly mention "algorithms" but is converting a design source into web code, building a design-to-code pipeline, or struggling to make incremental fidelity improvements without breaking previously-converted output.
Audits any developer documentation site across 33 checks in 7 categories and produces a scored report (out of 100) with Pass / Warn / Fail status per check. Use this skill whenever a user provides a docs URL and asks to audit it, review it, score it, check its quality, or evaluate it for AI discoverability, SEO, structure, content, or completeness. Also trigger when the user says "run a docs audit", "audit these docs", "check this documentation", "how good are the docs for X", or pastes a docs URL with any evaluative intent. Always use this skill — do not attempt a freeform docs review without following this structured workflow.
E2E test architecture and patterns with Playwright. Use when designing test suites, structuring Page Object Models, planning CI sharding strategies, setting up authentication flows, or organizing tests with tags and annotations. Use for test architecture, accessibility auditing with axe-core, network mocking strategies, visual regression workflows, HAR replay, and storageState authentication patterns. For Playwright API details, browser automation, or web scraping, use the playwright skill instead.
Checks if a community pull request is ready for human review. Verifies CLA signature, PR title format, description completeness, test coverage, and cubic-dev-ai issues, then triages to the right Linear team or recommends a close. Use when given a PR number or branch name to review, or when the user says /community-pr-readiness-check, or asks to check if a PR is ready for review.
Guides work on `packages/frontend/editor-ui` experiments. Use when creating, extending, wiring, testing, reviewing, or retiring editor-ui experiments, PostHog feature flags, experiment key indexes, variants, stores/composables, persisted experiment state, or experiment telemetry.
Owns Python code style for this stack: ruff for lint + format, numpydoc for docstrings. Three responsibilities — (1) place the project's `ruff.toml` from the bundled template once the stack and workspace are in place, (2) run ruff against any Python files Claude has just generated or edited, and (3) contextualize each touched file's comments to the data-science problem — rewriting any leftover template / workflow prose (skill names, gates, runner, digest, guard-rails) into concise, problem-specific docs so the user's committed files read like a colleague wrote them, not like a generated scaffold. Stops at "the touched files pass `ruff check` and document the problem, not the process." TRIGGER when (any of these): (1) a Python file was just created or edited via Write / Edit / MultiEdit — invoke this skill before declaring the task done so ruff is run AND the file's comments are contextualized to the problem; (2) a fresh ML workspace was just scaffolded by `organize-ml-workspace` and the project has no `ruff.toml` at its root yet — drop the bundled template; (3) the user asks about lint, format, docstring style, or reaches for `black` / `isort` / `flake8` / `pydocstyle` (redirect to ruff — the stack's canonical linter, owned by `data-science-python-stack` Tier 1). SKIP when: the project is non-Python; the only edits in this turn are to Markdown / TOML / JSON / YAML; the file lives in a third-party vendored directory the user doesn't own. HOW TO USE: run ruff manually on the files you just touched — do not configure a PostToolUse hook for this. **Read the "Stop conditions" block and emit the Pre-flight checklist as visible text in your response — both are mandatory before running ruff.**
Ship a release: the final pre-tag gate for OrchestKit itself, run once a change is already known good. Walks 12 ordered steps (build, full test suite, security tests, typecheck, manifest count validation, changelog entry, version bump across package.json and CLAUDE.md, selective staging, conventional release commit, tag, confirmed push), halting at the first failed step with remediation guidance. Use when cutting and shipping a version, not for judging whether a change is correct (that is /ork:verify).
Intelligent test selection based on code changes. Maps source files to tests via import analysis, implements tiered testing (fast < 1 min, impacted < 5 min, full suite), and tracks test reliability. Use when running tests after code changes to optimize feedback loops and CI time.
Triage and remediate vulnerabilities found by a Strix pentest (open-source CLI or app.strix.ai cloud), then re-run Strix to verify each fix. Use after a Strix scan reports findings, or when the user asks to fix security issues from a strix_runs report, vulnerabilities.json, findings.sarif, or a cloud scan's vulnerabilities.
Use when a Bagisto change is about to be called done, or when asked to run the verification gates — code style, tests, end-to-end tests and translation completeness. Trigger phrases include "verify", "is this done", "run the gates", "pint", "pest", "playwright", "translations check", "ready to commit".
Import QA test cases from CawPlan (Requirement/TestPoint or INLINE session cases) into TestRail with preview-first workflow. Use when: SQA wants to push T1-generated cases to TestRail; after test-point generation, expand cases and import; re-import after requirement updates (skip-only, no overwrite). All title-only cases → optional Steps-confirm (guide to `/cawplan-testcase-generate` or continue without steps); ≥1 expanded case → seamless import. NOT for: generating test points (use `cawplan-testpoint-generate`); archiving requirements (use `cawplan-requirement-analyze`); creating Test Plans/Runs (use `cawplan-testplan-layout`); failure-to-defect (use `cawplan-defect-ticket`); release risk (A5).
Chaos engineering principles, controlled failure injection, resilience testing, and system recovery validation. Use when testing distributed systems, building confidence in fault tolerance, or validating disaster recovery.
Clean, maintainable, human-readable code principles combined with anti-over-engineering discipline: naming, single responsibility, DRY, and scoping changes to exactly what was requested. Use when writing new code, refactoring existing code, reviewing code for quality, or deciding how much abstraction a change actually needs.
Audit a design proposal or diff against Exarchos's architectural invariants — event-sourcing integrity (INV-1), facade equivalence over shared dispatch core (INV-2), basileus-forward (INV-3), platform-agnosticity (INV-4), and agent-first interface design (INV-5a input ergonomics, INV-5b spec-aligned output contract, INV-5c Aspire-inspired control-plane verbs, INV-5d action discriminator pattern). Pairs with /axiom:backend-quality — this skill is project-specific (axiom is generic). Triggers: 'check invariants', 'design conformance', 'check #1118 / #1109', or /design-invariants.
Use when writing, changing, or reviewing Cucumber and Playwright tests under `e2e/`, including feature files, step definitions, support code, scenario tags, locators, and assertions. Do not use for Vitest, React Testing Library, backend tests, or generic browser automation outside the E2E suite.
Visual quality review for FrameVideo compositions. Use when checking or fixing material overlap, text overlap, clipped or overflowing text, low text/background contrast, unsafe title/logo/CTA/caption placement, dynamic text fitting, visual collisions during animation, or final composition QA after creating or significantly editing FrameVideo HTML.
Centralized help URL reference for accessibility remediation. Maps axe-core rule IDs to Accessibility Insights info-examples pages, document rule IDs to Microsoft Office and Adobe PDF help pages, and WCAG criteria to W3C Understanding documents. Use when generating CSV exports, markdown reports, or any output that links findings to external remediation documentation.
Authors a new Instance AI workflow or Agent eval case — written locally as JSON, calibrated against a real build, then pushed to the LangTracer suite CI runs — build cases, behaviour/process cases, credential cases, and seeded (mid-conversation) cases — with intent-driven expectations. Use when adding or changing an Instance AI eval, or debugging why one is flaky.
(NS) Security/compat/a11y + Vercel Web Interface Guidelines. Use when hardening security, fixing CSP/CORS/headers, reviewing UI / checking a11y / auditing UX, or Web Interface Guidelines — even without saying "best practices"; also ns-proto-creator close-out. Do NOT use for MR/SOLID (ns-reviewer), visual redesign (ns-frontend-design), or features (ns-coder / ns-spec-driven).
Validate and review Wallet MCP plugin files against the Plugin Specification. Use when writing a new plugin, preparing a plugin PR for submission, self-checking a plugin before opening a PR, or reviewing someone else's plugin PR. Triggers on requests like "check my plugin against the spec", "validate this plugin file", "review my plugin PR", "does this plugin conform to the spec", "prepare my plugin for submission", "write a plugin for X protocol".
Select and run Python SDK verification with nox, Makefile targets, Ruff, mypy, pytest markers, sanity tests, type inference checks, and build checks. Use when adding Python tests, diagnosing Python CI, or validating Python SDK/provider changes. Do not use for TypeScript-only checks.
Audits any SDK documentation site and produces a fully scored, downloadable HTML report. Use this skill whenever a user provides a documentation URL and asks to audit, review, analyse, score, or check the quality of SDK docs. Also trigger for phrases like "audit the SDK docs at X", "check if the docs at X are complete", "review this SDK documentation", "how good are these docs", "run an SDK audit on X", or any time a user pastes a URL alongside words like audit, review, crawl, check, analyse, quality, completeness, or gaps. ALWAYS use this skill — do not attempt the audit without following this structured crawl → analyse → cross-reference → report workflow.
Trigger when the user asks to audit, review, or security-check a Solana smart contract, or the codebase contains solana_program, anchor_lang, pinocchio, #[program], or #[derive(Accounts)]. Also trigger for requests such as "check for vulnerabilities", "find exploits", or "security review" on Rust code with Solana dependencies.
Scores how completely an implementation fulfills a PRD/spec, case by case, and produces a single comparable final grade. Invoke only when explicitly named (e.g. run spec-driven-eval); do not auto-trigger. Use when benchmarking spec-driven implementations, grading acceptance criteria, evaluating whether a feature was 100% implemented, comparing multiple implementations of the same PRD, or auditing implementation and test coverage (unit and e2e) against product requirements. Do NOT use for planning or building features (use tlc-spec-driven), writing PRDs, or general code review unrelated to a spec.
Use this skill to instrument a Jetpack Compose composable with `@TraceRecomposition` from `skydoves/compose-stability-analyzer` so per-recomposition diffs (which state or parameter changed, what value transition) print to logcat under the `Recomposition` tag. Works in release-with-debug-symbols builds where Android Studio Layout Inspector cannot reach, and feeds the IntelliJ / Android Studio plugin's live recomposition heatmap (green under 10, yellow 10–50, red 50+). Covers the Gradle plugin setup, the `ComposeStabilityAnalyzer.setEnabled(BuildConfig.DEBUG)` runtime gate that keeps the instrumentation out of production, and the handoff to debug-time Layout Inspector and CI `stabilityCheck`. Use when the user mentions `@TraceRecomposition`, "trace recomposition", "compose-stability-analyzer", "recomposition logcat", "recomposition heatmap", "release-mode recomposition counts", or needs to confirm a stability fix in a release-like build.
Use when the user asks for agent-assisted code review of a git diff, working tree, branch, base ref, git-backed plan or document change, or review/fix loop where scope triage, delegated reviewers, specification gaps, or cascade-safe fixes may matter.
Reviews watchOS code for app lifecycle, complications (ClockKit/WidgetKit), WatchConnectivity, and performance constraints. Use when reviewing code with import WatchKit, WKExtension, WKApplicationDelegate, WCSession, or watchOS-specific patterns.
Bounded research experiment loop that tries hypotheses, measures benchmark evidence, keeps what works, and records what fails. Use when the user asks to optimize a research metric, run an experiment loop, improve model/retrieval/evaluation performance iteratively, or benchmark a research hypothesis.
Write, modify, or debug Docker-based Composio CLI end-to-end tests under ts/e2e-tests/cli, including binary invocation, fixture isolation, output assertions, and package manifests. Use for CLI E2E test suites only; use cli-command for CLI source implementation.
Keep TypeScript and Python SDK behavior, generated client usage, public API naming, and docs examples aligned. Use when a change affects both SDKs, when generated client pins move, when comparing TS/Python behavior, or when a backend API contract changed. Do not use for single-language internal-only changes.
Use this skill when the user is tuning a live or captured `doca-flow` pipeline with `doca_flow_tune`—snapshotting pipe / counter / KPI state, picking a tuning axis (rule placement, resource hints / table sizing, HW-offload mode) and a matching measurement (rule-install rate, lookup latency, hardware-counter delta), running offline or online (read-only or state-changing) modes, reading the dumper CSV / analyze JSON / visualize mermaid, or applying a recommendation back into the Flow program. Trigger even when the user does not explicitly mention "doca_flow_tune"—typical implicit phrasings include "Flow rule-install rate is low on BlueField", "table sizing looks wrong for this pipe", "tune visualize step is empty", "before/after counters don't move", or "which doca-flow knob does this recommendation hit". Refuse and route elsewhere for measuring baseline numbers (doca-flow-perf), writing the doca-flow application, DOCA install, or streaming Flow telemetry—those belong to other skills.
Evaluate and improve the usability of existing front-end interfaces — either local source code or live websites by URL. Use this skill whenever the user has existing front-end code OR a live website URL and wants to understand why users struggle with it, find usability problems, or make it easier to use. Triggers on direct requests like "review my UI", "audit my interface", "check accessibility", "evaluate my design", "improve the UX", "audit this website", "review https://example.com" — but ALSO on indirect symptom descriptions like "users keep abandoning this form", "something feels off about this page", "we're getting complaints about the checkout flow", "conversion dropped after the redesign", "people say it's confusing", or "make this less painful to use". If someone has a front-end component, page, or URL and the problem sounds like usability, user confusion, or poor interaction design rather than a bug or performance issue — use this skill. NOT for building new interfaces from scratch, performance optimization, or backend logic.
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.