Menu
Skills in this repository
SkillsMP has collected 495 skills from Kayforkind/skill-slice. Open a skill to review its source and details.
Kayforkind/skill-sliceShowing 40 of 495 collected skills.
Provisions a 72-hour unclaimed Neon Postgres via neon.new (REST, CLI, SDK, or Vite plugin) with no signup, writing DATABASE_URL and a claim URL. Use when the user needs a throwaway DATABASE_URL for prototyping, demos, or tests. Not for production Neon…
Read-only analyzer that recommends Claude Code hooks, subagents, skills, plugins, and MCP servers from codebase signals; does not create those files. Use when the user asks how to set up Claude Code, which automations to add, or how to optimize a Claude Code…
Diagnoses macOS Claude in Chrome MCP native-host conflicts between Claude.app (Cowork) and Claude Code CLI sockets. Use when mcp__claude-in-chrome__* tools fail, return Browser extension is not connected, or behave erratically after updates. Not for Linux or…
Implements Clerk on Next.js App Router: ClerkProvider, clerkMiddleware/auth.protect(), organizations, Svix user-sync webhooks, and server auth()/client hooks. Use when the user mentions Clerk, ClerkProvider, sign-in/sign-up, organizations, SSO, or Clerk…
Runs OpenAI CLIP for zero-shot image classification, image-text cosine matching, semantic search, and cross-modal retrieval in a shared embedding space. Use when labeling images without training, ranking images by a text query, or scoring vision-language…
Deploys a Cloudflare Worker to a live workers.dev URL with wrangler deploy --temporary (Wrangler 4.102.0+), a 60-minute claim URL, and no OAuth. Use when shipping agent-written Worker code to a link without an account, or iterating a throwaway prototype. Not…
Implements Mondrian-style color-blocking layouts with large flat geometric panels, a 3-4 color high-contrast palette, and thick black grid lines via CSS Grid or native stacks. Use when the user wants color blocking, Mondrian grids, or bold geometric color…
Crafts checkpoint-specific ComfyUI prompt pairs (FLUX, SDXL, SD1.5, Wan, AnimateDiff) with matching negatives, CFG ranges, and identity-method rules (InstantID, PuLID, FaceID, Kontext, character LoRA). Use when generating or adapting prompt text for a known…
Runs local GPU Hugging Face Hub evaluations with inspect-ai and lighteval, choosing vLLM, Transformers, or accelerate backends and smoke-test limits. Use when benchmarking a Hub model on local hardware before a full run. Not for writing model-card eval tables…
Embeds live Zoom meetings in a custom web container via Meeting SDK Component View (ZoomMtgEmbedded.createClient, Promise init/join, lowercase password). Use when integrating real Zoom meetings into React/Vue/Angular layouts with embeddable UI. Not for Zoom…
Conditions raw gravitational-wave detector strain (PyCBC TimeSeries) for matched filtering: high-pass, resample, crop filter wraparound, then Welch PSD with inverse-spectrum truncation. Use when preprocessing LIGO/Virgo/KAGRA strain, GW TimeSeries, high-pass,…
Structures agent context as a hierarchy: persistent rules files (CLAUDE.md, AGENTS.md, .cursor/rules), scoped specs, relevant source, error output, then conversation compaction. Use when starting a session, output quality degrades, switching tasks, or…
Authors Convex backends: schema.ts validators and indexes, TypeScript queries/mutations/actions, React useQuery/useMutation, Convex Auth or Clerk, file storage, crons, and npx convex deploy. Use when adding or debugging Convex in React, Next.js, Vue, Svelte,…
Creates a git branch using Sentry's username/type/short-description convention, proposing feat, fix, ref, chore, or similar from the task or local diff after user confirmation. Use when asked to create, start, or switch to a new branch. Not for commits,…
Authors Cursor Agent Skills: SKILL.md frontmatter, third-person descriptions, progressive disclosure, and ~/.cursor/skills or .cursor/skills layout. Use when the user asks to create a skill, write a SKILL.md, or design skill structure and triggers. Not for…
Drives the crit CLI blocking-review loop: GitHub-style inline comments on diffs, plans, live URLs, or local HTML, then applies unresolved review.json comments; also covers UI/UX accessibility and responsive design review. Use when the user asks to review,…
Reads and writes pandas CSV with pinned dtypes, explicit NA tokens, formula-injection sanitization, and chunked processing. Use when ingesting sensor logs, simulation results, or in-memory time-series CSVs. Not for Polars LazyFrame pipelines (polars), Kafka…
Guides end users calling NVIDIA cuOpt for routing, LP, MILP, or QP via Python, C, REST, or CLI — install, formulate, and check solver status. Use when asking about the cuOpt SDK, installation, server deploy, or optimization problem setup. Not for modifying…
Runs stop-the-line triage for test, build, and runtime failures: stop shipping features, preserve evidence, reproduce/localize/reduce, fix the root cause, then add a regression guard and safe fallbacks. Use when tests fail, the build breaks, a bug report…
Builds Godot 4.3+ dedicated servers: headless/server export templates, dedicated_server feature tag, boot autoload, lobby and match flow, config, and systemd/Docker deploy. Use when a headless authoritative simulation must host clients with no GPU or display.…
Deploys a web app to Vercel as a preview (CLI, git-push, or no-auth claim URL), linking .vercel/ when possible. Use when the user asks to deploy, push live, or get a preview link. Not for Cloudflare wrangler --temporary Workers, production --prod unless…
Operates Vercel deployments and CI with vercel deploy, promote, rollback, inspect, build --prebuilt, plus GitHub Actions, GitLab CI, and Bitbucket pipeline YAML. Use when deploying, promoting, rolling back, or wiring Vercel into a CI workflow file. Not for…
Generates multiple standalone HTML design explorations of one UI component, each in a different company design language, then adapts the winner into the real theme. Use when the user wants design options, a design sprint, or to compare UI approaches for a…
Architects workflow SKILL.md files with numbered phases, routing/safety-gate/task-driven patterns, tool assignment, and progressive disclosure. Use when building or reviewing a multi-step skill structure rather than its domain content. Not for Cursor skill…
Writes and debugs Detox gray-box E2E tests for React Native (Jest-circus, .detoxrc.js, testID matchers, device sync). Use when authoring, running, configuring, or fixing flaky Detox tests on iOS simulators or Android emulators. Not for web or hybrid browser…
Plans DevRel work: conference CFPs, talk prep, live-coding demos (10-3-1 checkpoints), podcast pitches, building-in-public, and advocacy metrics. Use when the user mentions developer advocacy, DevRel, CFP, conference talk, live coding, podcast guesting, or…
Produces a research-grounded customer journey map: stages, touchpoints, emotional curve, pain points, moments of truth, and opportunity annotations, optionally as mermaid. Use when synthesizing a linear or cyclical journey for a named persona and goal. Not…
Reviews Django and DRF access control by tracing whether User A can read, change, or delete User B objects: get_queryset scoping, permission_classes, has_object_permission, tenant managers, and IDOR on pk routes. Use when auditing Django authorization, object…
Reviews Django and DRF code for validated ORM issues: N+1 queries, unbounded querysets, missing indexes, and write loops, reporting only proven hot-path findings. Use when asked to review Django performance, find N+1s, or audit queryset and database problems.…
Queries aggregate EAS Update health from eas-cli: crash rates, installs, unique users, payload size, and embedded-vs-OTA split per channel. Use when checking whether a published Expo OTA update is healthy or which update is most popular. Not for publishing…
Sends transactional and broadcast email from Next.js/Vercel apps with Resend and React Email templates (domains, batch, webhooks, idempotency). Use when installing resend or react-email, writing React Email templates, or building app/api/send routes. Not for…
Manages Vercel/Next.js environment variables: .env load order, vercel env pull/add/ls/rm, VERCEL_OIDC_TOKEN (~12h local), and production/preview/development scoping. Use when editing .env.local, running vercel env, or debugging expired OIDC on a linked…
Creates Expo Router server endpoints as +api.ts files on EAS Hosting for secrets, database calls, third-party proxies, webhooks, and rate limits. Use when the user needs API routes, server functions, or EAS Hosting in an Expo Router app. Not for public…
Maps a need to Expo's official expo/examples with-* integrations and adapts the canonical pattern (deps, config plugins, wiring) into an existing managed app, or scaffolds via create-expo --example. Use when integrating Stripe, Clerk, maps, Reanimated,…
Builds Expo native modules and views with the Expo Modules API (Swift, Kotlin, TypeScript), including create-expo-module, expo-module.config.json, config plugins, and lifecycle hooks. Use when adding native functionality, native views, or platform support to…
Adds JavaScript and TypeScript property-based tests with fast-check (randomized inputs, shrinking, roundtrips, custom arbitraries, fc.assert). Use when verifying invariants across generated inputs rather than single examples. Not for Vitest/Jest checks of…
Scores and ranks a product backlog with RICE, Kano, Value vs Effort, and WSJF so assumptions beat HiPPO. Use when ordering a candidate feature list, resolving stakeholder conflict, or choosing a scoring method. Not for problem discovery, OKR/vision…
Extracts, loudness-normalizes (loudnorm/LUFS), mixes, resamples, and analyzes audio with FFmpeg filter graphs. Use when pulling soundtracks from video, mixing narration over music, rearranging channels, or measuring levels. Not for container remux…
Applies fp-ts in React 18/19 and Next.js 14/15: Option state, Either form validation, TaskEither fetches, RemoteData UI, and React 19 use/useActionState/useOptimistic with functional patterns. Use when building React apps that already use or should adopt…
Writes a frontend-owned data and API requirements doc covering display needs, actions, and UI states without prescribing endpoints or field names, saved under .claude/docs/ai/<feature>/backend-requirements.md. Use when a frontend developer must document data…