Claude Skills
The open format is called Agent Skills and works in Claude Code, Codex, Cursor and other agents — most people know it as Claude Skills.
Every Agent Skill we could find on GitHub, deduplicated by content. 79 566 files from 1 758 authors, of which 61 913 are unique — the rest is the same skill repackaged into someone else's repository. For each one: what it weighs in tokens, whether it ships runnable scripts, and which MCP servers it needs.
Creates a project baseline of architecture drivers and constraints. Use before design or planning; not for target design, plan review, implementation, or architecture audit.
Creates evidence-backed current or target architecture diagrams when the diagram is the primary deliverable. Not for UI design, architecture audit, or invented structure.
Creates a decision-complete target system design from requirements and constraints. Use before implementation planning; not for requirements baselines, reviews, audits, or code changes.
Records one architecture decision with context, alternatives, tradeoffs, and consequences. Use for a significant choice; not for broad design, audit, or implementation.
Validates, commits, pushes, and remotely verifies approved repository changes. Use when publication is requested; not for releases, package publishing, or announcements.
Audits queries, transactions, data-path performance, and persistence resource lifecycle. Use when data correctness or scalability is at risk; not for general performance tuning.
Reviews standalone skills and their configured distribution surfaces before publication. Use for skill release readiness; not for product code or implementation-plan review.
Audits implemented architecture fitness, boundaries, contracts, dependencies, and configuration ownership. Use for system structure; not for current-state documentation, diagrams, or plan review.
Prepares and publishes a tagged GitHub release from repository evidence. Use for an explicit release request; not for ordinary commits, packages, or community news.
Audits whether an existing test suite proves important product behavior with strong, isolated tests. Use when test confidence is uncertain; not to design or implement new tests.
Drafts and publishes fact-checked GitHub Discussions announcements. Use for releases, updates, or project news; not for release creation or issue responses.
Optimizes a measured latency, throughput, memory, CPU, or I/O problem through profiling and keep-or-discard experiments. Use for a known bottleneck; not unbiased A/B comparison.
Upgrades dependencies across package managers with breaking-change research and rollback-safe verification. Use for dependency maintenance; not general code modernization.
Evaluates new product directions using current demand, acquisition, competition, economics, and validation evidence. Use before commitment; not for backlog or implementation planning.
Modernizes a bounded capability by removing obsolete custom mechanisms or reducing bundle and maintenance cost. Use for proven modernization value; not routine upgrades or tuning.
Compares tools or implementations through reproducible A/B workloads, correctness oracles, and controlled measurements. Use to choose alternatives; not to optimize a known bottleneck.
Reviews implementation plans against repository evidence and current authoritative guidance. Use before execution to expose gaps and risks; not for completed delivery review.
Creates and runs reproducible acceptance tests for stated requirements using project-native tooling. Use when executable acceptance evidence is needed; not for audits or product fixes.
Reviews a completed scoped change and its affected runtime and contract paths. Use to find change-caused defects and verify readiness; not for codebase audit, implementation, or repair.
Designs a risk-based test strategy and prioritized scenarios without changing code. Use when requirements need a test plan; not for auditing or implementing tests.
Build a measured exposure and grading path in Three.js. Use for a 64x36 encoded luminance meter, asynchronous readback, weighted log-average exposure, asymmetric adaptation, single tone-map ownership, and a generated 32-cube post-tone-map LUT.
Build coupled precipitation and affected surfaces in Three.js. Use for falling snow, snow accumulation, model snow caps, wet asphalt puddles, procedural ripple normals, splash flipbooks, rain streaks, shared weather envelopes, and surface wetness or coverage transitions.
Direct advanced Three.js camera systems. Use for scale-aware chase rigs, thrust lag, side/orbit cameras, body-relative up vectors, quaternion handoffs, authored cinematic framing, floating origins, pointer-look controls, camera collision constraints, projection ownership, and lifecycle restoration.
Implement production bloom in advanced Three.js scenes. Use for HDR signal ordering, bloom-node controls, dual selective bloom with guaranteed material restoration, scene-relative emissive hierarchy, and effect-isolation diagnostics.
Build a deliberate final-image pipeline for advanced Three.js scenes. Use for depth, normal, albedo, and history ownership; GTAO or bent normals; bloom; eye adaptation; tone mapping; 3D LUT grading; effect-local render targets; and pass diagnostics.
Implement physically motivated sky and aerial-perspective systems in Three.js. Use for planetary atmospheres, ground-to-space transitions, Rayleigh/Mie scattering, precomputed LUTs, depth-based transmittance and inscattering, sun/moon discs, and atmosphere-aware lighting.
Build advanced procedural animation in Three.js. Use for launch kinematics, gravity turns, staging, spin docking, target-frame decomposition, spring-follow motion, rotating-frame alignment, peeling debris, analytic transform timelines, frame-rate-independent response, and quaternion control.
Build silhouette-aware parallax occlusion mapping in Three.js WebGPU and TSL. Use for height-field ray marching, relief UVs, clipped flat or curved silhouettes, inflated shells, self-shadowing, relief-aware shadow depth, and height-derived normals.
Build authored procedural buildings and architectural kits in Three.js. Use for massing grammars, exposed-edge analysis, façade bays, profiles, arches, cornices, roofs, ornaments, material-slot mesh compilation, deterministic variants, and procedural city assets.
Build coherent procedural scalar and vector fields for Three.js materials and geometry. Use for terrain, planets, wear, biomes, clouds, water masks, displacement, roughness, normals, domain warping, and any visual where several channels must derive from shared causes.
Generate authored procedural trees, grass, ivy, and vegetation in Three.js. Use for surface-following vines, painted ivy paths, stylized or GPU grass, trunks, recursive branches, roots, canopies, leaf cards, species presets, deterministic growth, and rooted blade or petiole-hinge wind.
Build production procedural mesh systems in Three.js. Use for complete hard-surface object assemblies, tilted shell lofts, UV-owned apertures, sculpted rail and frame profiles, oriented branch rings, semantic mesh writers, deliberate skins and caps, fin lofts, custom normals, material slots, instancing decisions, and close-inspection geometry budgets.
Author production real-time VFX in Three.js. Use for raymarched aurora curtains, finite-footprint emissive slabs, uniform volume integration, equirectangular radiance probes, WebGPU voxel fire and smoke, coupled volumetric fluid fields, mesh-surface emitters, signed-distance fire collisions, ship-conforming reentry plasma, generated capsule wakes, instanced analytic sparks, timed dissolving debris, dense-swap effect pools, additive holographic projections, Fresnel rim shells, scanline banding, glitch displacement, swept shape-to-shape handovers, and explicit scene-relative HDR emission hierarchy.
Author procedural planetary bodies in Three.js. Use for spherical terrain, continents, ridges, craters, biome masks, coastlines, material variation, analytic normals, altitude LOD, and bodies that must hold up from orbit through close approach.
Author production procedural materials in Three.js. Use for hybrid texture-backed PBR soil and moss with procedural displacement and masks, upward-facing model moss accumulation, atlas filtering, specular AA, planet-space fields, terrain wetness, lava and emissive surfaces, per-instance dissolve, authored PBR identities, derivative normals, and custom direct-light shadow modulation.
Build raymarched space phenomena in Three.js. Use for black-hole lensing, accretion disks, wormholes, curved-ray integration, procedural star fields, relativistic-looking distortion, bounded volumetric structures, and GPU effects that need controlled numerical integration.
Implement stable, scalable directional-shadow systems for Three.js. Use for large procedural worlds, city scenes, terrain, moving cameras, WebGPU/TSL shadow nodes, cascades, cached clipmaps, texel stabilization, update budgets, and targeted invalidation.
Implement a production GTAO path in Three.js. Use for half-resolution horizon sampling, reversed-depth reconstruction, bent-normal encoding, full-resolution bilateral reconstruction, environment-light application, contact grounding, and halo diagnosis.
Validate advanced Three.js graphics as authored systems rather than subjective screenshots. Use for fixed-view visual contracts, field and pass diagnostics, no-post baselines, seed sweeps, camera-scale tests, temporal stability checks, GPU budgets, and regression evidence for procedural scenes.
Route ambitious Three.js graphics work to the smallest expert skill set. Use for new visual experiences, graphics rewrites, reference matching, or requests spanning geometry, materials, atmosphere, shadows, temporal effects, and final image treatment.
Build view-aligned and screen-space surface effects in Three.js. Use for touch-history frost and thaw, ping-pong accumulation, reduced-resolution blur, crystalline masks, two-scale refraction, and procedural rain droplets that refract and blur a background through wet glass.
Implement volumetric cloud systems in Three.js. Use for weather-driven density, bounded raymarching, shape/detail erosion, vertical profiles, lighting cones, silver lining, temporal reconstruction, cloud shadows, multiple layers, and scalable quality modes.
Build large procedural oceans in Three.js from directional wave spectra. Use for WebGPU/TSL FFT oceans, multi-cascade wavelength bands, hybrid FFT plus Gerstner clear-water oceans, stylized above/below surface optics, permanently submerged Snell-window views, total internal reflection, forward-refracted structures through an interface, pixel-footprint spectral LOD, aquatic perspective, caustic god rays, choppy displacement, spectral derivatives, Jacobian whitecaps, windrow and temporal foam, analytic sky reflection, underwater absorption, crest scatter, and GPU validation.
Build production analytic and bounded water in Three.js. Use for shared multi-wave displacement and normals, bounded RGBA heightfield pool simulation, local drops, object-driven ripples, differential-area caustics, ray-traced pool/water/sphere volume optics, derivative-filtered normal bands, analytic sky reflection, side-aware Fresnel, heuristic screen refraction, Beer-Lambert absorption, and crest foam.
Map knowledge components and skill hierarchies for a cognitive tutoring system or adaptive learning platform. Use when designing intelligent tutoring software or skill-based mastery systems.
Design AI-supported collaborative tasks that structure group interaction and address participation problems. Use when students struggle to collaborate effectively on group tasks.
Generate a cascading hint sequence for a problem type, revealing progressively without giving answers. Use when designing tutoring dialogues or scaffolded worksheets.
Create an interactive digital worked example sequence with fading for online or blended delivery. Use when building e-learning modules, LMS content, or app-based instruction.
Explain and configure individual spacing algorithms using student performance data and forgetting curves. Use when personalising retention schedules in adaptive learning platforms.
Design an adaptive assessment loop where each student response triggers the next instructional move. Use when building technology-enhanced formative assessment cycles.
Audit and redesign AI-generated feedback for pedagogical quality, timing, and learning impact. Use when building or reviewing automated feedback in digital learning tools.
Design deliberately flawed examples that develop error-detection skills and deepen understanding. Use when students make characteristic errors and need practice spotting mistakes.
Script a multi-turn tutoring dialogue with branching responses for anticipated student difficulties. Use when designing AI tutors, chatbot interactions, or structured one-to-one support scripts.
Interpret learning analytics data and translate dashboard findings into actionable teaching decisions. Use when reviewing LMS data, quiz patterns, or engagement metrics.
Create self-explanation prompts that deepen understanding of worked examples, texts, or diagrams. Use when students read material passively without engaging with underlying principles.
Redesign a direct instruction sequence to include productive struggle before the explanation phase. Use when teaching concepts that benefit from failure-first approaches.
Design metacognitive checkpoints that prevent AI-assisted learning from bypassing genuine understanding. Use when students use AI tools and may overestimate their own comprehension.
Design the transition from worked examples to independent problem-solving using expertise-reversal principles. Use when students follow examples but cannot solve problems alone.
Design a Funhouse Mirror activity where students use their own domain expertise to detect AI distortions, omissions, and overconfidence. Use when students know a subject well enough to evaluate AI claims about it.
Design a fact-checking protocol for AI-generated text, extending SIFT with AI-specific adaptations for hallucination detection. Use when students need to verify AI claims and citations.
Claude Skills — questions
Answers built from the skills we actually parsed.