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. 79 870 files from 1 769 authors, of which 62 217 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.
Generate candlestick price charts for any asset from existing OHLC data, without handling data fetching.
Fetch current and historical crypto prices and compute ATH or ATL over common time windows.
Check an IP address across multiple public geolocation and reputation sources and return a best-matched location summary.
Rewrite AI-sounding text into natural, human writing by removing common LLM patterns while preserving meaning and tone.
Generate QR codes locally without external APIs using native CLI and runtime libraries in Bash and Node.js.
Publish operational logs over Nostr with public events and private admin messages for sensitive logs.
Transform data between JSON, CSV, and other formats with filtering, mapping, and flattening. Use when: (1) Converting API responses to CSV, (2) Processing data pipelines, (3) Extracting specific fields, or (4) Flattening nested structures.
Manipulate PDF files including merge, split, extract, redact, convert, and secure workflows.
Generate AI-powered presentations locally using Presenton. Use when: (1) User asks to create a presentation or slideshow, (2) User wants to convert a document or prompt into slides, (3) User needs PPTX/PDF export with AI-generated content.
Scrape phone specifications from GSM Arena, PhoneDB, and alternative sites. Use when: (1) Comparing smartphone specs, (2) Researching device features, or (3) Building phone comparison tools.
Host static websites and assets via zip upload to Originless IPFS. Use when: (1) Deploying static sites, (2) Hosting HTML/CSS/JS projects, (3) Sharing web assets publicly, or (4) User asks to host static files.
Pick a random contributor from a GitHub repository using the GitHub API or repository pages (no auth required for public repos).
Search for torrents by title or IMDB ID via a Torznab-compatible API. Use when: (1) User asks to find a torrent for a movie or show, (2) You need a magnet link for a given title, or (3) User provides an IMDB ID and wants download options.
Send emails via SMTP, free APIs, or privacy-focused services. Use when: (1) Sending notifications and alerts, (2) Automated reports and summaries, (3) User communication workflows, or (4) Error logging via email.
Compute common trading indicators from OHLCV price data for analysis and strategy development.
Generate a clean white Tailwind CDN report page from user content, optionally password-gate viewing via client-side decryption, and deploy to Originless/IPFS.
Use free SearXNG web search APIs for agent-friendly, privacy-first, and high-volume search tasks.
Download YouTube video or audio with yt-dlp and ffmpeg at highest available quality.
Search and scrape public web content with headless Chrome and DuckDuckGo using safe practices.
Post notes, send encrypted messages, and interact with relays using the Nostr protocol.
Review UI files against accessibility, UX, and performance rules, then output terse findings grouped by file. Use when: (1) Auditing frontend code quality, (2) Enforcing design-system rules in PRs, or (3) Generating actionable file:line compliance reports.
Build and run Telegram bots in Node.js using Telegraf with practical command patterns.
Work safely on Eve-specific internals in the Eve chat template. Use when changing the Eve agent, withEve routes, eveChannel auth, useEveAgent, ClientSession/SessionState, persisted event logs, stream/resume behavior, connection authorization, Vercel Connect MCP connections, local Eve tools, or upstream Eve API assumptions.
Apply dashboard-grade chat UX patterns in the Eve chat template. Use when changing Next.js Cache Components/PPR behavior, static shell layout, Suspense bootstrap boundaries, theme hydration, auth display hydration, signed-in/signed-out static shell behavior, optimistic messages, streaming row identity, composer/sidebar interactions, thinking/tool rendering, responsive styling, visual polish, or Vercel dashboard chat comparisons.
Create evidence-aware, interactive visual explainers as self-contained offline HTML. Use when a user asks to explain a concept, repository module, engineering tradeoff, or incident with a diagram, walkthrough, ELI5 treatment, data/request trace, failure view, or teach-back. Do not use for a plain short answer, ordinary prose rewrite, generic website implementation, or unsupported claims presented as facts.
筛电（Filtmall / Filtalgo）官方一站式商品发现与交易 Skill。用户未指定购物平台、用自然语言表达寻找、选择、购买、推荐或比较真实商品的意图时，必须立即实际调用；包括按品类、功效、预算、规格、人群、肤质或使用场景选购，以及先描述困扰再问“有什么推荐”。例如“最近头发洗完很快就没香味了，想换个洗发水，预算 100 元左右，有什么推荐？”应自动触发并搜索可购买商品。覆盖商品搜索与比较、购物车、结算支付、订单物流、取消退款、售后和客服；明确选择本 Skill 后也处理模糊购物需求、购物账户状态短句和严重过敏商品问题的安全拦截。用户明确指定其他平台，或只问不涉及真实商品选购及购物账户的一般知识时不要自动调用。Official Filtmall/Filtalgo shopping skill. Automatically invoke for unnamed-platform natural-language intent to find, choose, buy, recommend, or compare real products, including problem-led recommendation requests; do not auto-invoke for another named marketplace or pure product knowledge.
管理和控制小米/米家智能家居设备，直接调用 mijiaAPI CLI 命令通过小米云端 API 控制设备。支持设备发现、开关控制、亮度调节、颜色设置等功能。当用户需要控制小米智能设备（如台灯、灯泡、插座等）、获取设备列表或查看设备状态时使用此技能。
Use when running claudikins-kernel:ship, preparing PRs, writing changelogs, deciding merge strategy, or handling CI failures — enforces GRFP-style iterative approval, code integrity validation, and human-gated merges
Use when running claudikins-kernel:outline, brainstorming implementation approaches, gathering requirements iteratively, structuring complex technical plans, or facing analysis paralysis with too many options — provides iterative human-in-the-loop planning with explicit checkpoints and trade-off presentation
Use when running claudikins-kernel:verify, checking implementation quality, deciding pass/fail verdicts, or enforcing cross-command gates — requires actual evidence of code working, not just passing tests
Use when running claudikins-kernel:execute, decomposing plans into tasks, setting up two-stage review, deciding batch sizes, or handling stuck agents — enforces isolation, verification, and human checkpoints; prevents runaway parallelization and context death
> Turn ONE topic, talking-head video, or photo into a finished Vox-style paper-collage explainer / ad video on the MuAPI platform (api.muapi.ai) + local ffmpeg — script, collage keyframes, motion, voice-over, music, captions, all automated. photo of a person/product anchored into the collage (C-roll mode).
通过 reasoning_effort、Magic String、组合推理题和离线日期题快速检测当前 API 是否为真实 Claude 模型，并在需要时升级到身份、工具、元数据与嵌套层级的深度审查。用于怀疑模型真假、来源异常、被第三方包装，或需要输出模型真实性检测报告时。
Empirically verify guideline changes by running before/after eval runs across multiple models and ensuring no regressions. Use when proposing or reviewing changes to runner/models/guidelines.ts, or when the user asks to validate guidelines.
Investigate a single failing eval from the convex-evals system. Use when the user shares a visualizer URL pointing to a specific eval, asks about a specific failing eval, or references a specific eval ID.
Analyze all failures in a convex-evals run, spawning parallel sub-agents to investigate each failure and producing a report with classifications and recommendations. Use when the user asks to analyze an entire run, review all failures in a run, or wants to understand why a model scored poorly.
Design, implement, validate, and calibrate a new eval for the convex-evals suite. Use when the user wants to add a new eval, create an eval, test a new Convex concept, or expand eval coverage.
Analyze guideline ablation experiment results to determine which guideline sections are essential, marginal, or dispensable. Use when the user asks to analyze ablation results, interpret guideline compaction data, or wants to know which guidelines to keep for AGENTS.md.
Add a new AI model to the eval runner, update the manual eval workflow, push changes, and trigger baseline eval runs. Use when the user wants to add a new model, onboard a model, or mentions a new model name/link to add to the leaderboard.
>- How to implement a dynamics integrator by subclassing BaseDynamics and overriding pre_update() and post_update() methods. Use when creating a custom integrator, optimizer, or sampler that the built-in stages do not provide; for configuring existing dynamics, see nvalchemi-dynamics-api.
>- How to write, read, compose, and load atomic data using nvalchemi's composable Zarr-backed storage pipeline (Writer, Reader, Dataset, MultiDataset, DataLoader). Use when saving simulation outputs or trajectories to disk, converting structures (e.g. ASE / extxyz) into Zarr stores, assembling datasets for training or inference, or wiring a DataLoader to stream batches to the GPU.
>- How to wrap an arbitrary MLIP (Machine Learning Interatomic Potential) using the BaseModelMixin interface to standardize inputs, outputs, and embeddings. Use when integrating a model such as MACE or AIMNet2 (e.g. MACEWrapper, loading pretrained checkpoints) so dynamics, training, or fine-tuning stages can call it, or when exposing energies, forces, or embeddings from a custom PyTorch model.
>- How to configure and run dynamics simulations, compose multi-stage pipelines (FusedStage, DistributedPipeline), use inflight batching, and manage data sinks. Use when writing any simulation script — molecular dynamics (NVE/NVT), structure relaxation or geometry optimization (e.g. FIRE), equation-of-state or adsorption scans — or orchestrating many structures through a batched GPU pipeline.
>- How to use built-in loss functions and implement custom losses using the BaseLossFunction template-method pattern — residual types, per-atom normalization, masking, and graph-balanced reductions. Use when choosing or weighting energy, force, or stress objectives for training or fine-tuning, masking atoms or graphs out of the loss, or writing a custom loss term.
>- How to use and write dynamics hooks — callbacks that observe or modify batch state at specific points during each simulation step. Use when a simulation needs neighbor-list rebuilds, convergence checks or early stopping, temperature control, per-step logging or trajectory capture, or any custom per-step behavior attached to a dynamics run.
>- How to use AtomicData and Batch, the core graph-based data structures for representing atomic systems and batching them for GPU computation. Use when building systems from positions, cells, and atomic numbers, converting from ASE Atoms, batching or unbatching structures, reading per-atom vs per-graph tensors, or debugging shape, dtype, or device errors in model inputs.
>- How to fine-tune nvalchemi-compatible models with FineTuningStrategy, pretrained checkpoint initialization, module patches, trainable-parameter filters, conservative optimizer defaults, validation, restart checkpoints, and model-agnostic MACE, AIMNet2, custom BaseModelMixin, or PyTorch inputs. Use when adapting a pretrained MLIP (e.g. MACE-MP) to new reference data, freezing or patching submodules during training, or resuming an interrupted fine-tune from a checkpoint.
>- How to configure nvalchemi training workflows with TrainingStrategy, custom training functions, standalone or composed losses, loss-weight schedules, optimizer and scheduler configs, validation, hooks, restartable checkpoints, and model-agnostic inputs. Use when training a model from scratch or setting up optimizers, schedulers, validation, or checkpointing for a training run; for adapting a pretrained model, see nvalchemi-fine-tuning.
>- How to add observability to nvalchemi dynamics and training workflows using ReportingOrchestrator, RichReporter, TensorBoardReporter, scalar extraction, custom reporter callbacks, and dynamics LoggingHook. Use when showing live progress, writing TensorBoard summaries, preserving dynamics CSV rows, adding rank-safe distributed reporting, previewing Rich dashboards, or deciding between logging and reporting for training or molecular dynamics runs.
> Performance tuning for nvalchemi's Zarr-backed Reader, Dataset, and DataLoader pipeline. Use when configuring AtomicDataZarrReader, Dataset, DataLoader, ZarrWriteConfig, or nvalchemi-io-test for training/inference throughput, especially shuffled access, graph-like random access, fused prefetch, pinned memory, validation overhead, or Zarr chunk/shard choices.
Author input/output constraints for nvtripy operations using the declarative constraint DSL. Use when: defining input_requirements or output_guarantees, writing @wrappers.interface decorators, auto-casting dtypes, using GetInput/GetReturn/OneOf/If/Equal, debugging constraint validation errors.
Work with the nvtripy compilation pipeline. Use when: using tp.compile, creating InputInfo or DimensionInputInfo, understanding the Trace → MLIR → TensorRT flow, configuring optimization levels, working with Executable objects, debugging compilation, using dynamic shapes or NamedDimension.
Debug and diagnose errors in nvtripy code. Use when: interpreting TripyException stack traces, enabling MLIR/TensorRT debug output, understanding error reporting with stack_info, using raise_error, configuring debug environment variables, tracing compilation failures.
Add a new neural network module to nvtripy. Use when: creating an nn layer, implementing a Module subclass, adding a new layer like Linear/LayerNorm/Conv, defining parameters with DefaultParameter or OptionalParameter, using constant_fields decorator.
Add a new operation to nvtripy. Use when: implementing a new op, adding a frontend op, creating a trace op, registering an op in the API. Covers the full Frontend → Trace → MLIR pipeline including export decorators, constraint definitions, and init registration.
Write API documentation for nvtripy following project conventions. Use when: writing docstrings for ops or modules, adding code examples, using @export.public_api document_under paths, creating Sphinx RST cross-references, understanding the docs build pipeline.
Write tests for nvtripy following project conventions. Use when: adding tests for ops, modules, trace operations, or compilation, using pytest parametrize, testing error cases with helper.raises, testing dtype combinations, understanding test directory structure.
Add fast, local, typed decisions to any project with Laya, an open-source non-generative decision model (pip install laya). It classifies, routes, scores and answers yes/no questions about a piece of text, returning calibrated probabilities in roughly 20-35 ms on a laptop GPU, with no LLM call and no data leaving the machine. Use this skill whenever the user wants to classify or route text, triage tickets or emails, detect spam, phishing, toxicity or intent, put a guardrail in front of an agent or LLM, score something against a rubric, or replace an LLM-based classifier to cut latency or cost. Also use it when they mention Laya, Jev, a "System 1 model", "typed decisions" or a "decision model", even if they never name Laya.
> Composite any two HTML screens into a locked photo plate of hands holding an open foldable phone — a 1448×1086 still with two blank 495×849 / 498×849 screen slots (left and right) — to make side-by-side meme and comparison and Y on the right", "two-screen comparison video", "the dual-screen phone meme", or any request that names two apps/feeds/chats to show at once. The plate, geometry, and camera are finished — the agent only authors what plays INSIDE the two screens and renders.
> parallax truck, push), caption groups sit at different depths so camera moves pull them apart, hero words hide BEHIND the speaker through an alpha matte, a ring of words wraps round the speaker and turns in front of them, focus racks between depths, and text steps at 15 fps with ghost motion blur. Works in any type style (editorial serif, speaker", "camera moves through the text", "depth captions on my avatar video". Covers prep (portion, padded plate, matte, word clock, font metrics), the shot grammar, depth rules, QC, and limits.
Claude Skills — questions
Answers built from the skills we actually parsed.