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. 79 600 files from 1 763 authors, of which 61 947 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.
Instrument a new AI agent in a Python codebase for Monte Carlo Agent Observability. Detects AI libraries, installs the Monte Carlo OpenTelemetry SDK, and proposes tracing setup and decorator placements as diffs. Asks before editing any file.
Create, edit, validate, and import Monitors-as-Code YAML files. CLI-first; falls back to MC MCP tools, then manual validation.
Guide users from coverage analysis to monitor creation. USE WHEN user asks what should I monitor, where are my gaps, improve coverage, or wants a systematic approach to monitoring across their data estate.
| Diagnoses pipeline performance issues -- slow jobs, expensive queries, latency trends -- using Monte Carlo's cross-platform observability. affected tables, then drill into root causes. Activates when a user asks about slow pipelines, expensive queries, or performance regressions.
| Reinforces an AI agent by turning Monte Carlo's reinforcement loop diagnosis into code fixes. Reads the daily reinforcement loop report for an agent's workflows, ranks the diagnosed issues, proposes what to fix, and — with the user's approval at each step — opens a pull request. Activates on "fix my agent", "improve my agent's health", "reinforce my agent", "what should I fix in my agent". Not for investigating a specific agent alert or trace (monte-carlo-troubleshoot-agent-traces), creating agent monitors (monte-carlo-monitoring-advisor), or instrumenting a new agent (monte-carlo-instrument-agent).
Shift-left safety net for dbt/SQL model edits. Runs change impact assessment before edits, generates SQL validation queries after, and executes them via `/mc-validate run`. Delegates health and monitor creation to peer skills.
Analyze data coverage, create monitors for warehouse tables and AI agents. Covers coverage gaps, use-case analysis, data monitor creation, and agent observability.
> Expert guide for Monte Carlo's push ingestion model. Use this skill whenever a customer build me a collection script, push metadata/lineage/query logs, invocation_id tracing, custom lineage nodes or edges, deleting push tables, or any question about why pushed data is not showing up. Also trigger when they ask to generate code that collects metadata, table schema, row counts, freshness, lineage, or query history from any data warehouse or data source and sends it to Monte Carlo. If the user mentions any warehouse, database, or data platform alongside any Monte Carlo topic, this skill is almost certainly relevant.
Analyze a warehouse for stale, unused, or redundant tables via the analyze_storage_costs MCP tool. Classifies waste patterns and table categories, computes safety tiers, and handles category drill-downs and lineage follow-ups.
Investigate and remediate data quality alerts using Monte Carlo MCP tools. Runs root cause analysis, assesses blast radius, discovers available tools (MCP/CLI/API), proposes and executes fixes, or escalates with full context when uncertain.
Analyze a Monte Carlo monitor and recommend config changes to reduce alert noise. Supports metric, custom SQL, validation, table, and agent (metric, evaluation, trajectory, validation) monitors. Fetches the report, identifies patterns, and suggests tuning.
Troubleshoots Monte Carlo AI agent alerts and traces — eval score drops, latency/token spikes, trajectory and validation breaches. Not for data incidents (monte-carlo-analyze-root-cause) or monitor creation (monte-carlo-monitoring-advisor).
Interactive copilot that offloads grunt work to a cheaper engine while Claude orchestrates. Greets each session with your live token limits and a driving mode (auto-pilot / you-drive / hybrid), auto-conserves when your Claude window runs hot, and picks models by live benchmarks. Triggers: outsource, offload, delegate, conserve tokens, use GLM/Devin/Codex, or get a second opinion.
Process and analyze CSV, JSON, and text files with data transformation, cleaning, analysis, and visualization capabilities
> Automate Adobe After Effects via ExtendScript. Use when the user asks to create, modify, or query anything in an After Effects project — layers, keyframes, expressions, effects, compositions, assets, rendering, batch operations. Generates and executes JSX ExtendScript via osascript on macOS.
>- Agent-native DeepRefine refinement loop — same control flow as DeepRefine.refine(), graphify search instead of FAISS, session LLM, dry-run review before approved graph writes.
>- Agent-native DeepRefine refinement loop — same control flow as DeepRefine.refine(), graphify search instead of FAISS, session LLM, dry-run review before approved graph writes.
>- Claude Code adapter for the DeepRefine agent-native refinement loop. Use when the user invokes /deeprefine, or asks to refine, diagnose, review, or apply changes to a Graphify / LLM-Wiki knowledge graph. Must follow the canonical DeepRefine skill rules and stop for review before graph writes.
Use when audited OpenClaw conversations or outgoing Feishu/Slack/channel messages contain concrete business numbers such as revenue, percentages, stock prices, contract values, costs, market share, or funding and need evidence checks plus red/yellow/green audit stamps after sends.
Adds freely-licensed maps and photos to textbook chapters by sourcing from Wikimedia Commons and US government archives, then inserting them with captions and proper attribution.
Scaffolds a new intelligent textbook project from scratch — mkdocs.yml, docs/ directory tree, starter pages, and license files. Use at the very start of a new project, before chapters, learning graph, or MicroSims exist.
Generates a status report of all diagrams and MicroSims across an intelligent textbook's chapters, including difficulty, Bloom's level, and UI complexity. Use to audit visualization coverage before a content review.
Creates interactive causal-loop diagram (CLD) articles for MkDocs Material sites. Use when the user wants to visualize feedback loops, systems archetypes (reinforcing/balancing loops, limits to growth, tragedy of the commons), or any topic with competing runaway and stabilizing dynamics. Produces vis-network diagrams — not static images or Mermaid.
Generates a 13-slide LinkedIn carousel (a "document post" — PPTX/PDF) that showcases an intelligent textbook's key features, with real screenshots, mascot art, and metrics pulled from the project. Use when the user wants a LinkedIn slideshow/carousel/document post for a textbook, as opposed to linkedin-announcement-generator which produces post *text* only.
Generates an interactive diagram MicroSim with numbered callout markers or rectangular zones over a scientific illustration, supporting explore, quiz, and edit modes. Use for anatomy diagrams, labeled components, or comparison posters needing interactive annotations.
Generates an interactive Docker Python lab block for MkDocs textbook pages, where students write and run real Python code inside an isolated Docker container. Use this skill whenever someone asks to add a Python lab, code runner, interactive Python exercise, or runnable code block to a textbook page that uses Docker (not Skulpt). Also use it when adding multiple labs to a single page, setting up the shared CSS/JS infrastructure, or creating a timing/benchmark lab that shows students how long each phase takes. Always invoke this skill instead of writing docker lab HTML by hand.
Generates a LinkedIn post announcing an intelligent textbook milestone by pulling book metrics, chapter count, and key statistics. Use when announcing textbook completion or major content milestones on social media.
Convert text to speech using ElevenLabs voice AI. Use when generating audio from text, creating voiceovers, building voice apps, or synthesizing speech in 70+ languages.
Generates slide decks in the MARP (Markdown Presentation Ecosystem) format and publishes them as live, embeddable presentations on an MkDocs Material site. Use this skill whenever the user wants to turn a topic, description, or existing chapter/document into a presentation, slide deck, or set of slides — trigger on phrases like "make a slide deck", "turn this chapter into a presentation", "create slides for X", "I need to present this", "build a deck about Y", or "add a presentation to the site", even if the user never says "MARP" by name. Produces a self-contained docs/slides/<deck-name>/ directory (MARP source, exported HTML, thumbnail, documentation page), adds the deck to the docs/slides/index.md gallery, and updates mkdocs.yml nav.
Creates or updates the GitHub README for a textbook project with badges, project overview, site metrics, and getting-started instructions.
Generates an AP-style press release for an intelligent textbook or course milestone — headline, lead, attributed quotes, boilerplate, and media-contact block. Use when pitching to journalists or education trade press; distinct from linkedin-announcement-generator (social post) and readme-generator (GitHub docs).
> Generate an MP3 pronunciation of a glossary term using ElevenLabs TTS API and insert a "Pronounce" button into the term's entry in a markdown file. Trigger when the user says "Create a pronounce button for the term X" or "Add pronunciation for X". Defaults to glossary.md if no file is specified.
Wires Google Analytics 4 into an MkDocs Material site — creates a GA4 property, writes the Measurement ID into mkdocs.yml, verifies the tag, and deploys. Use when adding analytics to a textbook for the first time.
Generates a lecture PowerPoint from an intelligent textbook using pptxgenjs — visual design, 4-act storytelling structure, and speaker notes. Use when converting a textbook into slides for classroom delivery.
Generates illustrated graphic novel narratives about scientists, mathematicians, and historical figures for intelligent textbooks, with image prompts for each panel. Use when adding historical-figure or case-study stories to a textbook's Stories section.
Generates fact-checked infographic posters by verifying every statistic against peer-reviewed sources before producing the image prompt. Use for any poster or infographic containing numeric claims or cited data; skip for purely decorative images.
Generates media for intelligent textbooks - slide decks and presentations (MARP web decks in docs/slides/ or PowerPoint .pptx lecture downloads), illustrated stories and graphic novels, fact-checked infographic posters, freely-licensed chapter images from Wikimedia and government archives, and audio (text-to-speech voiceovers, glossary pronounce buttons via ElevenLabs). Routes to the appropriate media guide.
Designs the chapter structure for an intelligent textbook by analyzing the learning graph and concept dependencies. Use after the learning graph is complete and before generating chapter content.
Publishes and promotes a finished intelligent textbook - GitHub README with badges and site statistics, LinkedIn announcement posts, LinkedIn carousel document posts (PPTX/PDF slideshows), and AP-style press releases. Use when announcing a book milestone or updating repository documentation. Routes to the appropriate guide.
Installs and configures intelligent-textbook infrastructure - scaffold a brand-new MkDocs Material textbook (init textbook), install any of 40 features (math, mascot, learning graph viewer, Google Analytics GA4, custom 404, kanban board), and generate book metrics. Routes to the appropriate installation guide.
Generates detailed chapter content for an intelligent textbook — text, diagrams, MicroSims, and exercises at the appropriate Bloom's level. Use when a chapter's index.md exists with title, summary, and concept list.
Validates or creates a course description for an intelligent textbook, scoring completeness against required elements (title, audience, prerequisites, topics, Bloom's Taxonomy outcomes). Use before running the learning-graph-generator.
Converts .docx files (papers, briefs, reports) into styled React component pages in a Next.js content-catalog site. Use when publishing a Word document as a structured web page.
Generates a FAQ set for an intelligent textbook from course content, learning graph, and glossary terms. Use after the learning graph and glossary exist and at least 30% of chapters are written.
Utility tools for MicroSim management including quality validation, screenshot capture, icon management, index page generation, iframe height synchronization, iframe control-visibility testing, visual layout review, and diagram/MicroSim coverage reports across chapters. Routes to the appropriate utility based on the task needed.
Creates interactive educational MicroSims, routing to the best-matched generator - p5.js, Chart.js, Plotly, Mermaid, vis-network, timelines, maps, Venn, causal-loop/feedback-loop diagrams (CLD), concept-classifier sorting quizzes, infographic overlays with callout labels, and Docker Python labs (runnable code blocks). Generates complete MicroSim packages with HTML, JavaScript, CSS, documentation, and metadata.
Generates a glossary from the learning graph's concept list with ISO 11179-compliant definitions (precise, concise, non-circular). Use after the learning graph concept list is finalized.
Generates 10 curated academic references per chapter, prioritizing Wikipedia plus credited textbook authors known for innovative explanations, with relevance descriptions, stored in chapter references.md files. Use when an intelligent textbook chapter needs citations.
Generates a comprehensive learning graph from a course description, including 200 concepts with dependencies, taxonomy categorization, and quality validation reports. Use this when the user wants to create a structured knowledge graph for educational content.
Generates multiple-choice quiz questions for each chapter, aligned to the learning graph and distributed across Bloom's Taxonomy levels. Use after chapter content and the learning graph both exist.
This skill should be used when the user asks to "connect to Trading 212", "authenticate Trading 212 API", "place a trade", "buy stock", "sell shares", "place market order",, "place pending order", "place limit order", "cancel order", "check my balance", "view account summary", "get positions", "view portfolio", "check P&L", "find ticker symbol", "search instruments", "check trading hours", "view dividends", "get order history", "export transactions", "generate CSV report", or needs guidance on Trading 212 API authentication, order placement, position monitoring, account information, instrument lookup, or historical data retrieval.
Design production-ready prompts, shot plans, storyboards, extensions, and video edits for Jimeng Seedance 2.5. Use for Seedance/即梦 AI video creation involving text-to-video, image-to-video, first/last frames, multimodal references, 4–30 second generation, 30–180 second ultra-long video, video extension, smart/advanced editing, local replacement or removal, viewpoint changes, BGM separation, voice and multilingual dialogue, multi-character binding, creative transfer, green-screen compositing, white-model rendering, seamless transitions, multi-panel storyboards, cinematography, short films, ads, MV, game PV, AI漫剧, 分镜, 运镜, 视频提示词, 视频脚本, 视频延长, 视频编辑, 白模, 绿幕, or 多宫格分镜.
Authoritative architecture reference for Lagune, covering repository layout, the command/template split, the core/adapter boundary, what it scaffolds, and the tracking-map model. Use before adding an agent or when a decision depends on repo shape.
Authoritative reference for writing Lagune's prose. Use before writing, editing, or translating any artifact, charter, command text, doc, or user-facing message.
Authoritative reference for the Lagune dashboard, a live view of a project's .lagune/ chain with a locked-down local action surface. Use before changing anything under src/dashboard/ or src/types/dashboard/.
Design engineering principles for making interfaces feel polished. Use when building UI, reviewing frontend code, or working on any visual detail, from animations, hover states, shadows, borders, and typography to optical alignment and tabular numbers.
Authoritative engineering reference for Lagune, covering the toolchain, code conventions, type rules, the build and distribution path, and how the tracking hooks work. Use before writing or changing source under src/ or test/, or before the build.
How to simulate a Lagune command end to end so the user sees both the process and the results in chat. Use when the user asks to simulate, demo, preview, run, or see in action any lagune command. Read this before attempting any such simulation.
Author a new built-in Lagune sub-skill inside the Lagune source, not a scaffolded `.lagune/` target. Use when adding or refining a security knowledge module that ships with Lagune, against the native layout (`spec/skills/*.md` plus the catalog).
Claude Skills — questions
Answers built from the skills we actually parsed.