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.
Build and verify a PII gold set with HUMAN annotators (first-class). Launch the browser annotator, label spans per the codebook, export per-annotator label files, then compute inter-annotator agreement (Cohen's/Fleiss' kappa) and draft an adjudicated gold. Use when the user says "annotate PII", "label this transcript", "build a gold set", "inter-annotator agreement", "review annotations", "adjudicate labels", or wants to measure/defend a de-identification gold standard. Local-only: synthetic or consented data only; annotators' names and transcript text stay on the machine — only labels/stats are collected, nothing PII is re-shared.
Query Chrome browsing history with natural language. Filter by date range, article type, keywords, and specific sites.
Query browsing history from all synced devices (iPhone, Mac, iPad, desktop). Supports natural language queries for filtering by date, device, domain, and keywords. Uses LLM classification for content categories. Can output to stdout or save as markdown/JSON to Obsidian vault.
De-identify a session transcript (file or folder) by redacting PII LOCALLY before any sharing or cloud use. Produces a redacted GREEN copy with unique reserved-sentinel placeholders ([CONFIDE_PERSON_0001], [CONFIDE_EMAIL_0001], [CONFIDE_DATE_0002]...) plus a counts-only stats summary, and a local secret <name>.map.json (0600, gitignored) that enables confide:rehydrate to restore real values after a cloud analysis. Use when the user says "anonymize this transcript", "redact PII", "de-identify session", "make safe to share", "strip personal data", "anonymize notes before sending to an LLM", or points at a transcript/folder that should be scrubbed. Local-only by default — raw text never leaves the machine; the map is the only artifact with originals and stays local; nothing printed is PII; human review is still required before sharing.
Run a corpus-scale, STATS-ONLY PII audit over a folder of session transcripts LOCALLY and produce an aggregate report — counts by type and by layer, the per-session redaction-rate distribution, document lengths, and a coarse residual proxy. Use when the user says "audit my sessions", "scan folder for PII", "how much PII across these transcripts", "PII stats for my corpus", "is my redaction holding at scale", or points at a directory of transcripts and asks how much personal data it contains. Fully local — raw text never leaves the machine; the report carries ZERO PII values, transcript substrings, or filenames (only anonymized own-NN ids and counts), so the aggregates are safe to surface. Run it on a RED (raw) corpus to size the PII, or on a GREEN (already-redacted) corpus to check residual leakage.
This skill should be used to summarize coaching or therapy session transcripts after a Fathom/Granola sync. The agent analyzes the transcript itself (no API key, runs on the subscription) and appends key insights, decisions, action items, and trail connections. Supports quick extraction or deep analysis with cross-session pattern detection.
>- Residual re-identification RISK CHECK on text you have ALREADY redacted (defensive, dual-use). Use when the user asks to "check residual re-id risk", "red-team my redaction", "what can an attacker still infer", "is this safe to share", or assess "re-identification risk" after anonymizing. Re-runs the CONFIDE detectors on the redacted output to surface surviving identifiers (singling-out), checks multiple files for linkability, and optionally probes a local model for still-inferable attribute CATEGORIES (inference) — mapped to GDPR Art-29. Reports risk categories/counts only,
Set up, install, and configure CONFIDE local de-identification — installs Python deps (natasha, scrubadub, phonenumbers, pymorphy2), ensures Ollama + pulls the default qwen2.5:3b model, detects optional llama.cpp, and writes the optimal-default config so confide:anon and confide:red work with zero further config. Everything is local-first; raw text never leaves the machine. Use when the user says "set up confide", "install confide", "configure confide de-id", "confide setup", or "get confide ready".
Put the real values back into an analysis that was produced from GREEN (placeholder) text — LOCALLY, using the user's own reversible map. Completes the confide round-trip (redact -> cloud-analyze the green -> rehydrate locally). Use when the user says "rehydrate", "restore real names", "unmask the analysis", "put the names back", "de-redact this output", "reverse the placeholders", or hands you an analysis full of [CONFIDE_PERSON_0001]/[CONFIDE_DATE_0002] plus a *.map.json. Runs only on the user's own map; the map never leaves the machine; nothing fetched or transmitted. Prints counts only — never echoes restored PII. Warns on placeholders not in the map (possible LLM hallucination).
Build a self-contained interactive HTML that lets you SEE de-identification and restoration — original ↔ redacted ↔ rehydrated, with color-coded PII spans and All/None/Selected toggles. Use when the user says "show me what was redacted", "visualize the de-id", "compare original and redacted", "highlight the PII", "view redaction diff", "see what rehydrate restored", or wants to inspect exactly which spans each type/layer removes. LOCAL only — the HTML embeds REAL values (like a vault artifact): written locally, gitignored, banner-marked private, never shipped or committed.
>- Set up and verify the CONFIDE THREE LOCKS for storing RED (real, identifiable) session data at rest — device FileVault, a dedicated encrypted store, and per-file sops/age encryption. Use when the user says "set up confide vault", "encrypt my session data", "three locks", "secure store for transcripts", "sops/age for RED data", or asks how to status and prints the EXACT command to fix any gap; it never moves, deletes, or encrypts data, and never runs `fdesetup enable`/`hdiutil`/`age-keygen` without an explicit flag and your confirmation. Probes are read-only (`fdesetup status`, which sops/age, key path).
Use when the user explicitly asks to prepare or bump a Cull patch, minor, or major release, curate its changelog and compatibility review, or create its focused release commit without publishing.
Generate interactive AI transformation context-builder prompts for consulting clients. Use when creating structured discovery session prompts that guide a company through context gathering about their business, pain points, tech stack, and AI opportunities. Produces a resumable, multi-section prompt with Express/Deep Dive modes.
Use when assessing whether Cull is ready to release, auditing or dry-running a Cull release, identifying version blockers, or before any Cull prepare or publish operation.
This skill should be used when the user wants to view, review, rate, organize, search, or export images / AI-art generations with the Cull app. Trigger on "show me these images", "review this batch", "open these in Cull", "rate / shortlist / collect these", "find similar images", "make a smart collection", "run a quality pass", "export the keepers", "publish this collection". Works via the `cull` CLI by default (no MCP required); the `mcp__cull__*` tools are optional for richer interactive control.
Use when a Cull release is stuck, inconsistent, failed in a workflow or artifact gate, missing Homebrew promotion, failed after publication, or needs a safe recovery or patch plan.
Use when the user explicitly asks to publish or complete a prepared Cull release, or when the authorized cull-release orchestrator reaches publication after every repository and artifact gate passes.
Use when verifying or auditing a Cull release, DMG, updater archive, notarization, Homebrew cask, installed version, launch health, or post-publication distribution state.
Use when orchestrating or resuming Cull's complete release cycle, reporting its current release state, or fulfilling an explicit Cull patch, minor, or major release request through verified GitHub and Homebrew distribution.
Evidence-based health research using tiered trusted sources with GRADE-inspired evidence ratings. Integrates Apple Health data for personalized context. Use when user asks health, nutrition, exercise, sleep, or wellness questions.
Audit the Claude Code ecosystem — skill health and staleness, project activity pulse, CLAUDE.md instruction drift, Mac Mini service status. Use this skill whenever the user asks about ecosystem health, stale skills, abandoned projects, system status, infrastructure check, "what's broken", "what's stale", "how's my setup", or any request to review the state of their Claude Code environment. Also triggers on "/ecosystem", "audit my ecosystem", "ecosystem health", "ecosystem audit".
This skill should be used to set up, validate, resolve, and export design tokens following the DTCG (Design Tokens Community Group) Format Module 2025.10 standard. Use when the user wants to define a design token set globally or per project, compile tokens to CSS variables, layer a project's tokens over a global brand base, or produce an on-brand context file for other generation skills. Triggers on "set up design tokens", "create a token set", "compile tokens to CSS", "design system variables", "brand tokens".
This skill should be used for elimination-style research where the user wants to choose from a shortlist of products, tools, services, vendors, or other options using explicit criteria, numeric evidence, tournament-style comparison, source/domain classification, image-supported consumer reports, raw data tables, and ownership-cost estimates for replaceable parts. Use this skill whenever the user asks to compare options, buy something, shortlist candidates, rank alternatives, generate a "don't make me think" report, or produce a full audit report with raw numeric data.
This skill converts text to high-quality audio files using ElevenLabs API. Use this skill when users request text-to-speech generation, audio narration, or voice synthesis with customizable voice parameters (stability, similarity boost) and voice presets (rachel, adam, bella, elli, josh, arnold, ava).
Fetch meetings, transcripts, summaries, and action items from Fathom API. Use when user asks to get Fathom recordings, sync meeting transcripts, or fetch recent calls.
This skill should be used when taking a single software feature from intent to shipped as a solo developer — goal-first, TDD, deterministic verification, evidence only where it earns its keep, and human judgment at the two moments that matter (goal approval, merge). Trigger when the user says "let's build feature X", "ship this feature", "run this through the factory", "write a goal contract", "feature-factory", or wants a disciplined intent→merge loop that resists process bloat. NOT for whole-product planning, multi-feature roadmaps, or autonomous multi-agent swarms.
This skill should be used when the user requests to research topics using FireCrawl, enrich notes with web sources, search and scrape information, or write scientific/academic papers. It extracts research topics from markdown files, creates research documents with scraped sources, generates BibTeX bibliographies from research results, and provides Pandoc/MyST templates for academic writing with citation management.
This skill should be used when inspecting or applying advanced OpenType features of a font (woff2/otf/ttf) — ligatures, stylistic sets (ss01–ss20), character variants (cvXX), texture healing, slashed zero, tabular/oldstyle figures, fractions, small caps, case-sensitive forms — and generating the CSS to enable them. Interviews the user via cenno to pick features. Triggers on "OpenType features", "font features", "stylistic sets", "ligatures", "texture healing", "tabular figures", "what can this font do".
This skill should be used when searching, fetching, or downloading emails from Gmail. Use for queries like "search Gmail for...", "find emails from John", "show unread emails", "emails about project X", or "download attachment from email".
Search and download images via Google Custom Search API with LLM-powered selection. This skill should be used when finding images for articles, presentations, research documents, or enriching Obsidian notes with relevant visuals. Supports simple queries, batch processing from JSON config, automatic config generation from terms, and full note enrichment with automatic image insertion below headings.
Generate and edit images using OpenAI's GPT Image 2 API. Interactive skill that guides users through image creation with style presets, cost-aware draft/final workflow, thinking mode, carousels, and photo editing. This skill should be used when the user requests image generation via OpenAI/GPT Image 2, wants to create social media carousels, edit photos into artistic styles, or needs images with readable text (infographics, diagrams, posters).
Publish files or Obsidian notes as GitHub Gists. Use when user wants to share code/notes publicly, create quick shareable snippets, or publish markdown to GitHub. Triggers include "publish as gist", "create gist", "share on github", "make a gist from this".
This skill should be used when importing, listing, or exporting Granola meeting recordings and transcripts. Queries Granola's Personal API to list meetings, extract transcripts, and export to Obsidian notes in Fathom-compatible format.
This skill should be used when interacting with Google Workspace services via the gws CLI — Gmail (search, triage, send, labels, filters, drafts), Calendar (agenda, events, Meet conferencing), Drive (upload, list, share, download), Sheets (read, append), Docs, Tasks, Chat (send), People/Contacts, and cross-service workflows (standup, meeting prep, weekly digest, email-to-task). Triggers on queries like "check my email", "search Gmail", "send email", "calendar agenda", "create calendar event", "upload to Drive", "read spreadsheet", "create a task", "triage inbox", "find contact", "post to Chat".
Query Apple Health SQLite database for vitals, activity, sleep, and workouts. Supports Markdown, JSON, and FHIR R4 output formats. This skill should be used when analyzing health metrics, generating health reports, answering questions about fitness or sleep patterns, or exporting health data in standard formats.
This skill should be used when editing, translating, or reviewing an Astro-style i18n string corpus (files of the form export default { en: {...}, ru: {...} } under src/i18n/strings), or when the user wants to fill in missing translations, audit coverage, accept/review translations, propagate an edit across duplicate strings, get translation candidates, bulk-edit UI microcopy, or open a visual/keyboard translation editor. Drives the standalone i18n Studio tool at ~/ai_projects/i18n-studio (AST-safe minimal-diff saves via ts-morph, acceptance review state, duplicate propagation, hot-reload, and Claude translation suggestions).
Scaffold a new Tauri v2 project with the cenno/cull house conventions — delegates boilerplate to `npm create tauri-app`, then layers an opinionated core plus opt-in modules (CLI+MCP, SQLite, tray/updater, release/preflight, Swift sidecar). Use when the user wants to start a new Tauri desktop app, "init a tauri project", or "scaffold a tauri app".
Scaffold a standalone SwiftUI app via XcodeGen — pick macOS, iOS, or iOS+watchOS; ships house conventions, a Swift Testing target, swiftformat/swiftlint, optional CloudKit/CI/Release/Push modules, and optional JTBD product context. Use to "init an xcode project", "scaffold a swiftui app", "new mac/ios app".
Terminal-first JTBD engine for founders and product people. Interview fast, kill jargon, capture real switching forces (Push/Pull/Habit/Anxiety), score opportunities, and export structured artifacts (JSON + one-pager + messaging angles + GTM brief). Use when the user says "help me figure out what to build", "analyze these customer reviews", "what are people actually hiring this for", "I need messaging for my product", "turn this interview into insights", "what should I prioritize", or any variation of articulating what a project does, why it matters, who it's for, or converting interview/review/transcript signal into a decision-grade brief. Also triggers on "describe my project", "JTBD", "jobs to be done", "switching forces", or "mine these reviews".
Final retrospective and self-assessment for participants of Claude Code Lab. Runs four sequential interactive parts — progress audit, best prompt, monthly plan, and feedback — using AskUserQuestion. Triggers on "/lab-retro", "lab retrospective", "claude code lab final", or after completing the 6-week Claude Code Lab cohort.
Generate a dedicated Obsidian learning vault for any certification, course, or study goal. Creates structured notes with domains, concepts, lessons, scenarios, MoCs, dataview queries, action items, and multiple navigation paths. Inspired by the genome vault pattern. Use when the user wants to create a study vault, learning vault, certification prep vault, or structured knowledge base for a learning goal.
Manage Linear issues, projects, and workflows via CLI. This skill should be used when the user wants to create, list, update, or search Linear issues, manage projects or milestones, or interact with their Linear workspace. Triggers on "create a task", "add a Linear issue", "list my issues", "update GLE-123", "what's in my backlog", or any Linear-related request.
Process textual and multimedia files with various LLM providers using the llm CLI. Supports both non-interactive and interactive modes with model selection, config persistence, and file input handling.
This skill should be used when processing meeting transcripts to auto-detect meeting type (leadgen, partnership, coaching, internal) and extract type-specific structured analysis. Triggers on "process meeting", "analyze meeting", "meeting summary", or after syncing new Fathom/Granola transcripts.
Run quick, offline, private LLM tasks on local models via llama.cpp, reusing models already downloaded by Ollama. Use for cheap/bulk text work (summarize, classify, extract JSON, anonymize PII, translate, proofread, keywords), local embeddings, and offline image description — and prefer it over a cloud API whenever a task is privacy-sensitive, must run offline, is high-volume/low-stakes, or just needs a fast throwaway answer. Provides an `lm` CLI wrapper plus an OpenAI-compatible local server.
Prepare for upcoming meetings — pulls Cal.com bookings, researches participants, audits previous sessions from Obsidian vault, and creates prep notes. Also links prep notes to post-meeting session notes.
Execute Claude Code commands in the telegram_agent project directory. Use when the user wants to work on the telegram agent itself, fix bugs, add features, or modify the bot code. This is a COMMAND HANDLER, not a script executor.
Run candidate product, brand, company, or benchmark names through an audition — authoritative domain-availability checks, collision research across SaaS/GitHub/packages/the target adjacent domain, a light trademark and ownability read, a ranked callback list, and an interactive casting report of finalists with optional draft branding. Use when the user is naming a product, app, company, feature, or benchmark; asks "is this name taken", "check these domains", "help me pick a name", "is X available", "name my product", "brand name research", "audition names"; or wants to compare and pressure-test a shortlist of candidate names before committing.
Generate and edit images using Google's Gemini image generation models (Nano Banana family). Supports style presets, platform-specific sizing (YouTube/slides/blog), variants, image editing via inlineData, reference images for style transfer, and organized output with metadata. Default model is Nano Banana 2 (gemini-3.1-flash-image-preview). Key is auto-decrypted via SOPS.
This skill should be used to run a formal heuristic evaluation of a design artifact against Jakob Nielsen's 10 usability heuristics, producing an evidence-backed, severity-scored report. Use it when the user wants a "heuristic evaluation", "usability review", "Nielsen heuristics check", "UX heuristic audit", or asks whether a screenshot, live URL, HTML page, codebase UI, interface description, or JTBD/spec document holds up against usability principles. Accepts five input types (screenshot/image, live URL, codebase/HTML, interface description, JTBD/spec doc) and adapts its rigor and output honestly to what is actually observable. Can render the report as plain markdown (default) or as a Tufte-style HTML report, and can export findings above a severity threshold as Linear or Beads (bd) tasks after confirmation.
Professional PDF generation from markdown using Pandoc with Eisvogel template and EB Garamond fonts. Use when converting markdown to PDF, creating white papers, research documents, marketing materials, or technical documentation. Supports both English and Russian documents with professional typography and color-coded themes. Mobile-optimized layout (6x9) by default for Telegram bot context, desktop/print layout (A4) for other contexts.
Generate interactive HTML presentations with professional ElevenLabs voiceover narration synced to slides. Supports dual article/slides mode, scroll-reveal animations, GPT Image 2 illustrations, and configurable detail levels. Use this skill when the user wants to create a presentation, slide deck, narrated briefing, research report with voiceover, or any content that should be presentable as both a readable article and a navigable slide deck. Also triggers on "make a presentation", "create slides", "present this", "narrated deck", "voiceover slides", "briefing with audio", or requests to turn research/notes into a shareable presentation. Works with any content — research findings, meeting summaries, proposals, educational material.
Use when the user wants Claude Code, Codex, or other AI coding/business agents to work together as peers. This skill should be used whenever the user mentions coordinating Claude Code and Codex, agent handoffs, multi-agent workflows, parity, respect, pushback between agents, deciding which agent should lead, or turning a business/code workflow into a two-agent operating model.
Build a compressed, visualizable "portrait" of a consulting/coaching client before a session, so the paid hour is spent solving, not scoping. Runs a 7-lens JTBD-inspired interview (where / how / what / problem / ideal / tension / jobs-to-be-done) that takes rich open answers in and compresses them to an 11-field YAML portrait out. Delivers three ways: raw paste-into-a-clean-chat prompt, a secret GitHub gist link, or a Codex CLI one-liner. Use when preparing for an upcoming client call, when the user says "prep an intake", "portrait interview", "questions before our session", "send a client a pre-session questionnaire", or wants a reusable client-intake instrument.
This skill should be used when publishing a new or updated skill to the claude-skills-site Astro website. Use this skill for both adding new skills AND updating existing ones on the site. Triggers on "/publish-skill skillname", "add skill to site", "publish skillname to skills site", "update skill on site", "edit skill on site", "sync skill to site", "republish skill". Reads SKILL.md from the skills repo, generates MDX frontmatter and body, picks the right bundle, updates bundle skills array, and commits/pushes both repos. For existing skills, pass --update to preserve apothecary_name, hero_image, and activity data while regenerating the rest. A hero image is generated by default (pass --no-image to skip).
This skill should be used to search the local Obsidian vault / markdown knowledge base by meaning, not just keywords, using the on-device qmd engine (BM25 + vector + LLM rerank). Trigger when the user asks to "search my vault/notes", "find notes about X", "what do my notes say about Y", "do I have anything on Z", "semantic search my knowledge base", or wants concept/cross-lingual retrieval over markdown. Fully local — nothing leaves the machine.
Iterate on RAG systems with structured evals instead of eyeballing. This skill should be used when the user is tuning a RAG pipeline — changing retrieval prompts, swapping models, adjusting chunking, or debugging poor answers — and wants a cheap, ranked set of experiments with cost tracking and structured feedback on the stack. Also use when the user asks "how do I know if my RAG is working?", "this RAG eval is burning money", or "what should I try next on retrieval?".
Demo/recording mode that redacts personally identifiable and sensitive information from Claude Code's outputs. Use when the user invokes /recording or says they are about to record, screen-share, or demo their Claude Code session and want PII scrubbed in real time.
Interactively prepare a code repository for publication — LICENSE, NOTICE, AUTHORSHIP, README sections, package metadata, .gitignore, community docs (CONTRIBUTING/CODE_OF_CONDUCT/SECURITY/CHANGELOG), .github templates (issues/PR/CI/dependabot), a promo "more from the author" block, and conditional EU/Germany legal compliance (CRA, AI Act, GDPR, Impressum, product liability). Includes a dedicated authorship wizard that documents human decisions, judgment, direction, and art direction with jurisdiction-aware legal framing. Use when setting up a new repo for release, adding missing legal/meta/community files, writing an AUTHORSHIP record for AI-assisted work, or asking "prepare this repo for GitHub / open-sourcing".
Interactive post-session retrospective that captures learnings, updates skills, and saves memories. Use when the user says "/retrospective", "let's do a retro", "what did we learn", "session review", "retro", or "wrap up". Also use at the end of long productive sessions when significant patterns or corrections emerged. Supports multi-session mode — by default processes all of today's sessions across projects.
Claude Skills — questions
Answers built from the skills we actually parsed.