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.
Index and search Claude Code sessions using semantic embeddings (Gemini). Find past sessions by topic, relaunch the best match. Triggers on "find session", "which session did I", "relaunch the session where", "session about X".
Three-layer PII anonymization for session transcripts (therapy, coaching, consulting, mentoring). Runs Natasha (Russian NER), OpenAI Privacy Filter, and local LLM (Ollama) in sequence for maximum coverage. Fully local by default. This skill should be used when anonymizing session transcripts, notes, or any text containing client PII before AI analysis. Triggers on "anonymize", "redact PII", "anonymize session", "protect client data", "strip personal data", "anonymize transcript".
This skill should be used when designing, running, validating, or auditing statistical experiments on personal or observational time-series data (health metrics, speech/text corpora, behavioral logs, diaries, n-of-1 self-tracking). It enforces pre-registration, exact permutation tests, FDR discipline, data-validation gates, adversarial code review, and cross-validation with external models. Triggers on "design an experiment", "test this hypothesis on my data", "is this correlation real", "audit these findings", "pre-register", "validate this dataset", or any n-of-1 / quantified-self analysis request.
This skill should be used when searching Claude Code session transcripts with semantic understanding. Triggers on queries like "find sessions about X", "when did I work on Y", "search previous conversations". Supports natural language queries with synonym matching.
Pre-consultation diagnostic questionnaire for clients building websites with AI tools (Claude, Codex, Cursor, Bolt, v0, etc.) who have concerns about quality, design, or maintainability. Collects structured answers about their project, tools, pain points, and goals, then generates a consultation brief. Use when preparing for a website review consultation, when a client asks for a site audit, or when someone says their AI-built site has problems. Supports Russian and English — asks the client to choose language first.
Interview-driven automation design tool. This skill should be used when the user wants to design a new skill, agent, automation, shortcut, or any other automatable workflow. Runs a coverage-driven JTBD interview (text or voice), then exports a one-page markdown spec plus an SVG design map. Can also analyze Claude Code sessions (current or by ID) to extract workflows, subagent patterns, and skill usage, then propose new skills based on observed patterns. Use this skill whenever the user mentions analyzing a session, extracting workflows, proposing skills from past work, reviewing what tools or agents were used, or turning a session into a skill. Triggers on "analyze this session", "what skills could I build from this", "propose skills from session", "what workflows did I use", "what did I do in this session", "extract patterns", "turn this into a skill".
Scan and clean macOS caches, package-manager data, crash dumps, and app caches to reclaim disk space. Deterministic — a config registry (targets.json) plus two scripts (survey.py read-only, clean.py executor) do all the measuring and deleting; the agent only relays a compressed summary and makes the few human-judgment calls. IMPORTANT — use this skill whenever the user's request on macOS involves: freeing disk space, cleaning/clearing caches, "disk is full", "clean up my Mac", "free up space", "what's eating my disk", "running low on disk", needing space for an install, or any low-storage complaint. Covers the whole workflow survey → choose → clean → empty Trash.
Controls the cmux macOS terminal app (Ghostty-based, AI-agent-aware) via its CLI socket API. Use this skill to: open browser preview panes when a dev server starts or an HTML file is created; spin up named workspaces for parallel subagents; show live sidebar progress/status during long tasks; open utility splits on demand (logs, URLs, terminals) in any direction. Also handles `/cmux demo` — an interactive walkthrough with 3 screens of options that runs personalized live demos. Trigger when: a server starts and needs previewing, an HTML/URL artifact is created, parallel subagents are dispatched, a long task needs visible progress, the user asks to open/show something in a split or tab, or the user invokes `/cmux demo`.
This skill should be used to watch a long-running background job (ffmpeg/media encode, qmd or other embedding/vector-DB run, batch agent/LLM pipeline, or a real-browser/agent-browser daemon) until it finishes or wedges, then deliver a verdict (done, needs-attention, or blocked) plus the exact next command, without burning dozens of manual poll commands. Triggers on "babysit this job", "watch this until it's done", "ping me when the encode/embed/batch finishes", "is this background process stuck", "monitor this ffmpeg/qmd run", or any request to wait on a long-running process and be told when it's complete or hung.
Sync and manage bilingual (EN/RU) library content for agency-docs. Use when adding, updating, or reviewing library articles. Handles translation, sync checks, and Russian stylistic review.
>- Cut a software release and maintain a tiered compatibility policy. Use when the user wants to release, ship a version, bump the version, tag a release, write a changelog, or update COMPATIBILITY. Config-driven via release.config.json; bumps version files, runs a readiness gate, updates COMPATIBILITY.md tiers and deprecations, tags (→ release workflow), and reports closed issues. Teaches the underlying standards as it runs.
Automate getSorted.de (Sorted) for freelancer invoicing, expense tracking, and German tax submissions (VAT, ZM, annual returns). Uses real-browser automation via Chrome Beta + agent-browser. Triggers on "create invoice", "sorted invoice", "download invoice", "submit VAT", "tax report", "sorted expenses", "/sorted", or any getSorted.de interaction.
This skill should be used to generate realistic, persona-consistent synthetic coaching and therapy session transcripts for evals, demos, and training data. It produces fictional but believable coach/client (or therapist/client) dialogue grounded in a chosen modality (ICF/GROW coaching, CBT, IFS parts work, ACT/motivational interviewing) and exports to Fathom/Granola transcript style, plain dialogue, structured JSON, or Obsidian markdown. Triggers on requests like "generate a synthetic coaching session", "make fake therapy transcripts for evals", "create demo session transcripts", "synthetic CBT dialogue", "persona-consistent coaching transcript", "test data for my session summarizer", or "mock coaching call".
This skill should be used when the user wants to implement features or fix bugs using test-driven development. Enforces the RED-GREEN-REFACTOR cycle with vertical slicing, context isolation between test writing and implementation, human checkpoints, and auto-test feedback loops. Uses multi-agent orchestration with the Task tool for architecturally enforced context isolation. Supports Jest, Vitest, pytest, Go test, cargo test, PHPUnit, and RSpec.
Generate a 3D interactive knowledge map (Inner Temple) from any Obsidian vault or document set. Supports multi-scale abstraction layers and dual-graph common maps between two vaults.
This skill should be used for comprehensive Telegram automation via Telethon API. Use for sending/receiving messages, monitoring chats, running a background daemon that triggers Claude Code sessions, managing channels/groups, and downloading media. Triggers on "telegram daemon", "monitor telegram", "telegram bot", "spawn Claude from telegram", or any Telethon-related request. IMPORTANT: Use `draft` command for "драфт/draft", use `send` for "отправь/send"; if ambiguous, ASK before sending.
Review and send Telegram response drafts, manage follow-ups for unanswered outbound messages. Use when the user says "/tg-responder review", "/tg-responder status", "/tg-responder follow-ups", "check telegram drafts", "review pending messages", "telegram inbox", "who hasn't replied", or "follow up".
This skill should be used when fetching, searching, downloading, sending, editing, or publishing messages on Telegram. Use for queries like "show my Telegram messages", "search Telegram for...", "get unread messages", "send a message to...", "edit that message", "publish this draft to klodkot", or "add Telegram messages to my notes".
A Theory-of-Constraints diagnostic for deciding what to automate with AI agents. Before building any automation, skill, Goal, loop, or schedule, it walks Goldratt's Five Focusing Steps over the user's work system to find the real bottleneck, then recommends the single highest-leverage automation aimed at the constraint plus a what-NOT-to-automate list. Use when the user asks "what should I automate", "where do I point my agents", "prioritize my automation backlog", "which workflow should I agentify", "is this worth building", "find my bottleneck", "what's the highest-leverage thing", when they are about to build a Claude Code skill/Goal/loop/schedule and aren't sure it matters, or during a review of their automations. Guards against the common failure of automating busywork (a local optimum) instead of the constraint.
Control timeBuzzer hardware LED via MIDI — set color, effects (pulse, strobe, rainbow, fade), and semantic status signals. Use when the user asks to change the buzzer LED color, signal status through the buzzer, or sync the buzzer with other lighting.
This skill analyzes meeting transcripts to extract decisions, action items, opinions, questions, and terminology using Cerebras AI (llama-3.3-70b). Use this skill when the user asks to analyze a transcript, extract action items from meetings, find decisions in conversations, build glossaries from discussions, or summarize key points from recorded meetings.
Create Tufte-inspired data reports and infographic dashboards as standalone HTML files. Uses EB Garamond for text, Monaspace Argon for numbers, Chart.js for interactive charts, and inline SVG sparklines. Produces publication-quality reports with 2-column narrative+data layouts, status dashboards, scroll animations, and responsive mobile support. Use this skill whenever the user wants to create a data report, activity dashboard, infographic, personal analytics page, health tracker visualization, or any document that combines narrative text with interactive charts and tables. Also triggers for "make a report like Tufte", "create an infographic", "build a dashboard", "visualize my data", or requests for beautiful data-driven documents.
This skill should be used when applying proper typography to prose text or files in Russian, English, German, or French — smart quotes per locale («ёлочки», “curly”, „Gänsefüßchen“, « guillemets »), correct dashes (тире, em/en dash, Gedankenstrich, tiret), non-breaking spaces, ranges, ellipsis, and French espaces insécables before ! ? ; :. Fully deterministic via a pinned typograf-based CLI; never apply these rules by hand. Triggers on "типографика", "typograf", "оттипографь", "smart quotes", "fix typography", "неразрывные пробелы".
Score and compare images using vision LLMs as judges. YAML-defined criteria presets for 11 use cases (text-to-image, photorealism, document OCR, charts, UI, portrait, product, scientific, invoice, alt-text, artistic style). Supports OpenAI, Anthropic, Gemini, Mistral, and OpenRouter as judge providers. Keys auto-decrypted via SOPS + age.
Research, score, and publish a weekly industry digest on any topic. Casts a wide net via web search (20+ candidates), verifies sources for AI-generated slop, scores each item on five parameters (Novelty, Relevance, Slopiness, Technical depth, Feasibility), selects the top items, and generates both markdown files and a Tufte-style HTML report. Use this skill whenever the user says "/weekly-digest", "run the weekly digest", "industry digest", "what's new in [topic]", "weekly roundup", "news digest on [topic]", "research what happened this week in [topic]", or any request to create a curated, scored summary of recent developments in a field. Also use when the user wants to monitor a topic area with quality filtering — the slopiness scoring is the key differentiator from a simple news search.
Audit a white paper or long-form technical document against a research-grounded best-practices checklist. Two lanes — deterministic script checks (readability, undefined acronyms, structure blocks, broken links) plus an LLM-judge review (overclaims, inconsistent numbers, buried lede, limitations honesty, audience fit). Produces a prioritized P0–P2 findings report by default; applies fixes on explicit request (TDD for any code changes). Use when the user says "audit this white paper", "review my paper against best practices", "check this doc for overclaims", "whitepaper QA", "is this paper ready to publish", or wants prioritized recommendations on a technical document.
This skill should be used when analyzing Wispr Flow voice dictation history for self-reflection, work patterns, mental health insights, or productivity analytics AND when managing the Wispr Flow dictionary (adding terms, fixing mishears, exporting/importing, suggesting improvements). Triggered by requests like "/wispr-analytics", "analyze my dictations", "what did I dictate today", "wispr reflection", "add to wispr dictionary", "improve dictation", "wispr suggest", "export wispr dictionary", or any request to review voice dictation patterns or manage dictation quality.
Queue and batch-apply Wispr Flow dictation corrections. Use when the user invokes /wispr-fix or writes "wispr fix: X -> Y" to correct a speech-to-text mishear.
Daily digest of 3-7 genuinely surprising items from newsletters and Telegram channels. Scores content for epistemic friction, not just relevance. Appends to daily note. Use when the user says "/wow-digest", "run the wow digest", "what's surprising today", "morning reading", or "digest my newsletters".
Extract YouTube video transcripts with metadata and save as Markdown to Obsidian vault. Use this skill when the user requests downloading YouTube transcripts, converting YouTube videos to text, or extracting video subtitles. Does not download video/audio files, only metadata and subtitles.
Create and manage Zoom meetings and access cloud recordings via the Zoom API. Use for queries like "create a Zoom meeting", "list my Zoom meetings", "show my Zoom recordings", or "schedule a meeting for tomorrow".
Make AI-generated writing read as human-written, in fiction and in professional prose. Repairs the narrative architecture of fiction and stories (based on StoryScope, arXiv:2604.03136); routes professional text through domain rules for release notes, announcements, PR and issue replies, code-review comments, incident postmortems, tickets, work orders, technical articles, and blog posts. Four operations - write, review (diagnose AI tells without editing), refactor (minimal in-place edits), recreate (full rewrite). Use when asked to humanize, de-AI, unslop, or strip AI flavor from any text; when writing or revising any of these document types; or whenever output must not read as machine-written.
Audit websites for SEO, technical, content, security, JS rendering, and AI readiness using SEOmator CLI. Returns LLM-optimized reports with health scores across 251 rules and 20 categories. Use when analyzing websites, debugging SEO issues, or checking site health.
Audit websites for SEO, technical, content, security, JS rendering, and AI readiness using SEOmator CLI. Returns LLM-optimized reports with health scores across 251 rules and 20 categories. Use when analyzing websites, debugging SEO issues, or checking site health.
把复杂剧情/故事大纲拆分为 Seedance 2.0 的"镜头1/镜头2/镜头3"多分镜结构。每个分镜按 4 维度组织（运镜 + 主体动作与表情 + 位置/空间变化 + 音频）。用于"帮我把这个故事写成 Seedance 提示词"、"分镜脚本"、"多镜头视频"、"剧情复杂的视频"、"剧本转分镜"等触发场景。
生成符合 Kling 3.0（可灵 3.0，快手）规则的视频提示词。三种写法自适应：4 部分基础公式（短视频）/ 5 层进阶公式（剧情+音频）/ 图生视频专用（只描述运动）。Kling 是 2026 年中文理解最强、原生音画同步、最长 2 分钟、支持角色定向发声、Motion Brush 的电影级模型。用于"用 Kling 生成视频"、"可灵 AI 提示词"、"中文视频生成"、"带原生音频的剧情视频"、"图生视频"、"角色对话视频"等触发场景。
生成符合 HappyHorse 1.0 严格规则的紧凑提示词（30-55 词），主体先行 + 明确镜头技术 + 音频激活路径（with X audible / speaking English at natural pace）。可选加入"8s 时序节拍"结构。用于"用 HappyHorse 生成视频"、"做个 3-15 秒短片"、"要原生带音频的视频"、"ASMR 视频提示词"等触发场景。
把用户的自然语言视频需求转换为符合 Doubao Seedance 2.0 进阶公式的提示词（8 要素：精准主体+动作细节+场景环境+光影色调+镜头运镜+视觉风格+画质+约束条件）。用于"帮我写一个 Seedance 提示词"、"生成视频提示词"、"做个产品广告视频"、"用 Seedance 生成 XX"等触发场景。如果用户没明确说 Seedance，但描述的是单一镜头的复杂叙事/多主体/电影感视频，也优先用此 skill。
诊断 Seedance 2.0 生成视频时出现的常见问题（人物 ID 漂移/双胞胎/字幕/Logo/风格漂移/延长跳变/画质劣化/特效不对/中文发音/音色不准/结尾噪音 等 12 类），定位根因并给出修复后的提示词。用于"我的提示词生成出来不对"、"视频里出现奇怪的字幕"、"人脸不像参考图"、"出现两个一样的人物"、"风格变了"、"怎么修这个提示词"等触发场景。
把一条 AI 视频提示词从源模型(如 Sora 2)的写法风格转换为目标模型(如 Kling 3.0 / Wan 2.7 / Veo 3.1 等)的最佳实践写法。基于 110 条 10 场景 × 11 模型对照基准数据(prompts/data/cross-model-matrix.json),不是凭直觉重写,而是查表式 in-context learning。10 场景:产品 / 双人对话 / 物理动作 / 图生视频 / 多人会议 / 恐怖 / 自然延时 / 抽象 / 武侠 / 萌宠。用于"Sora 已 EOL 帮我把这条提示词改成 Veo"、"我有 Kling 提示词想跑 Wan"、"跨模型 A/B 测试"、"把英文 prompt 优化成 Kling 中文版"等触发场景。
根据用户的视频需求(场景、时长、音频、语言、平台限制、预算、是否需要本地部署/角色一致性等),从 16 个主流 AI 视频模型(12 商业 + 4 开源)中推荐最匹配的 1-3 个,并解释为什么。覆盖商业:Seedance 2.0、HappyHorse 1.0、Kling 3.0、Sora 2、Veo 3.1、**Gemini Omni**(2026-05 新)、Runway Gen-4.5/Aleph、Pika 2.5、Hailuo 02、Hunyuan Video 1.5、Wan 2.7、即梦 AI;开源:LTX-Video 0.9.7、Mochi 1、CogVideoX 5B、Higgsfield Soul。用于"用哪个 AI 视频模型好"、"Sora 还是 Kling"、"国内有什么模型"、"哪个免费"、"哪个支持中文"、"哪个能编辑已有视频"、"哪个能本地部署"、"哪个角色一致性最强"等触发场景。
Call a peer agent's A2A JSON-RPC endpoint. Supports tasks/send (send a task message) and tasks/get (query task status).
Fetch an A2A agent card JSON from a given endpoint URL. Tries direct GET first, then falls back to /.well-known/agent.json.
Generate an image from a text prompt using the default OpenAI-like LLM config (gpt-image-1-mini or compatible image generation API). The image is uploaded to the gofile.io public file sharing service and ONLY the public download page URL is returned.
Use when the user wants to manage a manual execution run or update case-level results inside a run — listing runs, creating runs for a release, inspecting a run, assigning cases, or marking case results (passed/failed/blocked/skipped/retest/untested). Accepts counter-style IDs like RUN-12 and TC-156.
Use when the user wants to check TestDino connection status, validate their PAT, discover available organizations and projects, or find the right projectId. Always call this first when the project context is ambiguous before any other TestDino tool.
Use the `quickdesign` CLI to generate AI media — UGC promo videos, image edits, product creatives, video upscales — through Seedance, Kling, Sora2, Nano Banana, and GPT Image. Invoke this skill whenever the user asks for a talking-avatar video, multi-segment ad / promo / explainer, image edit (object swap, angle change, state change), product photoshoot, or video upscale via QuickDesign.
Use when the user wants to inspect automated test runs, list failed or flaky tests, debug a failing testcase with historical context, or filter runs by branch, commit, author, environment, browser, status, or tags. Includes list_testruns, get_run_details, list_testcase, get_testcase_details, and debug_testcase.
Use when the user wants to browse, inspect, create, or update exploratory testing sessions in TestDino — by state, status, session type, assignee, release, or tags. Covers list_sessions, get_session, create_session, and update_session. Accepts counter-style IDs like SES-12.
Use only when the user explicitly asks for a TestDino audit of Playwright automated test code. Routes through the audit tools the TestDino MCP server exposes (get_audit_report + submit_audit_report, or the legacy test_audit). For generic code review or non-Playwright targets, do a normal review instead.
Use when the user wants to create, update, or browse manual QA test cases and suites in TestDino — not execution runs. Covers list_manual_test_suites, list_manual_test_cases, get_manual_test_case, create_manual_test_case, update_manual_test_case, create_manual_test_suite.
Use when the user wants to browse, inspect, create, or update releases/milestones in a TestDino project. Covers list_releases, get_release, create_release, and update_release. Accepts counter-style IDs like MS-12.
> Validate wallet balances in TRES Finance against DeBank and generate a discrepancy report. Use this skill whenever the user asks to validate, verify, cross-check, or audit their TRES balances against on-chain data or DeBank. Also trigger when the user asks if their balances are correct or wants to see discrepancies. Only EVM-compatible wallets are supported.
> Add a blockchain explorer transaction to the TRES Finance ledger. Use when the user provides a blockchain explorer URL (Etherscan, Polygonscan, Arbiscan, Basescan, Snowtrace, BscScan, etc.) or a raw transaction hash and wants to record it in the platform ledger. Trigger phrases include "add this transaction", "import tx from etherscan", "record this on-chain tx", "add to ledger".
> (FIFO, LIFO, AVG, MAX_GAINS, MAX_LOSSES, FIFO_IMPAIRMENT), trigger recalculation, review financial issues, manage reevaluations/impairments, set bulk fiat prices, configure spec-ID lot disposal rules, view per-asset cost basis results, and export cost basis reports. Trigger this skill whenever the user asks about cost basis, gains/losses, realized gains, unrealized gains, COGS, inventory lots, tax lots, cost basis strategy, reevaluation, impairment, fiat value override, spec-ID rules, financial issues (negative balance, missing fiat), or anything related to how their crypto assets are costed. Also trigger when the user says "calculate cost basis", "what is my cost basis strategy", "show me financial issues", "create a reevaluation", "run cost basis", "change strategy to LIFO", or similar.
Claude Skills — questions
Answers built from the skills we actually parsed.