glebis/claude-skills

这个仓库里有 112 个技能，GitHub 星标 ★ 378。

• agency-docs-updaterEnd-to-end pipeline for publishing Claude Code lab meetings. Accepts optional ar
• agency-meetup-publishEnd-to-end pipeline for publishing AGENCY Community meetup recordings to YouTube
• agency-socialsGenerate social media covers and assets for AGENCY Community events, meetups, an
• agent-cliAdd agent-friendly --json NDJSON output to Python CLI scripts, or scaffold a com
• app-releaseEnd-to-end pipeline for releasing an iOS / watchOS app to TestFlight and the App
• automation-advisorInteractive automation decision advisor using the Automation Decision Matrix fra
• balancedConstructive, evidence-based dialogue mode that avoids sycophancy. This skill sh
• brand-agencyApplies Agency brand colors and typography to artifacts including presentations,
• browser-mateUse when automating a real, logged-in Chrome WITHOUT disturbing the user's open
• browsing-historyQuery browsing history from all synced devices (iPhone, Mac, iPad, desktop). Sup
• chrome-historyQuery Chrome browsing history with natural language. Filter by date range, artic
• coaching-session-summarizerThis skill should be used to summarize coaching or therapy session transcripts a
• codexUse when the user asks to run Codex CLI (codex exec, codex resume) or references
• cognitive-toolkitEvidence-based CBT and DBT intervention skills — guided thought records, opposit
• annotateBuild and verify a PII gold set with HUMAN annotators (first-class). Launch the
• anonDe-identify a session transcript (file or folder) by redacting PII LOCALLY befor
• auditRun a corpus-scale, STATS-ONLY PII audit over a folder of session transcripts LO
• red>-
• rehydratePut the real values back into an analysis that was produced from GREEN (placehol
• setupSet up, install, and configure CONFIDE local de-identification — installs Python
• vault>-
• viewBuild a self-contained interactive HTML that lets you SEE de-identification and
• context-builderGenerate interactive AI transformation context-builder prompts for consulting cl
• cullThis skill should be used when the user wants to view, review, rate, organize, s
• cull-releaseUse when orchestrating or resuming Cull's complete release cycle, reporting its
• cull-release-checkUse when assessing whether Cull is ready to release, auditing or dry-running a C
• cull-release-prepareUse when the user explicitly asks to prepare or bump a Cull patch, minor, or maj
• cull-release-publishUse when the user explicitly asks to publish or complete a prepared Cull release
• cull-release-recoverUse when a Cull release is stuck, inconsistent, failed in a workflow or artifact
• cull-release-verifyUse when verifying or auditing a Cull release, DMG, updater archive, notarizatio
• daydream
• decision-toolkitGenerate structured decision-making tools — step-by-step guides, bias checkers,
• deep-researchThis skill should be used when conducting comprehensive research on any topic us
• design-tokensMOVED — design-tokens now ships in the humane plugin (glebis/humane-agentic-desi
• doctorgEvidence-based health research using tiered trusted sources with GRADE-inspired
• ecosystemAudit the Claude Code ecosystem — skill health and staleness, project activity p
• elevenlabs-ttsThis skill converts text to high-quality audio files using ElevenLabs API. Use t
• elimination-researchThis skill should be used for elimination-style research where the user wants to
• fathomFetch meetings, transcripts, summaries, and action items from Fathom API. Use wh
• feature-factoryThis skill should be used when taking a single software feature from intent to s
• firecrawl-researchThis skill should be used when the user requests to research topics using FireCr
• font-featuresThis skill should be used when inspecting or applying advanced OpenType features
• github-gistPublish files or Obsidian notes as GitHub Gists. Use when user wants to share co
• gmailThis skill should be used when searching, fetching, or downloading emails from G
• google-image-searchSearch and download images via Google Custom Search API with LLM-powered selecti
• gpt-image-2Generate and edit images using OpenAI's GPT Image 2 API. Interactive skill that
• granolaThis skill should be used when importing, listing, or exporting Granola meeting
• gwsThis skill should be used when interacting with Google Workspace services via th
• health-dataQuery Apple Health SQLite database for vitals, activity, sleep, and workouts. Su
• i18n-studioThis skill should be used when editing, translating, or reviewing an Astro-style
• init-tauri-appScaffold a new Tauri v2 project with the cenno/cull house conventions — delegate
• init-xcode-appScaffold a standalone SwiftUI app via XcodeGen — pick macOS, iOS, or iOS+watchOS
• insight-extractor
• jtbdMOVED — jtbd now ships in the humane plugin (glebis/humane-agentic-design), not
• lab-retroFinal retrospective and self-assessment for participants of Claude Code Lab. Run
• learning-vaultGenerate a dedicated Obsidian learning vault for any certification, course, or s
• linearManage Linear issues, projects, and workflows via CLI. This skill should be used
• llm-cliProcess textual and multimedia files with various LLM providers using the llm CL
• local-modelsRun quick, offline, private LLM tasks on local models via llama.cpp, reusing mod
• meeting-prepPrepare for upcoming meetings — pulls Cal.com bookings, researches participants,
• meeting-processorThis skill should be used when processing meeting transcripts to auto-detect mee
• metaExecute Claude Code commands in the telegram_agent project directory. Use when t
• name-auditionRun candidate product, brand, company, or benchmark names through an audition —
• nano-bananaGenerate and edit images using Google's Gemini image generation models (Nano Ban
• nielsen-heuristicsMOVED — nielsen-heuristics now ships in the humane plugin (glebis/humane-agentic
• pdf-generationProfessional PDF generation from markdown using Pandoc with Eisvogel template an
• peer-agent-collaborationUse when the user wants Claude Code, Codex, or other AI coding/business agents t
• pre-session-portraitBuild a compressed, visualizable "portrait" of a consulting/coaching client befo
• presentGenerate interactive HTML presentations with professional ElevenLabs voiceover n
• presentation-generatorGenerate interactive HTML presentations with neobrutalism styling, ASCII art dec
• publish-skillThis skill should be used when publishing a new or updated skill to the claude-s
• qmd-searchThis skill should be used to search the local Obsidian vault / markdown knowledg
• rag-evalIterate on RAG systems with structured evals instead of eyeballing. This skill s
• recordingDemo/recording mode that redacts personally identifiable and sensitive informati
• repo-prepInteractively prepare a code repository for publication — LICENSE, NOTICE, AUTHO
• repo-publishPublish a local repository to a remote with quality gates, version tagging, and
• retrospectiveInteractive post-session retrospective that captures learnings, updates skills,
• rigorous-experimentsThis skill should be used when designing, running, validating, or auditing stati
• session-anonymizerThree-layer PII anonymization for session transcripts (therapy, coaching, consul
• session-finderIndex and search Claude Code sessions using semantic embeddings (Gemini). Find p
• session-searchThis skill should be used when searching Claude Code session transcripts with se
• site-diagnosisPre-consultation diagnostic questionnaire for clients building websites with AI
• sketchOpen a Fabric.js-based SVG editor in the browser for collaborative visual protot
• skill-studioInterview-driven automation design tool. This skill should be used when the user
• cmuxControls the cmux macOS terminal app (Ghostty-based, AI-agent-aware) via its CLI
• disk-cleanupScan and clean macOS caches, package-manager data, crash dumps, and app caches t
• job-babysitterThis skill should be used to watch a long-running background job (ffmpeg/media e
• library-syncSync and manage bilingual (EN/RU) library content for agency-docs. Use when addi
• release>-
• sortedAutomate getSorted.de (Sorted) for freelancer invoicing, expense tracking, and G
• synthetic-session-generatorThis skill should be used to generate realistic, persona-consistent synthetic co
• tddThis skill should be used when the user wants to implement features or fix bugs
• telegramThis skill should be used when fetching, searching, downloading, sending, editin
• telegram-postCreate, preview, and send formatted Telegram posts from draft markdown files. Us
• telegram-telethonThis skill should be used for comprehensive Telegram automation via Telethon API
• temple-generatorGenerate a 3D interactive knowledge map (Inner Temple) from any Obsidian vault o
• tg-responderReview and send Telegram response drafts, manage follow-ups for unanswered outbo
• the-goalA Theory-of-Constraints diagnostic for deciding what to automate with AI agents.
• thinking-patterns
• timebuzzer-ledControl timeBuzzer hardware LED via MIDI — set color, effects (pulse, strobe, ra
• trail-checkinInteractive trail review and update process for Obsidian vault at ~/Brains/brain
• transcript-analyzerThis skill analyzes meeting transcripts to extract decisions, action items, opin
• tufte-reportCreate Tufte-inspired data reports and infographic dashboards as standalone HTML
• typographyThis skill should be used when applying proper typography to prose text or files
• vision-benchScore and compare images using vision LLMs as judges. YAML-defined criteria pres
• weekly-digestResearch, score, and publish a weekly industry digest on any topic. Casts a wide
• whitepaper-auditAudit a white paper or long-form technical document against a research-grounded
• wispr-analyticsThis skill should be used when analyzing Wispr Flow voice dictation history for
• wispr-fixQueue and batch-apply Wispr Flow dictation corrections. Use when the user invoke
• wow-digestDaily digest of 3-7 genuinely surprising items from newsletters and Telegram cha
• youtube-transcriptExtract YouTube video transcripts with metadata and save as Markdown to Obsidian
• zoomCreate and manage Zoom meetings and access cloud recordings via the Zoom API. Us