Skill 列表

已對齊 live registry。今日同步：2026-06-20 06:02:26 CST；目前 catalog 共 195 個 skill、18 個分類。

未分類

• dogfood：Exploratory QA of web apps: find bugs, evidence, reports.
• yuanbao：Yuanbao (元宝) groups: @mention users, query info/members.

autonomous-ai-agents

• claude-code：Delegate coding to Claude Code CLI (features, PRs).
• codex：Delegate coding to OpenAI Codex CLI (features, PRs).
• hermes-agent：Configure, extend, or contribute to Hermes Agent.
• kanban-codex-lane：Use when a Hermes Kanban worker wants to run Codex CLI as an isolated implementation lane while Hermes keeps ownership of task lifecycle, reconciliation, testing, and handoff.
• opencode：Delegate coding to OpenCode CLI (features, PR review).

creative

• animation-script-writing：Corporate/explainer animation script planning and delivery, with audience-first pacing and explicit split between voiceover and on-screen text.
• architecture-diagram：Dark-themed SVG architecture/cloud/infra diagrams as HTML.
• ascii-art：ASCII art: pyfiglet, cowsay, boxes, image-to-ascii.
• ascii-video：ASCII video: convert video/audio to colored ASCII MP4/GIF.
• baoyu-article-illustrator：Article illustrations: type × style × palette consistency.
• baoyu-comic：Knowledge comics (知识漫画): educational, biography, tutorial.
• baoyu-infographic：Infographics: 21 layouts x 21 styles (信息图, 可视化).
• chatgpt-image-generation-browser：Use the ChatGPT web app in the browser to generate social-media-style images, including how to enter the image-generation mode, feed a precise prompt, and verify the generated image in the DOM.
• claude-design：Design one-off HTML artifacts (landing, deck, prototype).
• comfyui：Generate images, video, and audio with ComfyUI — install, launch, manage nodes/models, run workflows with parameter injection. Uses the official comfy-cli for lifecycle and direct REST/WebSocket API for execution.
• comfyui-flux2-klein-image-generation：Use local ComfyUI with FLUX.2 Klein 4B on the user's Windows-side D:\ComfyUI install to generate still images reliably from WSL/Hermes, with customizable prompt, size, seed, and output prefix.
• design-md：Author/validate/export Google's DESIGN.md token spec files.
• excalidraw：Hand-drawn Excalidraw JSON diagrams (arch, flow, seq).
• humanizer：Humanize text: strip AI-isms and add real voice.
• ideation：Generate project ideas via creative constraints.
• manim-video：Manim CE animations: 3Blue1Brown math/algo videos.
• p5js：p5.js sketches: gen art, shaders, interactive, 3D.
• pixel-art：Pixel art w/ era palettes (NES, Game Boy, PICO-8).
• popular-web-designs：54 real design systems (Stripe, Linear, Vercel) as HTML/CSS.
• pretext：Use when building creative browser demos with @chenglou/pretext — DOM-free text layout for ASCII art, typographic flow around obstacles, text-as-geometry games, kinetic typography, and text-powered generative art. Produces single-file HTML demos by default.
• sketch：Throwaway HTML mockups: 2-3 design variants to compare.
• songwriting-and-ai-music：Songwriting craft and Suno AI music prompts.
• touchdesigner-mcp：Control a running TouchDesigner instance via twozero MCP — create operators, set parameters, wire connections, execute Python, build real-time visuals. 36 native tools.
• wsl-pil-slide-mockup-prototyping：Rapidly prototype presentation-slide mockups on WSL with Python PIL, then verify the rendered layout and Chinese text via vision analysis.
• wsl-slide-mockup-pil：Create one-page slide mockups directly in WSL with Pillow when the user wants a fast visual draft or when ComfyUI is unavailable. Verify Chinese rendering with vision_analyze and iterate for Slack-friendly readability.
• exhibition-visual-storytelling：Concept, copy, and layout workflow for educational exhibition wall sections that need metaphor, narrative clarity, and interactive reveal patterns.

data-science

• jupyter-live-kernel：Iterative Python via live Jupyter kernel (hamelnb).

devops

• cron-job-debugging：Diagnose Hermes cron job failures by checking job metadata, forcing a rerun, and inspecting request dumps and persistent cron state for root-cause signals such as empty responses, non-retryable client errors, or misconfiguration.
• gov-tender-monitor-pcc-sweep：Reusable workflow for monitoring Taiwan government tenders from PCC, deduping against Notion, filtering by current workday/time, and writing back concise summaries.
• hermes-gateway-stale-pid-recovery：Diagnose and recover Hermes Gateway when Slack/Telegram messages stop arriving because the systemd gateway service fails with a stale PID or lock file.
• kanban-orchestrator：Decomposition playbook + anti-temptation rules for an orchestrator profile routing work through Kanban. The "don't do the work yourself" rule and the basic lifecycle are auto-injected into every kanban worker's system prompt; this skill is the deeper playbook when you're specifically playing the orchestrator role.
• kanban-worker：Pitfalls, examples, and edge cases for Hermes Kanban workers. The lifecycle itself is auto-injected into every worker's system prompt as KANBAN_GUIDANCE (from agent/prompt_builder.py); this skill is what you load when you want deeper detail on specific scenarios.
• webhook-subscriptions：Webhook subscriptions: event-driven agent runs.
• wsl-windows-chrome-bridge-recovery：Recover and operate WSL-to-Windows Chrome remote-debugging sessions reliably for browser automation jobs.

email

• himalaya：CLI to manage emails via IMAP/SMTP. Use himalaya to list, read, write, reply, forward, search, and organize emails from the terminal. Supports multiple accounts and message composition with MML (MIME Meta Language).

gaming

• minecraft-modpack-server：Host modded Minecraft servers (CurseForge, Modrinth).
• pokemon-player：Play Pokemon via headless emulator + RAM reads.

github

• codebase-inspection：Inspect codebases w/ pygount: LOC, languages, ratios.
• github-auth：GitHub auth setup: HTTPS tokens, SSH keys, gh CLI login.
• github-code-review：Review PRs: diffs, inline comments via gh or REST.
• github-issues：Create, triage, label, assign GitHub issues via gh or REST.
• github-pr-workflow：GitHub PR lifecycle: branch, commit, open, CI, merge.
• github-repo-management：Clone/create/fork repos; manage remotes, releases.

mcp

• native-mcp：MCP client: connect servers, register tools (stdio/HTTP).

media

• comfyui-flux2-windows-desktop：Run ComfyUI Desktop installed on Windows from WSL, diagnose FLUX.2 model failures, and generate images via the ComfyUI HTTP API with practical fallbacks.
• gif-search：Search/download GIFs from Tenor via curl + jq.
• heartmula：HeartMuLa: Suno-like song generation from lyrics + tags.
• songsee：Audio spectrograms/features (mel, chroma, MFCC) via CLI.
• spotify：Spotify: play, search, queue, manage playlists and devices.
• vimeo-frame-extraction-analysis：Download Vimeo videos when the page is blocked or human verification appears, then extract timed frames with Python/PyAV and analyze them with vision. Optionally write a concise record back to Notion.
• vimeo-frame-sampling-analysis：Analyze Vimeo videos that are blocked by page verification by downloading accessible formats and sampling frames at a fixed interval.
• vimeo-video-analysis-fallback：Handle Vimeo links when the page is blocked by human verification by using Vimeo API/oEmbed when available, otherwise fall back to yt-dlp + PyAV frame extraction + vision analysis.
• youtube-content：Fetch YouTube video transcripts and transform them into structured content (chapters, summaries, threads, blog posts). Use when the user shares a YouTube URL or video link, asks to summarize a video, requests a transcript, or wants to extract and reformat content from any YouTube video.

mlops

• audiocraft-audio-generation：AudioCraft: MusicGen text-to-music, AudioGen text-to-sound.
• axolotl：Expert guidance for fine-tuning LLMs with Axolotl - YAML configs, 100+ models, LoRA/QLoRA, DPO/KTO/ORPO/GRPO, multimodal support
• dspy：DSPy: declarative LM programs, auto-optimize prompts, RAG.
• evaluating-llms-harness：lm-eval-harness: benchmark LLMs (MMLU, GSM8K, etc.).
• fine-tuning-with-trl：Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.
• huggingface-hub：HuggingFace hf CLI: search/download/upload models, datasets.
• llama-cpp：llama.cpp local GGUF inference + HF Hub model discovery.
• obliteratus：OBLITERATUS: abliterate LLM refusals (diff-in-means).
• outlines：Guarantee valid JSON/XML/code structure during generation, use Pydantic models for type-safe outputs, support local models (Transformers, vLLM), and maximize inference speed with Outlines - dottxt.ai's structured generation library
• segment-anything-model：SAM: zero-shot image segmentation via points, boxes, masks.
• serving-llms-vllm：vLLM: high-throughput LLM serving, OpenAI API, quantization.
• unsloth：Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization
• weights-and-biases：W&B: log ML experiments, sweeps, model registry, dashboards.

note-taking

• obsidian：Read, search, create, and edit notes in the Obsidian vault.
• woolito-obsidian-durable-writeback：Write durable Woolito facts into Obsidian with compact edits, index/log sync, and memory-capacity fallback handling.
• woolito-slack-channel-obisidian-writeback：Write durable Slack channel scope and decisions into the Woolito Obsidian vault, keeping per-channel briefs, indexes, and routing rules in sync.
• woolito-slack-mention-resolution：Resolve Woolito Slack mentions safely and accurately by checking the Obsidian staff and Slack member directories before composing messages or cron reminders.
• woolito-vault-document-ingest：Ingest a user-provided planning/requirements document into the Woolito Obsidian Vault with raw-source preservation, source summary, durable brief, and ops index/log updates.

productivity

• airtable：Airtable REST API via curl. Records CRUD, filters, upserts.
• client-facing-narrative-strategy：Turn rough ideas, source materials, and company facts into clear client-facing proposals, decks, and public brand/site copy with a single narrative axis, strong hierarchy, and reusable copy blocks.
• crm-reminder-to-g3u9wm13r：Post the fixed weekly CRM reminder message into Slack channel G3U9WM13R, not the originating thread.
• gmail-attachment-send-flow：Send email with file attachments through the authenticated Google Workspace/Gmail setup in Hermes.
• gmail-sender-display-name-debugging：Diagnose Gmail sender display-name/From-header issues in automated email sends, including cases where the cron job succeeds but the visible sender name appears as mojibake or the wrong account name.
• google-docs-browser-editing：Edit a Google Docs document from the browser tool, including anonymous/link-editable docs. Use synthetic paste into the Docs text-event iframe instead of typed key events.
• google-docs-html-import：Create a new Google Docs document by uploading HTML through Google Drive, then verify the converted document with the Docs API. Useful when browser editing is awkward or you want reliable headings/lists/CJK rendering.
• google-drive-public-docx-reading：Read publicly shared .docx files from a Google Drive folder when browser interaction is flaky or unauthenticated.
• google-slides-api-editing：Directly edit an existing Google Slides deck with the Google Slides API (batchUpdate / presentations.get), preserving template styling and avoiding local PPTX round-trips.
• google-slides-browser-editing：Edit a Google Slides deck directly in the browser tool when link-editable access is available but browser login/API editing is not. Covers how to insert text into slide title/body placeholders via the offscreen Docs text-event iframe, how to switch layouts, and known limitations.
• google-slides-drive-text-audit-fallback：Audit Google Slides decks by searching exported text/plain content from Drive when the Slides API is disabled, unavailable, or unnecessary.
• google-slides-pptx-drive-fallback：Reusable fallback workflow for creating or replacing Google Slides decks by building a local PPTX first and converting it to a Slides file via Google Drive when direct Slides editing or the Slides API is unavailable.
• google-workspace：Gmail, Calendar, Drive, Contacts, Sheets, and Docs integration for Hermes. Uses Hermes-managed OAuth2 setup, prefers the Google Workspace CLI (`gws`) when available for broader API coverage, and falls back to the Python client libraries otherwise.
• google-workspace-copy-and-share-back：Reusable workflow for copying or creating Google Slides, Docs, or Sheets files, then automatically sharing the new file back to the requesting user with editor access.
• gov-tender-monitor：在台灣政府上班日搜尋政府採購網中的關鍵字標案，依 Woolito 主力服務分類、去重後寫入 Notion 標案列表，並支援 cronjob 簡潔回報。
• government-tender-workflow：Woolito 政府標案處理流程：讀招標文件、同步 Notion 標案列表、建立投標準備頁面、必要時移動資料夾並回寫 Obsidian。
• linear：Linear: manage issues, projects, teams via GraphQL + curl.
• maps：Geocode, POIs, routes, timezones via OpenStreetMap/OSRM.
• nano-pdf：Edit PDF text/typos/titles via nano-pdf CLI (NL prompts).
• new-ideas-pool：Create or update a New Ideas Pool in Notion by following the user-specified workflow order from the Rocky Skills Management page.
• notion：Notion API + ntn CLI: pages, databases, markdown, Workers.
• notion-access-discovery：Verify whether a Notion integration can actually see a workspace hub page, then recursively discover child pages and child databases from that hub.
• notion-batch-writing：Write large amounts of content into a Notion page by splitting block payloads into small chunks and verifying each batch.
• notion-bulk-enrichment：Enrich an existing Notion database by discovering its data source ID, patching schema, bulk upserting rows, and verifying results. Use especially when a shared Notion URL points to a database and row writes need careful API handling.
• notion-child-db-id-writeback：Populate a parent database's DB_ID field with each row page's nested child database block id by traversing the page URL, reading page children, and patching the row page property.
• notion-contact-list-audit：Audit a Notion contact/person CRM for obvious bad data, missing titles, test records, duplicate/invalid status markers, and suspicious phone/email/company fields.
• notion-domain-company-repair：Repair blank company-name fields in a Notion contact database by inferring company names from email domains, using Notion API queries plus lightweight web title checks when needed.
• notion-ideas-pool-new-client-setup：Build a new client Ideas Pool from the blank template, including the main page, inline databases, child databases, registration-table fields, and final verification.
• notion-ideas-pool-template-diff-analysis：Compare a blank Ideas Pool template against a newly copied client Ideas Pool and report structural, schema, naming, and content differences before final handoff.
• notion-inline-database-clone-2025-09-03：Clone or create an inline Notion database under a page, then seed it with rows and verify the resulting database/data_source IDs under the 2025-09-03 API.
• notion-inline-page-content-clone：Clone or align content between Notion inline data-source pages by inspecting page blocks, then rebuilding target pages with matching structure and text.
• notion-live-registry-sync：Keep the Notion `Skill 列表` page synchronized with the live Hermes skill registry, and document the workflow in Woolito Obsidian.
• notion-script-table-workflow：Notion-based workflow for rewriting animation/script content in row-based tables, keeping narration and on-screen text separate.
• notion-static-site-export：Assess and export a Notion page hierarchy into a static website with a tree-style sidebar, recursively crawling child pages and nested blocks.
• notion-static-site-sync-and-publish：Build and maintain a static website from a Notion page tree, with a manual sync command, daily auto-sync, and GitHub push support.
• notion-table-block-writing：Notion API table-block write caveat for nested table.children payloads.
• notion-view-replication-from-reference：Replicate a Notion database's views, filters, and gallery/table presentation from a reference database into a template database using the 2026-03-11 Views API.
• notion-views-and-functional-data-sources：Reusable Notion API workflow for mirroring and managing database views plus companion functional data sources on 2026-03-11+, including create/update/list/delete caveats and validation tips.
• notion-workflows：Class-level Notion workflows for lightweight collaboration, live registry/catalog sync, and Ideas Pool setup/verification.
• ocr-and-documents：Extract text from PDFs/scans (pymupdf, marker-pdf).
• powerpoint：Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions "deck," "slides," "presentation," or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.
• slack-app-bot-display-name：Distinguish Slack App name from Bot display name and update the correct setting so the bot shows the desired name inside Slack.
• teams-meeting-pipeline：Operate the Teams meeting summary pipeline via Hermes CLI — summarize meetings, inspect pipeline status, replay jobs, manage Microsoft Graph subscriptions.
• toggl-work-status：Query Toggl Track work status, team activity, and recent time entries for concise human-readable summaries.
• tsmc-hr-ideas-pool-reminder-rule：Manage the annual TSMC HR idea pool reminder schedule, including calendar-based weekend rollover and durable Obsidian writeback.
• validate-docx-readability-on-wsl-mounted-drive：Verify whether .docx files on /mnt/* mounted drives are actually readable, and distinguish valid Word files from zero-filled or placeholder/corrupt files.
• w-chcf-comfyui-storyboard-generation：CHCF / 中基會分鏡圖生成流程：把文字分鏡 prompt 準備好並批次送進本機 Windows ComfyUI Desktop 產出黑白古風分鏡圖。
• w-chcf-comfyui-storyboard-monitoring：CHCF / 中基會分鏡圖監看與收尾流程：監控 ComfyUI queue/history、補跑失敗鏡頭、整理輸出檔案、回寫 Notion、驗證交付。
• w-chcf-comfyui-storyboard-pipeline：CHCF / 中基會分鏡圖總控流程：用一個 prompt 串起生成、監看、補跑、存檔與 Notion 回寫的全流程。
• w-chcf-comfyui-storyboard-revision：CHCF / 中基會特定分鏡圖修改流程：用 Notion API 找出未確認或需重修的鏡頭，依畫面說明與 comment 意見重跑同一套 ComfyUI workflow，並把新版與歷史版同步到 03_storyboard 與 Notion。
• w-chcf-ep-inline-db-bulk-rename：Bulk-rename CHCF / 中基會 EP pages plus their inline child databases and stage rows from placeholder names to person-based names.
• w-chcf-podcast-folder-clone：Clone a CHCF source folder (usually chcf00_OO(RockyVer)) into multiple episode/person folders based on the ordered podcast list on a Notion page, with Windows-safe renaming and post-copy verification.
• w-chcf-project-lookup：CHCF 專案查找規則的短版別名，對應 `woolito-chcf-project-lookup`。
• w-chcf-reference-source-copy：Copy CHCF reference materials from `參考資料/中基會補充資料` into each EP folder's `Footage/01_source`, preserving source filenames and handling both files and folders.
• w-chcf-script-rewrite：CHCF 腳本改寫 skill，從原始歷史文本整理成旁白腳本。
• w-chcf-stage-page-update：CHCF 專案製作項目頁更新流程：找出旁白腳本/文字分鏡等 stage row，寫入內容並同步狀態。
• w-chcf-text-storyboard：CHCF 文字分鏡 skill，將已確認的旁白稿拆成文字分鏡並寫入 Notion。
• woolito-ai-pipeline-planning：規劃 Woolito 類型的 AI 動畫製作流程時，先保留 legacy 專案資料夾，再外掛 stage-based AI pipeline workspace；Google Sheets 管 shot/version 狀態，Google Docs 管長文稿。
• woolito-case-proposal-slides：針對單一案件/客戶/專案製作或擴寫 Google Slides 提案簡報的標準流程；優先使用 Google Slides API，保留原簡報前段與已略過投影片，並依對話與案件資料生成投影片內容。
• woolito-cis-logo-asset-locator：Locate Woolito CIS logo/mark PNG assets in the Dropbox vault by navigating the CIS logo folder structure and identifying likely white/black variants.
• woolito-esg-animation-case-curation：Search for real external ESG / sustainable-finance animation case studies, filter out generic explainer videos, and add the selected cases into Woolito's Notion 電子報資料塊 with detailed introductions and bottom-of-page thumbnail images.
• woolito-esg-animation-direction-research：Research a Woolito/WooPort ESG animation brief by combining public portfolio pages, browser image inspection, and prior session recall to produce a usable animation direction summary when source folders or decks are inaccessible.
• woolito-esg-case-thumbnail-qa：Verify and fix Woolito ESG case-study slides by matching each thumbnail to the correct portfolio artwork, using Woolito works pages, DOM inspection, and final visual QA.
• woolito-esg-tender-monitor：在台灣政府上班日搜尋政府採購網中與 ESG／永續相關的新標案，排除財物與設備採購，避免重複後寫入 Woolito 的 Notion 標案列表。
• woolito-government-tender-ops：Handle Woolito government tender intake across folder storage, Notion tender database, and Obsidian durable records.
• woolito-ideaspool-client-brief-translation：Translate a client request into workable IDEAs POOL solution directions by reading the Notion IDEAs POOL page/database, identifying relevant idea types/use cases, and reframing the request into natural content structures.
• woolito-inbound-lead-research-brief：Research an inbound Woolito lead from Gmail, cross-check Obsidian and Notion for company/contact context, then produce a client brief with reply angles and durable writeback.
• woolito-knowledge-newsletter：Generate and send Woolito Knowledge Base weekly email digests from Notion, with compact HTML card layouts and optional Notion image extraction.
• woolito-no-reply-email-channel-posting：When an inbound email arrives from no-reply-woolito.com@knockers.com.tw, analyze the latest message, summarize the company and request, and post the result to Slack channel G3U9WM13R instead of replying in a thread.
• woolito-notion-case-study-curation：Find external case-study style references and add them into the Woolito Notion `電子報資料塊` database with bottom-of-page thumbnail images, concise descriptions, and unchecked human-review flags.
• woolito-notion-ideas-pool-view-filter-copy：Use when you need to copy a proven Notion Ideas Pool view/filter setup from a template or existing client into a new client pool, including gallery views, quick filters, and cleanup of default views.
• woolito-notion-ideas-pool-view-filter-diff-check：Use when you need to compare a target Woolito Ideas Pool page against the template Ideas Pool and produce a diff focused on Notion view names, types, filters, quick_filters, sorts, configuration, and the page tree beneath the functional section (including child pages and child databases) without making writes.
• woolito-notion-skill-catalog-sync：Keep the Notion `Skill 列表` page synchronized with the live Hermes skill registry, and document the workflow in Woolito Obsidian.
• woolito-notion-static-site-sync：Reusable workflow for exporting a Notion page tree to a static website, keeping a per-page sidebar tree, and syncing updates manually or on a schedule to GitHub/Cloudflare Pages.
• woolito-quote-check：Standard workflow for checking Woolito quotation documents for amount math, quote number consistency, validity dates, and client name consistency.
• woolito-slack-channel-knowledge-rollup：將最近 Slack 各 channel / DM 的對話濃縮回寫到 Woolito Obsidian Vault，建立每個 channel 的優先查詢知識頁。
• woolito-tender-notion-table-template：Standard Notion page structure for Woolito government tender items using table-based main pages, chapter overviews, and subpages.
• woolito-webform-client-research：Woolito inbound webform lead triage for the standard WOOLITO contact-form subject. Research the company, summarize the request, draft reply angles, and post the result to Slack channel G3U9WM13R.
• woolito-wooport-case-recommendation：Recommend WooPort portfolio works for a client brief by matching communication purpose, audience, format, and style rather than only industry labels.
• woolito-wooport-esg-tagging：Cross-check Woolito official website ESG portfolio pages against the WooPort Notion database, then tag matching works with 類型 = ESG.
• woolito-wooport-works-reconciliation：Compare Woolito WooPort Notion works against the public website works page, exclude collection pages, fuzzy-match naming variants, and triage real gaps.
• woolito-works-link-resolution：Find exact Woolito/WooPort portfolio pages and embedded video links for a named project, especially when the user only remembers a client, year, or theme.

red-teaming

• godmode：Jailbreak LLMs: Parseltongue, GODMODE, ULTRAPLINIAN.

research

• arxiv：Search arXiv papers by keyword, author, category, or ID.
• blogwatcher：Monitor blogs and RSS/Atom feeds via blogwatcher-cli tool.
• llm-wiki：Karpathy's LLM Wiki: build/query interlinked markdown KB.
• official-ui-reference-collection：Collect and shortlist official product UI/interface reference images from vendor sites for presentations or moodboards, using browser image discovery plus vision verification.
• polymarket：Query Polymarket: markets, prices, orderbooks, history.
• research-paper-writing：Write ML papers for NeurIPS/ICML/ICLR: design→submit.
• taiwan-company-lead-research：Research a Taiwan company from an inbound lead, especially when normal browser search is blocked by CAPTCHA. Verify company basics, official web presence, scale, and record the result in the Woolito Obsidian Vault when relevant.
• website-mailto-audit-via-sitemap：Audit a website for specific mailto/email strings across all sitemap-discovered URLs, including sub-sitemaps and custom content types.
• woolito-tipo-trademark-check：查詢台灣智慧局（TIPO）商標檢索系統，快速確認名稱是否已有同名或近似商標，並取回案件明細。適合品牌命名、公司名初篩、商標風險快速檢查。

smart-home

• openhue：Control Philips Hue lights, scenes, rooms via OpenHue CLI.

social-media

• threads-facebook-content-style-analysis：Inspect Threads and Facebook pages in Chrome to analyze recurring post formats, question styles, and response patterns by collecting multiple posts through direct URLs and DOM text extraction.
• threads-poll-posting：Compose and publish a Threads post with poll options using the browser/MCP UI, including the key interaction pattern for opening the composer, enabling poll mode, filling the question and choices, and verifying the post is published.
• xurl：X/Twitter via xurl CLI: post, search, DM, media, v2 API.

software-development

• debugging-hermes-tui-commands：Debug Hermes TUI slash commands: Python, gateway, Ink UI.
• hermes-agent-skill-authoring：Author in-repo SKILL.md: frontmatter, validator, structure.
• hermes-dashboard-skills-filters：Add or extend Hermes dashboard Skills page filtering/search, including skill tags, API plumbing, i18n strings, and tests.
• hermes-kling-wsl-bridge：Bridge a WSL-based agent to Windows ComfyUI Desktop for Kling API-node workflows, including auth token injection, submit, and history sync.
• hermes-s6-container-supervision：Modify, debug, or extend the s6-overlay supervision tree inside the Hermes Agent Docker image — adding new services, debugging profile gateways, understanding the Architecture B main-program pattern.
• node-inspect-debugger：Debug Node.js via --inspect + Chrome DevTools Protocol CLI.
• plan：Plan mode: write markdown plan to .hermes/plans/, no exec.
• python-debugpy：Debug Python: pdb REPL + debugpy remote (DAP).
• requesting-code-review：Pre-commit review: security scan, quality gates, auto-fix.
• rocky-manual-sidebar-ui-polish：Workflow for refining Rocky Manual sidebar UX and branding details such as collapsible edge handles, icon alignment, and favicon updates, with generator sync and browser verification.
• rocky-manual-skill-catalog-filters：Build and update Rocky Manual's Skill list page with search-based browsing, optionally tag filtering, then regenerate the static site and push to GitHub.
• spike：Throwaway experiments to validate an idea before build.
• static-shell-client-router：Reusable workflow for fixing static multi-page HTML sites so the left shell/sidebar stays stable while the content pane updates via client-side navigation, including file:// compatibility via preloaded page snapshots.
• static-skill-catalog-page：Build or extend a static skills catalog page from a directory of SKILL.md files, with search, tag filters, and tag-aware cards.
• subagent-driven-development：Execute plans via delegate_task subagents (2-stage review).
• systematic-debugging：4-phase root cause debugging: understand bugs before fixing.
• test-driven-development：TDD: enforce RED-GREEN-REFACTOR, tests before code.
• vendor-api-integration：Integrate third-party SaaS APIs/SDKs for content workflows, especially email/newsletter builders, while keeping auth, content generation, and delivery responsibilities separate.
• writing-plans：Write implementation plans: bite-sized tasks, paths, code.
• wsl-windows-chrome-mcp-bridge：Diagnose and set up Windows Chrome control from Hermes running in WSL using Playwright MCP, with guidance on when to switch to a Windows-side MCP runtime.
• wsl-windows-comfyui-bridge：Access a Windows-hosted ComfyUI instance from WSL when localhost is not reachable directly. Use PowerShell Invoke-WebRequest as the bridge, handle Windows encoding, and build valid ComfyUI prompt payloads for dynamic combo nodes.