全部技能介绍，一页总览

Delegate coding to Claude Code CLI (features, PRs).

Delegate coding to OpenAI Codex CLI (features, PRs).

Complete guide to using and extending Hermes Agent — CLI usage, setup, configuration, spawning additional agents, gateway platforms, skills, voice, tools, profiles, and a concise contributor reference. Load this skill when helping users configure Hermes, troubleshoot issues, spawn agent instances, or make code contributions.

Practical findings from running Hermes doctor and doctor --fix, especially around auto-remediation limits and follow-up checks.

Delegate coding to OpenCode CLI (features, PR review).

Exploratory QA of web apps: find bugs, evidence, reports.

Yuanbao (元宝) groups: @mention users, query info/members.

Dark-themed SVG architecture/cloud/infra diagrams as HTML.

ASCII art: pyfiglet, cowsay, boxes, image-to-ascii.

ASCII video: convert video/audio to colored ASCII MP4/GIF.

Knowledge comics (知识漫画): educational, biography, tutorial.

Infographics: 21 layouts x 21 styles (信息图, 可视化).

Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.

Design one-off HTML artifacts (landing, deck, prototype).

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.

Author/validate/export Google's DESIGN.md token spec files.

Hand-drawn Excalidraw JSON diagrams (arch, flow, seq).

Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, artifacts, posters, or applications (examples include websites, landing pages, dashboards, React components, HTML/CSS layouts, or when styling/beautifying any web UI). Generates creative, polished code and UI design that avoids generic AI aesthetics.

Humanize text: strip AI-isms and add real voice.

Generate project ideas via creative constraints.

Manim CE animations: 3Blue1Brown math/algo videos.

p5.js sketches: gen art, shaders, interactive, 3D.

Pixel art w/ era palettes (NES, Game Boy, PICO-8).

54 real design systems (Stripe, Linear, Vercel) as HTML/CSS.

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.

Throwaway HTML mockups: 2-3 design variants to compare.

Songwriting craft and Suno AI music prompts.

Control a running TouchDesigner instance via twozero MCP — create operators, set parameters, wire connections, execute Python, build real-time visuals. 36 native tools.

Iterative Python via live Jupyter kernel (hamelnb).

UI/UX design intelligence for web and mobile. Includes 50+ styles, 161 color palettes, 57 font pairings, 161 product types, 99 UX guidelines, and 25 chart types across 10 stacks (React, Next.js, Vue, Svelte, SwiftUI, React Native, Flutter, Tailwind, shadcn/ui, and HTML/CSS). Actions: plan, build, create, design, implement, review, fix, improve, optimize, enhance, refactor, and check UI/UX code. Projects: website, landing page, dashboard, admin panel, e-commerce, SaaS, portfolio, blog, and mobile app. Elements: button, modal, navbar, sidebar, card, table, form, and chart. Styles: glassmorphism, claymorphism, minimalism, brutalism, neumorphism, bento grid, dark mode, responsive, skeuomorphism, and flat design. Topics: color systems, accessibility, animation, layout, typography, font pairing, spacing, interaction states, shadow, and gradient. Integrations: shadcn/ui MCP for component search and examples.

Decomposition playbook + specialist-roster conventions + 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.

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.

Bind a static site to a custom domain in nginx, issue Let's Encrypt certificates with certbot, and enforce a canonical host (for example redirect apex to www) with verification steps.

Bind a subdomain to a local service port with nginx reverse proxy, verify DNS/port readiness first, and optionally issue Let's Encrypt certificates.

Install Nginx UI on a Linux server, verify the systemd service and listening port, retrieve the one-time install secret, and assess whether to expose it directly or behind an HTTPS subdomain.

Inspect a website backup archive, identify whether it contains a full stack or a static site, safely extract only the static homepage, preview it locally, and prepare for nginx deployment.

Configure a custom long-running app to auto-start with user-level systemd, verify readiness, and safely retire an older Hermes-managed background process if it conflicts.

Structured self-debugging workflow for AI agent failures using capture, diagnosis, contained recovery, and introspection reports.

Build an evidence-backed ECC install plan for a specific repo by sorting skills, commands, rules, hooks, and extras into DAILY vs LIBRARY buckets using parallel repo-aware review passes. Use when ECC should be trimmed to what a project actually needs instead of loading the full bundle.

Regression testing strategies for AI-assisted development. Sandbox-mode API testing without database dependencies, automated bug-check workflows, and patterns to catch AI blind spots where the same model writes and reviews code.

Create CodeTour `.tour` files — persona-targeted, step-by-step walkthroughs with real file and line anchors. Use for onboarding tours, architecture walkthroughs, PR tours, RCA tours, and structured "explain how this works" requests.

Interactive installer for Everything Claude Code — guides users through selecting and installing skills and rules to user-level or project-level directories, verifies paths, and optionally optimizes installed files.

Automatically extract reusable patterns from Claude Code sessions and save them as learned skills for future use.

Instinct-based learning system that observes sessions via hooks, creates atomic instincts with confidence scoring, and evolves them into skills/commands/agents. v2.1 adds project-scoped instincts to prevent cross-project contamination.

Convene a four-voice council for ambiguous decisions, tradeoffs, and go/no-go calls. Use when multiple valid paths exist and you need structured disagreement before choosing.

Playwright E2E testing patterns, Page Object Model, configuration, CI/CD integration, artifact management, and flaky test strategies.

Formal evaluation framework for Claude Code sessions implementing eval-driven development (EDD) principles

This skill should be used when the user asks to create a hookify rule, write a hook rule, configure hookify, add a hookify rule, or needs guidance on hookify rule syntax and patterns.

Pattern for progressively refining context retrieval to solve the subagent context problem

Write-time code quality enforcement using Plankton — auto-formatting, linting, and Claude-powered fixes on every file edit via hooks.

Use when auditing Claude skills and commands for quality. Supports Quick Scan (changed skills only) and Full Stocktake modes with sequential subagent batch evaluation.

Himalaya CLI: IMAP/SMTP email from terminal.

Host modded Minecraft servers (CurseForge, Modrinth).

Play Pokemon via headless emulator + RAM reads.

Inspect codebases w/ pygount: LOC, languages, ratios.

GitHub auth setup: HTTPS tokens, SSH keys, gh CLI login.

Review PRs: diffs, inline comments via gh or REST.

Create, triage, label, assign GitHub issues via gh or REST.

GitHub PR lifecycle: branch, commit, open, CI, merge.

Clone/create/fork repos; manage remotes, releases.

Find nearby places (restaurants, cafes, bars, pharmacies, etc.) using OpenStreetMap. Works with coordinates, addresses, cities, zip codes, or Telegram location pins. No API keys needed.

Use the mcporter CLI to list, configure, auth, and call MCP servers/tools directly (HTTP or stdio), including ad-hoc servers, config edits, and CLI/type generation.

MCP client: connect servers, register tools (stdio/HTTP).

Search/download GIFs from Tenor via curl + jq.

HeartMuLa: Suno-like song generation from lyrics + tags.

Audio spectrograms/features (mel, chroma, MFCC) via CLI.

Spotify: play, search, queue, manage playlists and devices.

AudioCraft: MusicGen text-to-music, AudioGen text-to-sound.

Axolotl: YAML LLM fine-tuning (LoRA, DPO, GRPO).

OpenAI's model connecting vision and language. Enables zero-shot image classification, image-text matching, and cross-modal retrieval. Trained on 400M image-text pairs. Use for image search, content moderation, or vision-language tasks without fine-tuning. Best for general-purpose image understanding.

DSPy: declarative LM programs, auto-optimize prompts, RAG.

lm-eval-harness: benchmark LLMs (MMLU, GSM8K, etc.).

TRL: SFT, DPO, PPO, GRPO, reward modeling for LLM RLHF.

HuggingFace hf CLI: search/download/upload models, datasets.

llama.cpp local GGUF inference + HF Hub model discovery.

Serverless GPU cloud platform for running ML workloads. Use when you need on-demand GPU access without infrastructure management, deploying ML models as APIs, or running batch jobs with automatic scaling.

OBLITERATUS: abliterate LLM refusals (diff-in-means).

Outlines: structured JSON/regex/Pydantic LLM generation.

Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.

Expert guidance for Fully Sharded Data Parallel training with PyTorch FSDP - parameter sharding, mixed precision, CPU offloading, FSDP2

SAM: zero-shot image segmentation via points, boxes, masks.

vLLM: high-throughput LLM serving, OpenAI API, quantization.

State-of-the-art text-to-image generation with Stable Diffusion models via HuggingFace Diffusers. Use when generating images from text prompts, performing image-to-image translation, inpainting, or building custom diffusion pipelines.

Unsloth: 2-5x faster LoRA/QLoRA fine-tuning, less VRAM.

W&B: log ML experiments, sweeps, model registry, dashboards.

OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M params). Use for speech-to-text, podcast transcription, or multilingual audio processing. Best for robust, multilingual ASR.

Write ML papers for NeurIPS/ICML/ICLR: design→submit.

Read, search, create, and edit notes in the Obsidian vault.

Airtable REST API via curl. Records CRUD, filters, upserts.

Gmail, Calendar, Drive, Docs, Sheets via gws CLI or Python.

Linear: manage issues, projects, teams via GraphQL + curl.

Geocode, POIs, routes, timezones via OpenStreetMap/OSRM.

Edit PDF text/typos/titles via nano-pdf CLI (NL prompts).

Notion API via curl: pages, databases, blocks, search.

Extract text from PDFs/scans (pymupdf, marker-pdf).

Create, read, edit .pptx decks, slides, notes, templates.

Jailbreak LLMs: Parseltongue, GODMODE, ULTRAPLINIAN.

Search arXiv papers by keyword, author, category, or ID.

Monitor blogs and RSS/Atom feeds via blogwatcher-cli tool.

Karpathy's LLM Wiki: build/query interlinked markdown KB.

Query Polymarket: markets, prices, orderbooks, history.

YouTube transcripts to summaries, threads, blogs.

Control Philips Hue lights, scenes, rooms via OpenHue CLI.

Interact with X/Twitter via the x-cli terminal client using official X API credentials. Use for posting, reading timelines, searching tweets, liking, retweeting, bookmarks, mentions, and user lookups.

X/Twitter via xurl CLI: post, search, DM, media, v2 API.

Debug Hermes TUI slash commands: Python, gateway, Ink UI.

Install and sanity-check GitBook from the GitbookIO GitHub repo, while distinguishing it from the legacy gitbook-cli and handling common environment pitfalls.

Author in-repo SKILL.md: frontmatter, validator, structure.

Behavioral guidelines to reduce common LLM coding mistakes. Derived from Andrej Karpathy's observations. Use ALWAYS when writing, reviewing, or refactoring code. Covers assumption surfacing, simplicity, surgical edits, and goal-driven execution with verification.

Debug Node.js via --inspect + Chrome DevTools Protocol CLI.

Plan mode: write markdown plan to .hermes/plans/, no exec.

Debug Python: pdb REPL + debugpy remote (DAP).

Pre-commit review: security scan, quality gates, auto-fix.

Throwaway experiments to validate an idea before build.

Convert a single-file static HTML app that stores data in localStorage into a lightweight SQLite-backed web app with a Python stdlib HTTP server, CRUD API, CSV import/export, and a minimal end-to-end verification script.

Execute plans via delegate_task subagents (2-stage review).

4-phase root cause debugging: understand bugs before fixing.

TDD: enforce RED-GREEN-REFACTOR, tests before code.

Write implementation plans: bite-sized tasks, paths, code.

Playwright E2E testing patterns, Page Object Model, configuration, CI/CD integration, artifact management, and flaky test strategies.