Skill Library

Delegate complex coding tasks — features, PRs, and refactors — to Claude Code CLI. Spawns an autonomous subagent with full terminal and file access in an isolated workspace.

Delegate coding tasks to OpenAI's Codex CLI. Spawns an autonomous coding agent for feature implementation, bug fixes, and PRs using GPT/Codex models.

Configure, extend, or contribute to Hermes Agent itself. Covers CLI setup, config management, model/provider configuration, tool enable/disable, skills, voice, gateway, and plugins.

Delegate coding tasks to OpenCode CLI. Supports feature implementation, code review, and PR workflows with an open-source coding agent.

Create dark-themed SVG architecture, cloud, and infrastructure diagrams rendered as interactive HTML. Supports system context, container, and deployment diagrams.

Generate ASCII art using pyfiglet, cowsay, boxes, and image-to-ASCII conversion. Text banners, code art, and terminal decorations.

Convert video and audio to colored ASCII MP4 or GIF. Renders frames as ANSI-colored ASCII art for terminal-friendly video playback.

Create knowledge comics (知识漫画) — educational, biographical, and tutorial comics with panel layouts and illustrated storytelling. Chinese and English supported.

Create infographics with 21 layouts × 21 style combinations. Supports information visualization, data storytelling, and educational graphics.

Design one-off HTML artifacts — landing pages, slide decks, interactive prototypes. Full control over layout, animations, and interactivity via raw HTML/CSS/JS.

Generate images, video, and audio using ComfyUI. Install workflows, run inference pipelines, and manage ComfyUI nodes for Stable Diffusion and other generative models.

Author, validate, and export Google-style DESIGN.md token spec files. Structured design documentation for AI-assisted development.

Create hand-drawn style diagrams as Excalidraw JSON files. Supports architecture diagrams, flowcharts, sequence diagrams — exported as .excalidraw files for import into excalidraw.com.

Rewrite AI-generated text to sound more natural and human. Strips AI-isms, adds authentic voice, adjusts tone, and makes content read like a real person wrote it.

Generate project ideas via creative constraints. Uses structured brainstorming techniques to produce novel, specific, and actionable project concepts.

Create mathematical and algorithmic animations using Manim Community Edition. 3Blue1Brown-style explanatory videos with geometric animations, equation rendering, and scene scripting.

Create interactive p5.js sketches — generative art, shaders, interactive visualizations, and 3D graphics. Runs in the browser as self-contained HTML pages.

Create pixel art with era-authentic palettes — NES (52 colors), Game Boy (4 shades), PICO-8 (16 colors). Output as PNG with proper scaling.

Access 54 real design system templates from Stripe, Linear, Vercel, Supabase, Tailwind CSS, and more. Use as starting points for HTML/CSS projects.

Throwaway HTML mockups — create 2-3 design variants to compare. Quick visual brainstorming for UI ideas without any build tooling.

Songwriting craft guidance and Suno AI music prompts. Generate lyrics, chord progressions, and Suno-optimized prompts for AI music generation.

Scaffold a static tutorial blog about a technical tool. Complete setup with posts, navigation, dark theme, and deployment config.

Iterative Python development via a live Jupyter kernel (hamelnb). Execute code blocks interactively, inspect variables, and build data analysis pipelines step by step.

Decomposition playbook for kanban-based multi-agent workflows. Break complex projects into tasks, assign to worker profiles, and track progress on the board.

Pitfalls, examples, and edge cases for kanban-based work execution. Worker profile that picks up tasks from the board, executes them, and reports results.

Event-driven agent runs via webhook subscriptions. Configure triggers that wake up the agent on external events (GitHub pushes, CI completions, API calls).

Exploratory QA testing of web applications. Systematic approach to finding bugs, collecting evidence, and producing QA reports.

Send, receive, search, and manage email from the terminal using the Himalaya CLI. IMAP/SMTP-based email client with folder management.

Host and manage modded Minecraft servers using CurseForge or Modrinth modpacks. Full setup from server installation to mod management and player configuration.

Play Pokemon games via a headless emulator with RAM reads. Automated gameplay with memory inspection for game state awareness.

Analyze codebases with pygount — get LOC counts, language breakdowns, file type distributions, and code ratio metrics.

Setup GitHub authentication — HTTPS tokens, SSH keys, gh CLI login, and fine-grained PAT configuration.

Review pull requests — fetch diffs, add inline comments via gh CLI or REST API, approve or request changes.

Create, triage, label, assign, and manage GitHub issues via gh CLI or REST API. Full issue lifecycle management.

Full PR lifecycle — create branch, commit changes, push, open PR, wait for CI, and merge. Follows best practices for conventional commits and PR descriptions.

Clone, create, fork repositories; manage remotes, releases, and repository settings via API.

Model Context Protocol (MCP) client built into Hermes. Connect to MCP servers (stdio or HTTP/SSE), register their tools, and use them alongside built-in tools.

Search and download GIFs from Tenor using curl and jq. Quick GIF lookup for reactions, explanations, and visual content.

Suno-like song generation from lyrics and style tags. Generate original music with AI from text descriptions of genre, mood, and structure.

Audio analysis and visualization — generate spectrograms, mel-spectrograms, chromagrams, and MFCC features from audio files. CLI-based audio feature extraction.

Full Spotify playback control — 7 tools for playback, devices, queue, search, playlists, albums, and library. Uses Spotify Web API with PKCE OAuth.

Convert YouTube video transcripts to summaries, social media threads, blog posts, and other content formats. Extract and repurpose video content.

Search, download, and upload models and datasets from Hugging Face using the hf CLI. Full hub interaction — model discovery, file management, and versioning.

Benchmark LLMs using the lm-eval-harness framework. Run standardized evaluations (MMLU, GSM8K, HumanEval, etc.) on any model with a supported backend.

Log ML experiments, run hyperparameter sweeps, manage model registries, and create dashboards with Weights & Biases integration.

Local GGUF model inference with llama.cpp. Run quantized LLMs on CPU/GPU from Hugging Face. Server mode, CLI chat, and API endpoints.

Abliterate (remove) LLM refusal behaviors using the diff-in-means technique. Modifies model weights to reduce safety refusals for research purposes.

High-throughput LLM serving with vLLM. OpenAI-compatible API, PagedAttention, continuous batching, and quantization (AWQ/GPTQ/FP8).

Audio generation with Meta's AudioCraft — MusicGen for text-to-music and AudioGen for text-to-sound. Generates audio from natural language descriptions.

Zero-shot image segmentation using Meta's Segment Anything Model. Segment objects by points, bounding boxes, or automatic mask generation.

Declarative LM programming with auto-optimization. Build RAG pipelines, classifiers, and multi-step LM programs that optimize their own prompts.

Read, search, create, and edit notes in an Obsidian vault. Wiki-link cross-referencing, full-text search, and multi-note operations.

Full Airtable API interaction via curl — records CRUD, filters, sorting, upserts, and base schema operations.

Run quality assurance on blog posts — frontmatter checks, content validation, description length limits, and formatting. Automated post review script.

Gmail, Calendar, Drive, Docs, and Sheets management via the gws CLI or direct Python integration. Read and manage Google services from the terminal.

Manage Linear issues, projects, and teams via GraphQL API + curl. Issue lifecycle, project management, team operations, and workflow automation.

Geocoding, POI search, routing, and timezone lookup using OpenStreetMap and OSRM. Free, no API key needed for basic usage.

Edit PDF text, fix typos, update titles, and modify PDF metadata using natural language prompts. CLI-based PDF editing without layout rebuilding.

Notion API integration via curl — pages, databases, blocks, and search. Read, create, and update Notion content from the terminal.

Extract text from PDFs and scanned documents using PyMuPDF (text PDFs) and marker-pdf (scanned/image PDFs with OCR).

Create, read, edit .pptx presentations with python-pptx. Full slide management: add slides, text, tables, charts, images, and speaker notes.

Operate the Teams meeting summary pipeline via Hermes CLI. Fully automated pipeline: join meeting → transcribe → summarize → distribute.

Research and security testing — techniques for understanding LLM safety boundaries. Includes Parseltongue, GODMODE, and ULTRAPLINIAN patterns for academic red-teaming.

Search arXiv papers by keyword, author, category, or ID. Get paper details, abstracts, and download links.

Monitor blogs and RSS/Atom feeds via blogwatcher-cli. Subscribe to feeds, detect new posts, and summarize updates.

Karpathy's LLM Wiki pattern — build and query an interlinked markdown knowledge base for personal research and learning.

Research and verify API pricing, free tier availability, rate limits, and model capabilities across LLM providers.

Query Polymarket prediction markets — see market prices, orderbooks, trade history, and market details.

Evaluate and compare websites, blogs, and online services. Design, UX, performance, and feature comparisons.

Control Philips Hue lights, scenes, and rooms via OpenHue CLI. Turn lights on/off, set brightness, change colors, and activate scenes.

Post, search, DM, and manage media on X/Twitter via xurl CLI. Full v2 API support for tweets, threads, DMs, and account management.

Build catalog/directory pages with search, category filters, and card grids — like the tools, plugins, gateway, and hosting pages on this blog.

Debug Hermes Agent's Terminal UI slash commands — Python backend, Ink Gateway, and React Ink component debugging.

Author Hermes Agent skills. YAML frontmatter, markdown body, validator checks, and SKILL.md structure guidelines.

Debug Node.js applications via --inspect flag and Chrome DevTools Protocol CLI. Inspect heap, CPU profiles, breakpoints, and async stacks.

Write markdown plans to .hermes/plans/ without executing anything. Pure planning mode for architecture design and implementation strategy.

Debug Python code with pdb REPL and debugpy remote debugging (DAP protocol). Set breakpoints, inspect variables, and debug remotely.

Pre-commit code review — security scan, quality gates, automated fixes, and review checklist generation.

Post-task reflection cycle. After every complex task, analyze mistakes, apply user corrections, and encode lessons as skills or memory.

Throwaway experiments to validate an idea before building. Quick prototypes to test feasibility, performance, or approach.

Execute implementation plans via delegate_task subagents with a two-stage review process. Decompose work, parallelize, and verify.

4-phase root cause debugging: understand the bug before fixing. Reproduce → isolate → identify root cause → fix with verification.

Red-Green-Refactor TDD cycle. Write tests before code, make them pass, then refactor. Enforces test-first discipline.

Full tutorial blog workflow — scaffold, theme, SEO/GEO optimization, JSON-LD structured data, reference pages, and Cloudflare deployment.

Write implementation plans with bite-sized tasks, file paths, and estimated code. Structured planning before execution.

Interact with Yuanbao (Tencent) group platform — @mention users, query group members, and retrieve group information.