NousResearch/hermes-agent

1-3-1 Communication Rule produces structured decision briefs that present one problem, three options with trade-offs, and one recommendation. Use it when a task has multiple viable approaches and the user needs a clear proposal.

Antigravity CLI operates the agy command-line interface for plugins, authentication, and sandbox configuration. Use it to install, debug, and run one-shot or interactive Antigravity coding-agent sessions from Hermes.

Article Illustrator analyzes articles, identifies illustration positions, and generates images with consistent type, style, and palette. Use it when asked to illustrate an article or add images to content.

AudioCraft Audio Generation provides guidance for using Meta's AudioCraft models (MusicGen, AudioGen, EnCodec) to generate music and sound effects from text descriptions. Use it when building applications that require text-to-music, melody-conditioned generation, or text-to-sound synthesis.

Blackbox delegates coding tasks to the Blackbox AI multi-model CLI, which runs prompts through multiple LLMs and selects the best result. Use it for one-shot coding tasks, PR reviews, background long-running work, and checkpoint-resumed sessions.

Concept Diagrams generates flat, minimal educational SVG visuals as a single self-contained HTML file with automatic light/dark mode. Use it for physics setups, biology diagrams, floor plans, and process flows when a more specialized diagram skill is not available.

Creative Ideation generates ideas via named methods from creative practice. It routes a user's situation to the right technique and produces specific, non-obvious output for any generative or selective question.

EVM Blockchain Data reads wallets, tokens, transactions, gas prices, ENS names, and contract details across eight EVM chains. Use it when an agent needs to inspect on-chain activity without an API key.

Google Meet joins a Google Meet call, transcribes live captions, optionally speaks in realtime, and handles follow-up work like notes and summaries. Use when the user asks the agent to sit in on a meeting, take notes, summarize, respond in-call, or action items from it.

Grok Build delegates coding tasks to xAI's Grok Build CLI through the Hermes terminal, supporting headless one-shot runs, interactive PTY sessions, and ACP agent mode. Use it for feature building, refactoring, PR reviews, and batch issue fixing.

HeartMuLa generates songs from lyrics and style tags using open-source music foundation models. Use it when you need a local or open-source alternative to commercial AI music services like Suno.

Honcho Memory configures and troubleshoots Honcho AI-native cross-session user modeling for Hermes agents. Use it to set up cloud or self-hosted memory, manage peer observation and recall modes, and tune cadence, depth, and context injection.

HyperFrames renders MP4/WebM videos from HTML compositions. Use it for motion graphics, captioned narration videos, audio-reactive visuals, social overlays, and website-to-video promos driven by HTML/CSS/JS animations.

Hyperliquid retrieves market data, account balances, positions, fills, and orders from the Hyperliquid exchange public API. Use it for market analysis, account review, and post-trade reporting.

Jupyter Notebook provides a stateful Python REPL via a live Jupyter kernel (hamelnb) so variables persist across executions. Use it for iterative data exploration, API inspection, and complex code development.

Kanban Video Orchestrator plans and runs multi-agent video production pipelines by decomposing video requests into specialized agent roles and managing them via a Kanban board. Use it to orchestrate complex video projects that require multiple specialist agents rather than a single script.

Knowledge Comic Creator generates educational, biography, and tutorial comics from user-provided text, URLs, or topics, with configurable art styles, tones, layouts, and languages. Use it when a user asks for a knowledge comic, 知识漫画, or Logicomix-style explanation.

OpenHands delegates coding tasks to the OpenHands CLI in headless mode. Use it when a user wants a coding agent that can run on any LiteLLM-supported provider, not just a single vendor.

Pixel Art converts images into retro pixel art using era-accurate palettes like NES, Game Boy, and PICO-8, then optionally animates them into short MP4 or GIF scenes. Use it when a user wants pixel-art styling or looping animations from a source image.

Social Media Content Calendar plans multi-platform social campaigns from briefs through scheduling and verified publishing. Use it when building a calendar, turning launches into posts, or repurposing content across X, LinkedIn, Instagram, and TikTok.

Solana queries wallets, token holdings, transactions, NFTs, whale transfers, and network stats on the Solana blockchain with USD pricing via CoinGecko. Use it when you need to inspect an address, verify a transaction, check token metadata or prices, or monitor network health.

Tldraw Offline lets an agent read and edit canvases in the tldraw offline desktop app via its local HTTP API, and embed document scripts for durable behavior like reactive shapes and interactive buttons. Use for diagrams, wireframes, and scripted interactions.

Unreal MCP automates Unreal Engine editor scenes, actors, and renders through the official MCP plugin. Use it for building levels, lighting, material tuning, camera framing, and screenshot capture in live Unreal Editor sessions.

Yuanbao lets an agent interact with Yuanbao groups: query group info and members, @mention users, and send direct messages. Use it when the agent needs to respond in a Yuanbao chat or notify someone.