One binary that authenticates, installs MCP into your AI client, and generates images, videos, and music. Built for Claude Code, Codex, and anyone who lives in the terminal.
Install once, log in once, then use the CLI directly or have it wire up your favorite assistant.
Globally install the Invoomen CLI from npm. (Not yet on npm — for now: git clone https://github.com/invoomen/cli && cd cli && npm install && npm run build && npm link.)
npm install -g @invoomen/cliOpens your browser, you copy your key once, the CLI stores it locally.
invoomen auth loginEither use the CLI directly or have it write the MCP config for Claude Code, Codex, Cursor, or Claude Desktop.
invoomen generate image "cinematic dolly-in of a hummingbird"Everything you can do without leaving the shell.
invoomen auth loginOpen your browser, create a key, store it locally in ~/.invoomen/config.json.
invoomen mcp install <client>Write the right MCP config for claude, claude-code, codex, or cursor.
invoomen generate image "…"Generate an image. Resolves a default model, waits, prints the URL.
invoomen generate video "…"Generate a video. Async — prints a generation_id and polls until ready.
invoomen generate music "…"Spin up a Suno music generation and download the result.
invoomen balancePrint your current credit balance and renewal date.
invoomen list modelsBrowse the model catalog. Filter with --kind image|video|music|audio.
invoomen list generationsRecent generations in your active workspace.
The CLI knows where each client stores its config and writes the right snippet for you.
invoomen mcp install claude~/.claude/claude_desktop_config.json (Claude Desktop / Claude Code)
invoomen mcp install claude-codeRuns 'claude mcp add' so Claude Code picks it up project-wide
invoomen mcp install codex~/.codex/config.toml (OpenAI Codex CLI)
invoomen mcp install cursor~/.cursor/mcp.json (Cursor IDE)
The CLI is a thin wrapper around the Invoomen MCP server — same tools, same models, same credits.
33+ image models including Seedream, Nano Banana, Imagen 4, Qwen2 Edit, GPT Image 2.
Kling, Wan 2.7, Seedance, Grok Imagine, and more — up to 4K, 15-second clips.
Full Suno v5 — generate tracks, lyrics, covers, vocal extraction, MIDI.
ElevenLabs dialogue v3, audio isolation, and Suno persona/vocal models.
Reference your saved characters by @name in any prompt — no re-upload needed.
Reusable props, scenes, and styles attach to generations with @mentions.
Bills the same Invoomen credits as the website. Team workspaces share the credit pool.
Works with Claude.ai, Claude Desktop, Claude Code, Codex CLI, Cursor, ChatGPT.
Scripts, automations, and side-by-side experimentation.
Generate product imagery, social ads, and short-form video for campaigns straight from chat.
Spin up thumbnails, b-roll, intros, and music beds without leaving your editor.
Pre-viz, storyboards, character consistency across shots, and animatic videos.
Catalog imagery, lifestyle shots, packaging variants, and product launch video assets.
Run multiple client brands from one workspace; reference saved brand kits as characters.
Texture references, mood boards, palette explorations, and asset variations.
One command to install. One command to log in. Generate from anywhere.
Create your API keyOpen source on GitHub.