For developers & AI agents

MCP & CLI

Use LensRoom 3.0 from your terminal, your IDE, or any MCP-compatible assistant. Generate cinematic images, videos, and audio across 25+ models — without leaving your workflow.

Quick start

1Install the CLI

npm install -g lensroom-cli

2Sign in

lensroom auth login

3Add the skills

npx skills add lensroom/Generative-Media-Skills

CLI

lensroom-cli

Generate images, videos, and audio from the terminal across 25+ AI models. Dual interface — colored human output plus JSON for agents (--output-json, --jq filtering). Async workflows, file uploads, credit tracking.

npm install -g lensroom-cli
lensroom auth login
lensroom image generate "a cyberpunk city" \
  --model nano-banana

View on GitHub

MCP

lensroom-mcp-server

Connect Claude, Cursor, Windsurf, and any MCP-compatible assistant to 25+ generative models. Hosted endpoint — no install. Structured tools with input/output schemas, async polling, and account management.

claude mcp add --transport http lensroom \
  https://v3.lensroom.ru/mcp \
  --header "Authorization: Bearer YOUR_KEY"

View on GitHub

Skills

Generative Media Skills

Multimodal toolkit for Claude Code, Cursor, and Gemini CLI. Cinema Director, Nano-Banana, UI Designer, Logo Creator, Seedance 2, AI Clipping, and YouTube Shorts presets. Agent-native with JSON outputs and semantic exit codes.

npx skills add lensroom/Generative-Media-Skills --all

View on GitHub

Examples

Image generation

lensroom image generate "a serene mountain lake at sunrise" \
  --model nano-banana --download ./outputs

Text-to-video

lensroom video generate "a dog running on a beach" \
  --model kling-3.0

Audio creation

lensroom audio create "upbeat lo-fi hip hop for studying"

Run a skill

bash library/visual/nano-banana/scripts/\
  generate-nano-art.sh --file image.jpg --view

Open-source · MIT licensed · Works with Claude, Cursor, Windsurf, and Gemini CLI