cowork-media-mcp
Lets Claude Cowork/Desktop generate images via Gemini Flash Image and videos via Google Veo through plain language requests.
README
cowork-media-mcp
An MCP connector that lets Claude Cowork / Claude Desktop generate media:
- Images through Google's Gemini Flash Image ("nano-banana")
- Video through Google's Veo
Once it's wired in, you just ask Claude in plain language ("make a logo of…", "generate a 5-second clip of…") and it routes image jobs to Gemini and video jobs to Veo on its own. Claude Cowork runs Claude, not Gemini or Veo, so it can't call them natively — this connector is the bridge.
What you need
A Google Generative Language API key (this is separate from anything on your BimRoss/Ross setup — it's your own personal key for your desktop Claude). Get one free at https://aistudio.google.com/apikey.
You also need Node.js 18+ installed. Check with node --version.
Install
Option A — no clone (recommended)
Add this to your Claude config file and restart Claude Desktop / Cowork:
{
"mcpServers": {
"media": {
"command": "npx",
"args": ["-y", "github:BimRoss/cowork-media-mcp"],
"env": {
"GEMINI_API_KEY": "PASTE-YOUR-KEY-HERE"
}
}
}
}
The config file lives at:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
If the file already has an mcpServers block, add the "media" entry inside
it instead of pasting a second block.
Option B — clone it locally
git clone https://github.com/BimRoss/cowork-media-mcp
cd cowork-media-mcp
npm install
Then point the config at the local file (swap in the real path):
{
"mcpServers": {
"media": {
"command": "node",
"args": ["/full/path/to/cowork-media-mcp/index.js"],
"env": {
"GEMINI_API_KEY": "PASTE-YOUR-KEY-HERE"
}
}
}
}
Use it
Restart Claude, then just ask:
- "Use the media tool to make a neon logo of a running shoe on a dark background."
- "Generate a 5-second video of waves at sunset."
Finished files save to ~/cowork-media/ by default. Images also come back
inline in the chat.
Tools
| Tool | Model | Notes |
|---|---|---|
generate_image |
Gemini Flash Image | Fast, cheap. Returns inline + saves to disk. |
generate_video |
Veo 2 | Async, 1–3 min. ~$2.80 per clip, billed every call. |
Config knobs (optional env entries)
GEMINI_IMAGE_MODEL— defaultgemini-3.1-flash-imageGEMINI_VIDEO_MODEL— defaultveo-2.0-generate-001(setveo-3.0-fast-generate-001for audio)COWORK_MEDIA_DIR— where files save, default~/cowork-media
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.