Enables AI agents to generate and manipulate image, video, audio, and 3D content using the Runware API.
Also known as Runware/mcp, @runware/mcp
- Stars
- 0
- Tools
- 12
- Downloads, 30 days
- —
- Latest version
- 1.3.2
- Last commit
- Nov 13, 2026
- License
- MIT
What is Runware MCP?
Runware enables AI agents to generate and manipulate image, video, audio, and 3D content using the Runware API. It exposes 12 tools, such as run, model_schema, list_models.
Summary generated from the project's README
About Runware MCP server
Generating and manipulating multimedia content through AI agents is possible with this server. It connects clients like Claude Desktop, Cursor, and Codex to the Runware API, allowing them to handle image, video, audio, and 3D generation tasks. Users can perform complex operations such as upscaling, background removal, and captioning by simply describing their intent to the agent.
Setup requires a Runware API key provided via the RUNWARE_API_KEY environment variable. Developers can run the server using npx for immediate use or install it globally via npm. The agent manages the technical details by selecting appropriate models and validating parameters against specific schemas before execution.
Beyond simple generation, the server provides tools to manage media storage, search for community models like Civitai fine-tunes, and check account balances. It offers deep visibility into the model catalog through capabilities like listing model pricing, viewing metadata, and retrieving sample input and output pairs for specific configurations.
What you can do with it
- 1Create multimedia assetsAsk an AI agent to produce a specific visual or auditory file from a text prompt.
- 2Enhance existing mediaInstruct your agent to increase the resolution of a specific image file.
- 3Research model specificationsQuery the agent to find out which models are available for a specific type of media generation.
- 4Manage account resourcesCheck how much credit remains in your Runware account.
- 5Discover community modelsSearch for specialized fine-tuned models within the community catalog.
Written from the project's README. Every use case is checked against a quote from it.
How to install Runware MCP
Formats verified 10 Nov 2026Add to CursorAdd to GooseInstall Runware MCP in Claude Code
$ claude mcp add --env 'RUNWARE_API_KEY=<YOUR_RUNWARE_API_KEY>' mcp -- npx -y @runware/mcpInstall Runware MCP in Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@runware/mcp"
],
"env": {
"RUNWARE_API_KEY": "<YOUR_RUNWARE_API_KEY>"
}
}
}
}Install Runware MCP in VS Code
.vscode/mcp.json
{
"servers": {
"mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@runware/mcp"
],
"env": {
"RUNWARE_API_KEY": "<YOUR_RUNWARE_API_KEY>"
}
}
}
}Install Runware MCP in VS Code CLI
$ code --add-mcp '{"name":"mcp","type":"stdio","command":"npx","args":["-y","@runware/mcp"],"env":{"RUNWARE_API_KEY":"<YOUR_RUNWARE_API_KEY>"}}'Install Runware MCP in Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@runware/mcp"
],
"env": {
"RUNWARE_API_KEY": "<YOUR_RUNWARE_API_KEY>"
}
}
}
}Install Runware MCP in Codex
$ codex mcp add mcp --env 'RUNWARE_API_KEY=<YOUR_RUNWARE_API_KEY>' -- npx -y @runware/mcpInstall Runware MCP in Codex config.toml
~/.codex/config.toml
[mcp_servers.mcp] command = "npx" args = ["-y", "@runware/mcp"] [mcp_servers.mcp.env] RUNWARE_API_KEY = "<YOUR_RUNWARE_API_KEY>"
Install Runware MCP in Gemini CLI
$ gemini mcp add -s user -e 'RUNWARE_API_KEY=<YOUR_RUNWARE_API_KEY>' mcp npx -- -y @runware/mcpInstall Runware MCP in Gemini settings.json
~/.gemini/settings.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@runware/mcp"
],
"env": {
"RUNWARE_API_KEY": "<YOUR_RUNWARE_API_KEY>"
}
}
}
}Runware MCP tools 12
Detected from the READMErunmodel_schemalist_modelsmodel_detailsmodel_examplesmodel_pricinglist_capabilitiesmodel_searchmedia_storagemodel_uploadaccountget_task_details
| run | — | |
| model_schema | — | |
| list_models | — | |
| model_details | — | |
| model_examples | — | |
| model_pricing | — | |
| list_capabilities | — | |
| model_search | — | |
| media_storage | — | |
| model_upload | — | |
| account | — | |
| get_task_details | — |
Configuration
RUNWARE_API_KEYRequiredSecretYour Runware API key. Create one at https://runware.ai.
| Name | Description | |
|---|---|---|
| RUNWARE_API_KEY | Your Runware API key. Create one at https://runware.ai. | RequiredSecret |
Automated checks
Ran Nov 13, 2026Full report →Passed our automated checks. That is not a security audit: read the code and give the server the narrowest access it needs. How checks work
- Source repository linked and reachablePass
- Open source license (OSI)Pass
- Commit or release in the last yearPass
- Package name not close to a popular packagePass
- Remote endpoint uses HTTPSNot checked
- Destructive tools are labelledPass
- No hidden instructions to the modelPass
- No critical advisory in the package or its dependenciesPass
README
View on GitHubRunware MCP
MCP server that gives AI agents (Claude, Cursor, Codex, etc.) access to the full Runware API — image generation, video generation, audio generation, 3D, upscaling, background removal, captioning, and more.
Install
Pick one — npx is simplest, global install is faster on repeat use.
# Use directly with npx (no install)
npx -y @runware/mcp
# Or install globally
npm install -g @runware/mcp
Connect
Get an API key at runware.ai. Then point your agent at the MCP.
Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"runware": {
"command": "npx",
"args": ["-y", "@runware/mcp"],
"env": {
"RUNWARE_API_KEY": "your-api-key"
}
}
}
}
If you installed globally:
{
"mcpServers": {
"runware": {
"command": "runware-mcp",
"env": {
"RUNWARE_API_KEY": "your-api-key"
}
}
}
}
Cursor
Settings → MCP → Add server — same config as above.
Codex
Add to your MCP configuration with the same command and env.
How to use it
Once connected, just talk to your agent. It picks the right tool, picks the right model, and fills in the parameters.
- "Generate an image of a cat in a forest"
- "What models do you have for video generation?"
- "Upscale https://example.com/photo.jpg by 4x"
- "Show me my account balance"
- "Find me a Civitai LoRA for anime style"
The agent figures out which model fits, what parameters to pass, and how to interpret the response. Parameters are validated against each model's schema before submission, so the agent gets fast feedback when it picks wrong values.
Tools the agent has access to
You don't call these directly — the agent does, based on what you ask for.
run— execute any Runware inference task (image, video, audio, 3D, upscaling, captioning, etc.) on a given modelmodel_schema— fetch the parameter schema for a specific modellist_models— list Runware's official, curated model integrations (supportscapability,category,creator,searchfilters)model_details— get full metadata for a curated model by AIRmodel_examples— get sample input/output pairs for a curated modelmodel_pricing— get pricing overview + per-configuration examples for a curated modellist_capabilities— list every model capability (e.g.io:text-to-image,op:upscale) with labelsmodel_search— search the community model catalog (Civitai fine-tunes, custom uploads)media_storage— store media (images, video, audio, 3D models) for use as input, or delete stored mediamodel_upload— upload a custom modelaccount— retrieve account information including balance and usageget_task_details— retrieve the original request and response for a previous task
Environment variables
| Variable | Required | Description |
|---|---|---|
RUNWARE_API_KEY |
Yes | Your Runware API key |
Development
# Run directly (no build step)
npm run dev
# Build
npm run build
# Test
npm test
# Lint
npm run lint
# Smoke-test the stdio protocol
echo '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"test","version":"1.0.0"}}}
{"jsonrpc":"2.0","method":"notifications/initialized"}
{"jsonrpc":"2.0","id":2,"method":"tools/list"}' | RUNWARE_API_KEY=your-key node dist/index.js
# Or use the MCP Inspector for a UI
npx @modelcontextprotocol/inspector node dist/index.js
Runware MCP questions
How do I add Runware MCP to Claude Code?
Run claude mcp add --env 'RUNWARE_API_KEY=<YOUR_RUNWARE_API_KEY>' mcp -- npx -y @runware/mcp in your terminal. Add --scope user to make it available in every project.
How do I add Runware MCP to Cursor?
Open ~/.cursor/mcp.json and add the server under mcpServers, using the Cursor config shown in the install section above.
What tools does Runware MCP provide?
It exposes 12 tools, including run, model_schema, list_models, model_details, model_examples.
How do I connect the server to Claude Desktop?
Add the server configuration to your claude_desktop_config.json file, specifying the npx command and your RUNWARE_API_KEY in the environment variables.
Can I search for specific styles or models?
Yes, you can search the community model catalog for things like Civitai fine-tunes and custom uploads.
How do I check my account balance?
The agent can retrieve your account information, including balance and usage, through the account tool.