1. Conduid
  2. Content
  3. Runware
MCP server · Content

Runware

by · github.com/Runware/MCP-Runware

Model Context Protocol (MCP) server that provides lightning fast image and video generation tools using the Runware API.

Unclaimed last commit a year ago content
InstallView source
34Low

Scored 4 days ago · breakdown

About Runware

Runware is an MCP server published by Runware in the Content category: model Context Protocol (MCP) server that provides lightning fast image and video generation tools using the Runware API.

The repository has 9 stars and 9 forks, with the last commit a year ago. Six months or more without a commit doesn't mean the server is broken, but check the open issues before depending on it in production.

Install

From source
git clone https://github.com/Runware/MCP-Runware

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Ask AI

✦

Ask AI about Runware

Powered by Claude · Grounded in docs

I know everything about Runware. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

Static checks derived from the repository 10 hours ago. Heuristics on the source, not an audit.

  • ✓README presentThe repository ships documentation.
  • !License declaredNo license file. Legally you have no right to use it in a product.
  • !Tests presentNo tests found. Behaviour changes may go unnoticed between versions.
  • ✓Dependencies pinnedA lockfile pins dependency versions.
  • ✓No dynamic code executionSource scan found no eval, dynamic Function, or shell execution.

1 check has not been run yet.

README

Runware MCP Server

A powerful Model Context Protocol (MCP) server that provides lightning fast image and video generation tools using the Runware API. This server supports both SSE (Server-Sent Events) transport for custom claude connector and direct claude desktop installation as well.

Features

Image Generation Tools

  • imageInference: Full-featured image generation with advanced parameters
  • photoMaker: Subject personalization with PhotoMaker technology
  • imageUpscale: High-quality image resolution enhancement
  • imageBackgroundRemoval: Background removal with multiple AI models
  • imageCaption: AI-powered image description generation
  • imageMasking: Automatic mask generation for faces, hands, and people

Video Generation Tools

  • videoInference: Text-to-video and image-to-video generation
  • listVideoModels: Discover available video models
  • getVideoModelInfo: Get detailed model specifications

Utility Tools

  • imageUpload: Upload local images to get Runware UUIDs
  • modelSearch: Search and discover AI models on the platform

Smart Features

  • Automatic Model Selection: I2V uses klingai:5@2, T2V uses google:3@1
  • Input Validation: Prevents Claude upload URL pasting and validates dimensions
  • Comprehensive Error Handling: Clear error messages and guidance

Demo

Watch the demo video to see the Runware MCP server in action:

https://github.com/user-attachments/assets/9732096b-8513-455c-9759-cc88363c42f9

Architecture

[ MCP Client / AI Assistant ]
           |
    (connects via SSE over HTTP)
           |
    [ Uvicorn Server ]
           |
    [ Starlette App ]
           |
    [ FastMCP Server ]
           |
    [ Runware API ]

Prerequisites

  • Python: 3.10 or higher
  • Runware API Key: Get your API key from Runware Dashboard
  • Dependencies: See requirements.txt or pyproject.toml

Installation

1. Clone the Repository

git clone https://github.com/Runware/MCP-Runware.git
cd MCP-Runware

2. Install Dependencies

# Using uv (recommended)
uv venv
source .venv/bin/activate
uv pip install .

# Or using pip
pip install -r requirements.txt

3. Environment Setup

Create a .env file in the project root:

RUNWARE_API_KEY=your_api_key_here

Deployment Methods

Method 1: SSE Server (Recommended for Production)

Docker Deployment

# Build the Docker image
docker build -t runware_mcp_sse .

# Run the container
docker run --rm -p 8081:8081 runware_mcp_sse

Method 2: MCP Install (Direct Integration)

Install in Claude Desktop

# From the project directory
mcp install --with-editable . runware_mcp_server.py

Model Recommendations

Image Generation

  • Default: civitai:943001@1055701 (SDXL-based)
  • PhotoMaker: civitai:139562@344487 (RealVisXL V4.0)
  • Background Removal: runware:109@1 (RemBG 1.4)

Video Generation

  • Image-to-Video (I2V): klingai:5@2 (1920x1080)
  • Text-to-Video (T2V): google:3@1 (1280x720)

You can find all additional models here: Runware Models

Configuration

Environment Variables

  • RUNWARE_API_KEY: Your Runware API key (required)

Input Validation

  • Rejects Claude upload URLs (https://files.*). Claude tends to include base64 strings in its reasoning/thinking process, which rapidly fills the context window with garbage data. Learn more about this issue
  • Supports local file paths, public accessible URLs (make sure it has proper file extension such as JPG, PNG, WEBP, etc), and Runware UUIDs

Support

README fetched from the repository 10 hours ago. The original is authoritative.

Questions

About Runware

How do I install Runware?

No package has been published for this server, so install it from source: clone the repository and follow its README.

Is Runware safe to use with an AI agent?

Its trust score is 34 out of 100 (low). It passes 3 of the 5 static security checks run so far; the failures are listed above. It has no ConduID identity yet, so agent calls to it are not receipted.

Is Runware still maintained?

The last commit was a year ago and 2 open issues. That's long enough that you should check whether the maintainer is responding to issues before depending on it.