# midjourney-mcp [Health: Active]

**Category:** 🎥 Multimedia Process  
**Repository:** https://github.com/runapi-ai/midjourney-mcp  
**GitHub Stars:** 2  
**npm Downloads (last month):** 635  
**Views:** 5  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/midjourney-mcp

## Description
RunAPI MCP server for Midjourney: create tasks, poll status, check pricing.

## Tools
Capabilities this server exposes over MCP:

- **login** — Authenticate RunAPI by opening a browser PKCE login flow and saving the API key to ~/.config/runapi/config.json.
- **edit_image** — Create a Midjourney task on RunAPI (edit image). Returns a task id, status, and output URLs.
- **extend_video** — Create a Midjourney task on RunAPI (extend video). Returns a task id, status, and output URLs.
- **image_to_video** — Create a Midjourney task on RunAPI (image to video). Returns a task id, status, and output URLs.
- **text_to_image** — Create a Midjourney task on RunAPI (text to image). Returns a task id, status, and output URLs.
- **get_seed** — Run a synchronous Midjourney operation on RunAPI (get seed). Returns the operation result.
- **image_to_prompt** — Run a synchronous Midjourney operation on RunAPI (image to prompt). Returns the operation result.
- **shorten_prompt** — Run a synchronous Midjourney operation on RunAPI (shorten prompt). Returns the operation result.
- **get_task** — Fetch the current status and latest result payload for a midjourney task.
- **check_pricing** — Look up RunAPI pricing for the midjourney model line.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "midjourney-mcp": {
    "command": "npx",
    "args": ["-y","@runapi.ai/midjourney-mcp"],
    "env": {
      "RUNAPI_API_KEY": ""
    }
  }
}
```

**Requires environment variables:** `RUNAPI_API_KEY` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What midjourney-mcp does

midjourney-mcp exposes RunAPI's Midjourney operations as MCP tools for compatible AI clients. It supports text-to-image generation, image editing, image-to-video conversion, and video extension. Creation tools return a task identifier, status information, and output URLs, with an option to wait for a terminal result.

The server also provides synchronous operations for extracting a seed, converting an image into a prompt, and shortening a prompt. Existing tasks can be inspected through `get_task`, while `check_pricing` retrieves pricing for a Midjourney model and endpoint. Pricing lookup is available without authentication.

The supported model identifiers include `midjourney-v8.1` for text-to-image, `midjourney-edit-image` for image editing, and `midjourney-image-to-video` for image-to-video and video extension workflows. Model availability may change, so pricing lookup or the RunAPI Midjourney model page should be used for the current catalog.

## How it works

The package runs as a stdio MCP server, allowing an MCP host to start it with `npx` and pass tool calls through the standard input and output streams. An agent can first inspect pricing, submit a task, and either wait for completion or retain the returned task ID for later polling with `get_task`.

Task-based operations are asynchronous at the RunAPI level. The creation tools can optionally wait for a terminal status; when they do not, the returned task ID becomes the handle for a later status request. Seed lookup, image-to-prompt, and prompt-shortening calls instead return their operation results directly.

## Setup and configuration

Install the package through an MCP client's stdio configuration. The command uses the package's npm distribution:

```json
{
  "mcpServers": {
    "midjourney": {
      "command": "npx",
      "args": ["-y", "@runapi.ai/midjourney-mcp"]
    }
  }
}
```

Authentication can be initiated with the `login` tool. It opens a browser-based PKCE login flow and stores the resulting API key at `~/.config/runapi/config.json`, which is also used by `runapi login`. For headless or CI environments, set `RUNAPI_API_KEY` before starting the MCP host. The server checks that environment variable first, then the saved configuration file. Without either credential, pricing lookup remains available, but authenticated operations cannot be used.

Do not place real API keys in source control. A pre-provisioned configuration file can contain an `apiKey` field under the RunAPI configuration path.

## Tools and capabilities

- `login`: authenticate RunAPI and save the API key locally.
- `text_to_image`: submit a Midjourney text-to-image task.
- `edit_image`: submit an image-editing task.
- `image_to_video`: create a video from an image.
- `extend_video`: continue an existing video workflow.
- `get_task`: retrieve the latest status and result payload for a task.
- `get_seed`: run a synchronous seed lookup.
- `image_to_prompt`: derive a prompt from an image.
- `shorten_prompt`: shorten a Midjourney prompt.
- `check_pricing`: inspect pricing without signing in.

## Limitations and notes

This package is scoped to the Midjourney model line on RunAPI rather than the full RunAPI model catalog. Several operations require RunAPI authentication, and generated task results depend on the asynchronous status returned by RunAPI. Model availability can change between releases, so clients should use `check_pricing` or the RunAPI model page when selecting a model or endpoint.

_Full upstream README: https://allmcps.com/mcp/midjourney-mcp/readme_