generate-image
Generates an image from a text prompt, with an illustration style, a colour and an aspect ratio
LABSCO SUMMARY
Video generation that admits it takes time, and hands you the polling pair instead of a call that hangs.
generate-video takes the asynchronous path by default and switches to a blocking one when the request says to wait, while submit-video-task and get-video-task make the same thing explicit — which is what an agent loop that cannot sit on an open call actually needs. The image side is narrower by comparison: text, style, colour and ratio go in, and nothing here edits what comes back.
What it is
A four-tool server over Jimeng AI's text-to-image and text-to-video models, with the video side offered both as one blocking call and as a submit-then-collect pair.
What you get
- An image generated from text, with an illustration style, a colour and an aspect ratio as arguments
- A video generated from a prompt, returning asynchronously by default and synchronously when the request asks to wait for the result
- A video task submitted on its own, returning its task ID immediately
- The finished video collected later by task_id
Requirements
JIMENG_ACCESS_KEY and JIMENG_SECRET_KEY — an access key pair for Jimeng AI, which is what every one of the four tools authenticates with.
Setup effort
One command plus a key — npm install -g jimeng-ai-mcp, then supply credentials
USE CASES
01
Getting an illustration for a post without leaving the editor
Describe the image, name a style, a colour and an aspect ratio, and get it back in the same turn you were writing in.
02
Producing a clip while you carry on with something else
Submit the video task and take the ID immediately, then come back for the result later — the pattern that keeps a long render from occupying the conversation.
03
Asking for a video and waiting for it in one turn
Use the single call and tell it to wait for the result, for the times when a task ID to chase later is more trouble than the render itself.
4 tools
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
generate-video
Generates a video from a prompt with Jimeng AI's text-to-video model, asynchronously by default and synchronously when the request asks to wait for the result
get-video-task
Returns the result of a video generation task, given its task_id
submit-video-task
Submits a video generation task and returns its task ID immediately, without waiting for the video
BEFORE YOU INSTALL
What to know first
Tools you get
4 tools when we connected. generate-video and the submit-video-task/get-video-task pair do the same job, one blocking and one submit-then-poll.
Requires
- npx on PATH
- JIMENG_ACCESS_KEY in your environment
- JIMENG_SECRET_KEY in your environment
Where it runs
All 3 documented setups run the server on your machine.
CONTEXT COST
Light Standing cost
Rated by the size of the 4 tool definitions that ride along on every turn, not by how many there are
INSTALLATION
Connect your client
Server setup
TERMINAL
npm install -g jimeng-ai-mcpMCP Clients
Cursor
CURSOR · JSON
{
"mcpServers": {
"jimeng": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"jimeng-ai-mcp"
],
"env": {
"JIMENG_ACCESS_KEY": "<YOUR_JIMENG_ACCESS_KEY>",
"JIMENG_SECRET_KEY": "<YOUR_JIMENG_SECRET_KEY>"
}
}
}
}Cursor — WSL
CURSOR — WSL · JSON
{
"mcpServers": {
"jimeng": {
"type": "stdio",
"command": "cmd",
"args": [
"/c",
"npx",
"-y",
"jimeng-ai-mcp"
],
"env": {
"JIMENG_ACCESS_KEY": "<YOUR_JIMENG_ACCESS_KEY>",
"JIMENG_SECRET_KEY": "<YOUR_JIMENG_SECRET_KEY>"
}
}
}
}The project gives this block for Cursor when the server runs under WSL: its README says the cmd /c prefix is needed there for the command to run correctly.
Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"jimeng": {
"command": "npx",
"args": [
"-y",
"jimeng-ai-mcp"
],
"env": {
"JIMENG_ACCESS_KEY": "<YOUR_JIMENG_ACCESS_KEY>",
"JIMENG_SECRET_KEY": "<YOUR_JIMENG_SECRET_KEY>"
}
}
}
}
