VE Text To Video MCP

Async MCP server for text-to-video generation via ve-backend proxy.

Environment

• API_URL: backend proxy url, default https://api.visionengine-tech.com/api/v1/text2video
• API_KEY: user API key from VisionEngine backend (required for submit/query and remote upload)
• MODEL: platform model id, default @preset/vec-1-0-text2video
• WORKDIR: local workspace root
• DEFAULT_OUTPUT_DIR: default relative output directory, default public/videos
• FILE_MODE: local file handling mode, local or remote, default remote
• REMOTION_WORK_DIR: shared mount root used in local mode, default /vec
• BASE_URL: backend public base url used for /save and /shared links, default https://api.visionengine-tech.com
• REMOTE_AUDIO_UPLOAD_PATH: remote upload path for local audios in remote mode, default public/audios

Tools

• submit
• query

submit

Submit an async text-to-video task and receive a taskId for later polling.

audioPath supports:

• public audio URL
• local audio file path

When local audio files are used:

• FILE_MODE=local: convert local absolute path to backend shared download URL by removing REMOTION_WORK_DIR prefix
• FILE_MODE=remote (default): upload local file to backend /save, then convert returned file path to /shared/...?...download=true

Recommended upstream constraints:

• prompt length: <= 1500 characters
• negative prompt length: <= 500 characters
• audio formats: mp3/wav preferred
• audio size: <= 15MB
• duration parameter: 2~15 seconds
• result URL may expire within 24 hours

query

Query a submitted task by taskId.

• If still running, the tool returns current taskStatus and suggests querying again after about 15 seconds.
• If succeeded, it returns outputVideoUrl and by default downloads the result locally.
• Set downloadResult=false if you only want the current status or the remote result URL.

Note: upstream result URLs may expire within 24 hours, so prompt download is recommended.