mint:generate:start
Start video generation
start_video_generation
Generate a playable clip from a prompt or owned media references.
Parameters
Input schema generated from the registry. Required parameters are tagged.
- promptrequired
string
- video_mode
"text_to_video" | "image_to_video" | "first_last_frame" | "reference_to_video"
- generation_preset
"fast" | "standard" | "production"
- aspect_ratio
"16:9" | "9:16"
Image and first/last-frame modes follow the start image in every preset. Its landscape/portrait orientation must match this setting; reframe the image before submitting to change orientation. Exact output dimensions are reported on completion.
- generate_audio
boolean
- references
array
- duration_seconds
integer
- display_name_hint
string
- chat_id
string
- project_id
string
Destination Project ID. Required when chat_id is absent. When both are supplied, the chat must belong to this Project.
- contextrequired
string
Describe the user's goal and why this tool helps in 15-25 words. Do not include credentials, personal data, identifiers, URLs, emails, or account addresses.
Example request
{
"project_id": "prjworkspace123",
"prompt": "A slow camera move through a sunlit courtyard",
"video_mode": "text_to_video",
"generation_preset": "standard",
"duration_seconds": 8,
"aspect_ratio": "16:9",
"generate_audio": true
}Example response
{
"status": "queued",
"generationStarted": true,
"assetType": "video",
"assetId": "vidcourtyard123",
"chatId": "j572z3a4b5c6d7e8f9g0h1i2j3",
"chatUrl": "https://mint.gg/chat/j572z3a4b5c6d7e8f9g0h1i2j3",
"chatCreated": true,
"pollIntervalMs": 30000,
"videoMode": "text_to_video",
"durationSeconds": 8,
"aspectRatio": "16:9",
"generateAudio": true,
"approvalRequiredForFinalGeneration": false,
"summary": "Video generation started.",
"nextSteps": [
{
"tool": "wait_for_status",
"action": "wait_for_final",
"description": "Poll until Mint reports final output eligibility before trying to use files in a code project.",
"input": {
"asset_type": "video",
"asset_id": "vidcourtyard123",
"until_stage": "final"
}
}
]
}Use cases
- Create a cinematic clip from text.
- Animate an uploaded image or guide a new clip with existing media references.