Multi-scene composition
start_compositionEstimate, start or retry a multi-scene video composition. Estimate first: action "estimate" returns the credit cost and spends nothing. "start" begins the composition, which needs a prompt. "retry" re-runs a composition by compositionId. Spends Galatic.AI credits. Returns a job id; poll get_job_status with kind "composition".
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"action": "<action>",
"context": "<context>",
"llm_model": "<llm_model>"
},
"name": "start_composition"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| model | No | Video model id for the scenes, from list_models. | |
| action | Yes | estimate: cost only, spends nothing. start: begin a composition. retry: re-run compositionId. | |
| prompt | No | What the composition is about. Required for start. | |
| context | Yes | Explain in 15-25 words, in third person, why this tool is called and how it supports the user's goal. For analytics only. You MUST describe only the abstract purpose of the tool call. NEVER include, repeat, paraphrase, or infer personal, sensitive, or identifying information from the user request or tool results, including names, emails, phone numbers, IPs, IDs, or credentials. You MUST generalize specific entities into roles such as "a user", "the customer", or "an account". Example: "Retrieving a customer's recent orders to investigate a billing issue and help support determine the appropriate resolution." | |
| mixMode | No | How narration and music are mixed. | |
| captions | No | Caption settings, passed through to the composition. | |
| costTier | No | Cost tier for the scenes. It sets the price and the maximum duration. | |
| llm_model | Yes | The exact model identifier you (the assistant) are running as, taken from your system prompt or environment (e.g. "claude-opus-4-8", "gpt-5.2"). Used for analytics only. If you do not know your model identifier with certainty, pass "unknown" — never guess. | |
| musicConfig | No | Background music settings for the composition. | |
| videoConfig | No | Scene video settings such as duration, costTier and extendMode ("multi-scene" or "extend"). | |
| voiceConfig | No | Narration voice. provider is one of polly, polly-generative, elevenlabs-clone or skip. | |
| compositionId | No | Required for retry: the id of the composition to re-run. | |
| galleryAssets | No | Library items to use as scene material, passed through as the route expects. | |
| narrationText | No | Narration script, read in the voice set by voiceConfig. | |
| compositionType | No | What kind of piece this is. It decides how the result is filed in the library. | |
| conversation_id | No | Pass the exact conversation_id from the server's previous response, unchanged. The server provides it on the first call — never invent one, and do not issue parallel tool calls until you have it. Keep passing the same conversation_id for the rest of the conversation, including after later user messages or on a different task; do not reset it when the user starts a new request. | |
| targetDurationSec | No | Target length of the whole composition, in seconds. |