Access

Cre8vid MCP interface

AI video MCP server. Give agents tools, limits, and receipts.

Discover models, price the exact request, build the prompt, submit one controlled job, and return the result through a typed tool surface.

tools/callcreate_video
{
  "model": "kling-3",
  "mode": "image-to-video",
  "prompt": "The subject turns...",
  "duration_seconds": 5,
  "aspect_ratio": "9:16",
  "generate_audio": true,
  "reference_ids": ["asset_123"],
  "max_cost_usd": 0.25,
  "idempotency_key": "shot-07-v2"
}
Cost ceiling enforced before submission

Tool surface

Six tools. One job lifecycle.

Read first. Estimate second. Start the paid action only after the prompt, settings, and budget are explicit.

01

list_models

Read model names, prices, inputs, outputs, and control limits.

READ
02

estimate_video

Price the exact model, duration, resolution, audio, and references.

READ
03

build_prompt

Return model-specific prompt structure without starting a generation.

READ
04

create_video

Submit the approved prompt and settings under a hard cost ceiling.

PAID
05

get_job

Read queue state, progress, route, cost, errors, and completion data.

READ
06

get_result

Return final metadata and a resource link for the completed video.

READ
Agent rule

A natural-language budget is advice. A typed maximum-cost field is a limit.

MCP questions

The whole workflow, answered.

What is an AI video MCP server?

It exposes video-generation actions as typed tools that an MCP client can discover and call. The client sends structured arguments, the server runs the video workflow, and the tool returns structured job, cost, status, or output data.

Read the full answer →
01

Discover

Read available routes, controls, limits, and rates before choosing a model.

02

Approve

Lock the prompt, output settings, references, estimate, and cost ceiling.

03

Retrieve

Poll the job handle, then return final cost, metadata, and the video resource.