> ## Documentation Index
> Fetch the complete documentation index at: https://docs.akapulu.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Update take video instructions

> Replace face_prompt and upper_body_prompt on every chunk. Window count and timing stay locked. Prompt edits are free. Fails after video has started.

Replace `face_prompt` and `upper_body_prompt` on every chunk.

Each object needs `index`, `upper_body_prompt`, and `face_prompt`. Send one object per stored chunk; indexes and timing stay locked. Prompt edits are free and do not regenerate audio.

Fails after video has started.

What to put in those strings is in [Workflow](/guides/clips/workflow#video-instructions).


## OpenAPI

````yaml openapi/akapulu.json POST /clips/{clip_id}/takes/{take_id}/prompts
openapi: 3.1.0
info:
  title: Akapulu Labs API
  version: 1.0.0
  description: Public Akapulu Labs API reference.
servers:
  - url: https://akapulu.com/api
security:
  - bearerAuth: []
paths:
  /clips/{clip_id}/takes/{take_id}/prompts:
    post:
      summary: Update take video instructions
      description: >-
        Replace face_prompt and upper_body_prompt on every chunk. Window count
        and timing stay locked. Prompt edits are free. Fails after video has
        started.
      parameters:
        - name: clip_id
          in: path
          required: true
          description: Clip ID.
          schema:
            type: string
            format: uuid
        - name: take_id
          in: path
          required: true
          description: Take ID.
          schema:
            type: string
            format: uuid
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/UpdateTakePromptsRequest'
            examples:
              default:
                value:
                  chunk_prompts:
                    - index: 0
                      face_prompt: >-
                        A lady is talking with animated facial expressions,
                        frequently raising and arching the eyebrows for
                        emphasis. Only the foreground is moving, the background
                        remains static.
                      upper_body_prompt: >-
                        A lady is talking to the camera with natural movements
                        while speaking. The torso and shoulders stay steady.
                        Only the foreground person is moving, the background
                        remains static.
      responses:
        '200':
          description: Video instructions updated
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/UpdateTakePromptsResponse'
              examples:
                success:
                  value:
                    id: 22222222-3333-4444-5555-666666666666
                    chunk_prompts:
                      - index: 0
                        start: 0
                        end: 2.92
                        text: Welcome to Acme.
                        face_prompt: >-
                          A lady is talking with animated facial expressions,
                          frequently raising and arching the eyebrows for
                          emphasis. Only the foreground is moving, the
                          background remains static.
                        upper_body_prompt: >-
                          A lady is talking to the camera with natural movements
                          while speaking. The torso and shoulders stay steady.
                          Only the foreground person is moving, the background
                          remains static.
        '400':
          description: Invalid request
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ApiError'
              examples:
                default:
                  value:
                    error: chunk_prompts must be a non-empty list
        '401':
          description: Missing or invalid API key
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ApiError'
              examples:
                default:
                  value:
                    error: Invalid API key
        '404':
          description: Take not found
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ApiError'
              examples:
                default:
                  value:
                    error: Take not found
        '409':
          description: Take cannot be edited
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ApiError'
              examples:
                default:
                  value:
                    error: Take video has already started
components:
  schemas:
    UpdateTakePromptsRequest:
      type: object
      required:
        - chunk_prompts
      properties:
        chunk_prompts:
          type: array
          items:
            type: object
            required:
              - index
              - face_prompt
              - upper_body_prompt
            properties:
              index:
                type: integer
              face_prompt:
                type: string
              upper_body_prompt:
                type: string
    UpdateTakePromptsResponse:
      type: object
      required:
        - id
        - chunk_prompts
      properties:
        id:
          type: string
          format: uuid
        chunk_prompts:
          type: array
          items:
            $ref: '#/components/schemas/ChunkPrompt'
    ApiError:
      type: object
      required:
        - error
      properties:
        error:
          type: string
          description: Human-readable error message.
    ChunkPrompt:
      type: object
      properties:
        index:
          type: integer
        start:
          type: number
          description: Chunk start in seconds.
        end:
          type: number
          description: Chunk end in seconds.
        text:
          type: string
          description: Spoken words in this chunk.
        face_prompt:
          type: string
        upper_body_prompt:
          type: string
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer

````