create_character
Create a reusable character for the user's videos and save it to their Digital Maker account, from a written description. (A character is placed INTO the scene pictures. For a talking on-screen host, an AI presenter or avatar, use draw_presenter instead.) Every character is made the same way: a FULL BODY figure, head to feet, standing on a plain white background, so it can be dropped into any scene. Once saved, pass its name to create_video and that same character appears in every scene. It also lands in the user's image library. Users keep up to 10.
IF THE USER SHARES A REFERENCE IMAGE (a screenshot of a channel they like, a style they want): look at it yourself and write your own description of the KIND of character it is, then pass that. Describe the archetype, art style, colour palette, build, hair, clothing and general vibe. Ignore the framing and background of what they shared, since the result is always a full-body figure on white regardless. Do NOT try to reproduce a specific individual's face, and do not describe a real identifiable person. The point is a character of a similar style that belongs to this user, not a copy of someone else's.
Costs one image from the user's monthly allowance. Show them the result and offer to redo it before they build a video around it. Available on any plan that includes images, which is Pro and above, even though faceless video itself needs Premium: a Pro user can build their character library now and use it after upgrading.
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": {
"name": "<name>",
"description": "<description>"
},
"name": "create_character"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| name | Yes | Short name the user will refer to the character by, e.g. Maya. Must be unique among their characters. | |
| style | No | The visual world this character belongs to. Use the SAME value you will later pass as style_preset on create_video, so the character matches its own video. When the user shares a reference image, pick the preset whose look matches it: flat 2D illustration is animated_cartoon; dark moody photoreal lit by a single rim light is motivational_dark; gold, marble and wealth imagery is finance_luxury; candlelit period detail is history_ancient; cosmic and nebula lighting is space_scifi; soft neutral tones with lots of negative space is minimalist_clean; ordinary present-day photography is cinematic_documentary. Defaults to cinematic_documentary. | |
| description | Yes | Full visual description: age range, build, hair, clothing, colour palette, art style, expression, overall vibe. The richer this is, the more consistent the character stays across scenes. Avoid naming real people. |