One POST in, a rendered video and its full editable scene graph out. Same price as the app. No markup, no minimum.
POST /v1/explainer
{
"input": "https://yourdocs.example.com/how-billing-works.md",
"length": "auto", // or "0:30" steps up to "6:00"
"aspect": "16:9", // 16:9, 9:16, 1:1
"voice": "sketchie:ana", // library voice, or elevenlabs:<id>, fish:<id>
"language": "en"
} Response: the MP4 URL, plus the scene graph.
{
"video_url": "...",
"scenes": [
{
"id": "scene-3",
"narration": "The invoice is generated on the first of the month...",
"elements": [ ... ],
"duration_s": 22.4
}
]
} Every other explainer-video API returns a file. Ours returns a document. Patch a scene and re-render just that scene:
PATCH /v1/explainer/{id}/scenes/scene-3
{ "narration": "Invoices are generated on the first business day of the month..." } Seconds later you have a new video with one scene changed and everything else untouched. Re-renders are metered per second of re-rendered output. Your revision loop stops costing full generations.
Sketchie ships an MCP server, so any MCP-capable agent can generate and edit explainer videos as native tools: generate from a doc, inspect the scene graph, patch a scene, re-render. If you are building agents that should answer with video instead of a wall of text, this is the shortest path that exists.
Pass elevenlabs:<voice_id> or fish:<voice_id>
and Sketchie narrates with your cloned or designed voice. Your brand
voice, your founder’s voice, your course instructor’s voice.
Library voices are included at no extra cost, including
audience-designed options like kid-explains-to-kids for young
learners.