Make 10 faceless shorts at once with an AI agent

Ten faceless shorts from a topic list

One prompt, ten topics: the agent plans and drafts a short per topic, shows you the total, then renders them in one pass.

Credits: 1000 — 100 credits per 30-second short with AI images at standard quality, times ten.

The task

Ten shorts from ten topics

Prompt

Make ten 30-second faceless shorts for YouTube Shorts, one for each topic in topics.md. Use the same voice and visual style for all of them, tell me the total credits, and render them after I confirm.

Calls it leads to

  1. 1read shotmatic://voicesPick a ready account voice; its id becomes voice_id.
  2. 2plan_video(platform: "youtube_shorts", duration_seconds: 30)Once per topic: a starter payload and an estimate each.
  3. 3create_sessionOne draft per topic.
  4. 4approve_and_renderOne render per draft, after you confirm the total.
  5. 5get_render_statusFollows each render.
  6. 6list_videosCollects the ten finished videos.
You get
Ten 9:16 shorts in your library.
API key scopes
sessions:read, sessions:write, videos:read, voices:read
Limits
  • create_session allows 10 calls per 5 minutes per API key and approve_and_render 10 calls per 5 minutes, so a larger batch has to pause between windows.
  • Each short is its own video; the batch is not a sequence and does not repeat.
  • Read every draft before you confirm the total. In the recorded run below (three topics, not ten), the drafts came back at 40 to 59 words, about half of what 30 seconds of narration needs, and were rewritten before rendering.
Recorded run

Why the sky is blue

  • The Eiffel Tower was meant to be temporary

  • How the Fresnel lens changed lighthouses

Three faceless shorts from one topic list

Made on October 4, 2026 with Shotmatic's hosted MCP server (shotmatic.com/mcp) and an API key for Shotmatic's internal account. The list had three topics, not ten.

Prompt

Topics: why the Eiffel Tower was meant to be temporary; how the Fresnel lens changed lighthouses; why the sky is blue.

Calls the run made

  1. plan_videoOnce per topic: a Faceless Short payload and its estimate.
  2. create_sessionOne draft per topic, all with the same voice (Bella), Hand-painted Anime style and Story captions.
  3. get_session_previewThe drafts ran 40 to 59 words, too short for 30 seconds, and two ended on filler (a quote, a slogan).
  4. update_sessionReplaced each draft with a fact-checked script of 81 to 83 words.
  5. approve_and_renderOne render per draft, after the 300-credit total was confirmed.
  6. get_render_statusFollowed the three renders to completed.
  7. list_videosCollected the three finished videos.
Format
Faceless short
Aspect ratio
9:16
Length
33 s, 33 s, 33 s
Visuals
AI Images
Image quality
Standard
Visual style
Hand-painted Anime
Captions
Story
Narration
AI voice
Soundtrack
Yes

At these settings a video like this is 100 credits today. Each of its 3 sessions was quoted at 100 credits before approval (300 in all); Shotmatic's own account is not billed for renders, so nothing was charged.

Run it from
Claude CodeCursorCodexAny MCP client

More recipes

    Make 10 Faceless Shorts at Once with an AI Agent | Shotmatic