Claude Code video generation with the Shotmatic MCP server

Ask Claude Code for a video. Approve it before it renders.

Claude Code reads your repository, so it can turn a changelog, a README or a script you keep in git into a narrated short. Shotmatic does the rendering in your account, against your credit balance.

Create an API keySee all 23 tools
Install

Connect Claude Code to Shotmatic

  1. 01

    Create an API key on your Shotmatic account page, in the Developer section. The key is shown once; give it only the scopes the prompts below need.

  2. 02

    Add the server to .mcp.json in your project (or your user config) and replace <SHOTMATIC_API_KEY> with the key. Keep the file out of git when it holds a real key.

  3. 03

    Restart Claude Code in that folder and approve the shotmatic server when it asks.

  4. 04

    Run /mcp in Claude Code: shotmatic should be connected and list its tools.

.mcp.jsonClaude Code Remote MCP
{
  "mcpServers": {
    "shotmatic": {
      "type": "http",
      "url": "https://shotmatic.com/mcp",
      "headers": {
        "Authorization": "Bearer <SHOTMATIC_API_KEY>"
      }
    }
  }
}

Add this server to Claude Code’s .mcp.json using a Shotmatic API key.

After connecting, call get_account and read shotmatic://capabilities. Use plan_video to choose a creation path. Add write scopes only for the actions you need. Claude web connectors that require OAuth are not available; this setup is for Claude Code.

Scopes the prompts below use: avatar_profiles:read, clip_tasks:read, clip_tasks:write, media:read, sessions:read, sessions:write, videos:read, voices:read.

Prompts

Three prompts and the calls they lead to

Shotmatic's server instructions steer the agent through these calls. It may add a read, such as the capabilities catalog, or check a status more than once.

01A short from your changelog

Prompt

Read the newest entry in CHANGELOG.md and turn it into a 30-second vertical short for YouTube Shorts. Write the narration from the entry, show me the credit estimate, and wait for my go-ahead before rendering.

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)Returns the recommended format, a starter session payload and the credit estimate.
  3. 3create_sessionCreates the editable draft from the starter payload.
  4. 4get_session_previewShows the scenes and narration for your review.
  5. 5approve_and_renderAfter you say go: queues the render.
  6. 6get_render_statusFollows the render to completion.
  7. 7get_videoReturns the finished MP4 link.
You get
A 30-second 9:16 MP4 in your Shotmatic library.
API key scopes
sessions:read, sessions:write, videos:read, voices:read

02Clips from a recording you uploaded

Prompt

Find the webinar recording I uploaded to Shotmatic today and cut it into three vertical clips: the setup demo, the pricing question, and the closing tip. Quote the clips before rendering them.

Calls it leads to

  1. 1list_media(type: "video", sort: "recent")Finds the recording and its media id.
  2. 2create_clip_task(source_type: "media", number_of_sections: 3, aspect_ratio: "9:16")Starts the analysis of that library video; this charges the source analysis.
  3. 3wait_for_clip_taskWaits up to 25 seconds per call until the suggested sections are ready.
  4. 4quote_clip_sectionsPrices the chosen sections; nothing is reserved yet.
  5. 5render_clip_sectionsAfter you approve the quote: renders the clips.
  6. 6get_videoReturns each finished clip.
You get
Three 9:16 clips, each a separate video in your library.
API key scopes
clip_tasks:read, clip_tasks:write, media:read, videos:read

03An avatar explainer from your README

Prompt

Write a 30-second presenter script from the first section of README.md, then make an avatar explainer with it. Use one of my avatar profiles and that profile's default voice, and show me the quote before rendering.

Calls it leads to

  1. 1read shotmatic://avatar-profilesLists your presenter profiles and their default voices.
  2. 2plan_videoWith script, avatar_profile_id and voice_id it returns a full-presenter session payload.
  3. 3create_sessionCreates the Avatar draft.
  4. 4get_session_previewShows the draft for review.
  5. 5quote_avatar_sessionReturns the final price for this revision and the approval arguments.
  6. 6approve_and_renderAfter you accept the quote: renders with the quote's approval arguments.
  7. 7get_render_statusFollows the render.
You get
A presenter-led explainer, priced before it rendered.
API key scopes
avatar_profiles:read, sessions:read, sessions:write
Real output

A video made through these tools

Every client reaches the same server, so an output looks the same whichever agent asked for it.

Paper Moon Assembly

Made on April 3, 2026 with Shotmatic's MCP session tools, on a local copy of the hosted server using the production render pipeline and an internal test account.

Script

A city of paper astronomers gathers every night to assemble a moon from glowing fragments. Crane-sized kites lift silver crescents into place. Children in raincoats deliver the final shard. When the moon clicks together, the entire city glows as if it has remembered its own sky.

Calls the run made

  1. create_sessionCreated a draft from the script with a 9:16 render.
  2. get_session_previewWaited for the preview before changing anything.
  3. update_sessionCut the caption density, keeping the vertical framing.
  4. approve_session (today: approve_and_render)Approved the revised draft into the final render.
Format
Faceless short
Aspect ratio
9:16
Length
18 s
Visuals
AI Video
Image quality
Ultra
Visual style
Hand-painted Anime
Captions
Viral
Narration
AI voice
Soundtrack
Yes

At these settings a video like this is 301 credits today. This run predates the credit ledger, so its own charge is not on record.

Next
Cursor guideCodex guideRecipes: step-by-step agent tasks