Shotstack vs Templated API + MCP

Shotstack

A public OpenAPI definition, Markdown docs, a hosted MCP server with OAuth and a free watermarked sandbox let an agent draft, check and render video with little setup. API keys have no scopes, webhooks are unsigned, no security contact or certification of Shotstack's own was found, and the published Node and Python SDKs date from July 2024.

Templated API + MCP

MIT-licensed MCP server with 25 tools, every one annotated with readOnlyHint and destructiveHint. One API key with full account access, and the MCP docs offer it as ?apiKey= in the URL.

Shotstack

1. Fetch https://shotstack.io/docs/api/api.edit.json and the conventions guide before writing an edit. Track order is reversed, so the first track renders on top
2. Render in stage first. It is free and watermarked, but AI generation there is charged from the production credit balance
3. Call POST /generate/quote before POST /generate, and send an Idempotency-Key header on generation. POST /render has no idempotency key, so do not resubmit blindly
4. Do not store the url from the render status response. It expires after 24 hours. Use the CDN address built from owner and the render id
5. On a 429, wait for the 60-second window to reset and retry with backoff. Limits are per API key, 300 a minute on the Edit API in production

Templated API + MCP

1. Call get_template_layers before create_render so the keys in layers match the template
2. Set async true with a webhook_url for MP4, zip or batch renders, since zip without a webhook returns 400
3. Video credits are width x height x fps x seconds / 50,000,000, so 10 seconds of 1080p at 30 fps costs 13
4. Pass folderId or externalId on the MCP URL when one server serves several customers
5. Back off on your own schedule after a 429, since the docs give no retry guidance