APITemplate.io vs Templated API + MCP

APITemplate.io

A public OpenAPI description covers 12 operations, with rate limits, four regional endpoints and a 99.9 per cent uptime SLA published. Each account or team has one unscoped API key, no changelog or deprecation notices were found, the SDK repositories date from September 2023, and two documentation pages name different regional hostnames.

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.

APITemplate.io

1. Send the key in the X-API-KEY header and pick the regional host first, because requests and output are processed and stored in that region
2. Call GET /v2/list-templates with with_layer_info=1 before POST /v2/create-image, so the name values in overrides match the template's elements
3. Set meta to your own reference on each generation and check GET /v2/list-objects before retrying, since there is no idempotency key
4. Set expiration in minutes (1 to 10080) or export_type=file for sensitive output. By default the download URL does not expire
5. For async=1 you must pass webhook_url. The callback is a GET with the result in the query string, retried three times
6. On HTTP 429 pause and retry on your own schedule. The docs give no Retry-After header or backoff interval

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