All products

Track AI video generation jobs

Route approved shot requests to an authorised provider and keep job IDs with their outputs. Check which shots were actually produced before handing them to an editor.

Track AI video generation jobs when a shoot contains several similar requests and a delayed reply could lead to a duplicate submission. The editor needs to know which shot produced a take, which settings were approved and whether an actual file was returned. A provider acknowledging a request is only one step in that trail.

The execute-media-generation workflow checks a compiled generation package and the provider route for each request. A dry run can check completeness without spending generation credits. Real execution depends on an authorised adapter, available credentials and a provider capable of the requested mode.

What it does

  • Check that an approved request has an execution route
  • Distinguish a simulated run from a submitted job
  • Retain the provider job ID with the associated shot
  • Record failures or blocked requests separately
  • Reconcile returned media with the approved reference and request

Install and use

  1. Use a compatible agent and the maker’s full shared schemas. For real output, supply compiled requests, an authorised provider adapter and an agreed budget. The skill does not include provider credentials.
  2. Ask, “Review these two pottery shots in dry-run mode. Once the route is approved, submit only the accepted request and return its job record with the output.”
  3. Check the actual media before sending it to the editor. Investigate a pending job by its identifier before deciding whether another submission is necessary.

FAQ

Source

Made by zhangzhangco. Original skill: execute-media-generation, published on GitHub under the MIT licence.

View on GitHub