See llms.txt for all machine-readable content.
This workflow takes ad copy, sends it to Spendict’s assess endpoint to score and classify it, and then routes the result based on the returned launch recommendation so only ads that pass the gate proceed to your next step.
/api/v1/assess endpoint to receive scores and a launch_recommendation verdict.run to the “ship it” path and fix_first to the “hold” path, with all other values sent to a fallback output.Authorization: Bearer <your_key> for the Spendict request.run, fix_first, and fallback outputs to decide how to ship, revise, or regenerate ads based on the verdict.