Interpret SendGrid Engagement Quality Responses in Codex
Distinguish a reported SendGrid Engagement Quality score from a pending result or a record that omits the required score fields.
中文Setup for Codex
Use a local fixture in Codex. Live scores are asynchronous, and account eligibility determines whether the API includes score and metrics data.
- Open the pinned Twilio AI repository revision and copy this skill directory, with its companion files, into your Codex project's .agents/skills directory.
- Open Codex in a scratch project and select the matching Twilio SendGrid skill. Use the local test fixture before configuring credentials for any live API check.
Example task
Give Codex three mocked cases: A has HTTP 200 with score 87, B has HTTP 202 with an empty body, C has HTTP 200 without score or metrics. Ask for a conservative label for each case and the value only when supplied.Expected result
The labels are A score available, 87. B pending and no score assigned. C score unavailable and no score assigned. It never converts pending or missing data into zero.
Upstream documentation
Original SKILL.md on GitHub · MIT license · Source reviewed
Frequently asked question
Does SendGrid return a zero score when engagement quality is missing?
Pending and omitted score fields are different states from a measured zero. Check the response status, field presence and the account's eligibility.