Modernize workflows#4668
Conversation
|
@codex fix any issues you find with the change |
There was a problem hiding this comment.
💡 Codex Review
codex/.github/workflows/issue-deduplicator.yml
Lines 108 to 110 in 2ac4dba
The Codex invocation now requests an object that wraps the candidate issues under an issues array, yet the comment script still treats the parsed output as a bare array (numbers.map(...)). When Codex returns the new object shape, numbers becomes an object and .map throws, causing the job to fail before posting any comment. The consumer needs to read issues from the parsed object before iterating.
codex/.github/workflows/issue-labeler.yml
Lines 93 to 97 in 2ac4dba
The gather step now defines an output_schema that returns an object with a labels array, but the apply step continues to validate codex_output with jq -e 'type == "array"' and iterates over it directly. With the new schema the output is a JSON object, so this check prints “Codex output was not a JSON array” and exits early, meaning no labels are ever applied. The script should parse the object and extract .labels before checking/looping.
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting
|
Summary
Testing
|
Use structured output and inline prompt messages.