What version of Codex is running?
N/A
Which model were you using?
N/A
What platform is your computer?
N/A
What steps can reproduce the bug?
Create an issue in this repository (like #4656).
What is the expected behavior?
The deduplicator action reports issues that may be duplicates of the new issue.
What do you see instead?
The workflow fails with an error:
https://github.com/openai/codex/actions/runs/18222330874/job/51885023079
SyntaxError: Unexpected token '...'
at new AsyncFunction (<anonymous>)
at callAsyncFunction (/home/runner/work/_actions/actions/github-script/v7/dist/index.js:36187:16)
at main (/home/runner/work/_actions/actions/github-script/v7/dist/index.js:36285:26)
at /home/runner/work/_actions/actions/github-script/v7/dist/index.js:36260:1
at /home/runner/work/_actions/actions/github-script/v7/dist/index.js:36317:3
at Object.<anonymous> (/home/runner/work/_actions/actions/github-script/v7/dist/index.js:36320:12)
at Module._compile (node:internal/modules/cjs/loader:1529:14)
at Module._extensions..js (node:internal/modules/cjs/loader:1613:10)
at Module.load (node:internal/modules/cjs/loader:1275:32)
at Module._load (node:internal/modules/cjs/loader:1096:12)
Error: Unhandled error: SyntaxError: Unexpected token '...'
Additional information
Based on observing the output of the "Identify potential duplicates" job it does seem to successfully run the script to identify duplicates, but fails when it tries to report them.
With that said, it also seems like there are some issues with the deduplication logic itself. For example, the most likely duplicate issue it reports is the original issue itself.
What version of Codex is running?
N/A
Which model were you using?
N/A
What platform is your computer?
N/A
What steps can reproduce the bug?
Create an issue in this repository (like #4656).
What is the expected behavior?
The deduplicator action reports issues that may be duplicates of the new issue.
What do you see instead?
The workflow fails with an error:
https://github.com/openai/codex/actions/runs/18222330874/job/51885023079
Additional information
Based on observing the output of the "Identify potential duplicates" job it does seem to successfully run the script to identify duplicates, but fails when it tries to report them.
With that said, it also seems like there are some issues with the deduplication logic itself. For example, the most likely duplicate issue it reports is the original issue itself.