test(openai-chat): declare the dangling-toolcalls destination accepts the developer role

[lidge-jun]

Summary

• tests/adapters/openai/openai-chat-dangling-toolcalls.test.ts asserts that a developer barrier lands in a specific slot relative to a pending tool call, reading the wire role to find it. fix(openai-chat): fold the developer role unless a destination records that it accepts it #5334 made that role fold to system unless a destination records acceptance, so three cases started failing on dev and on every branch cut from it.
• The suite's destination now records that it accepts the role, so the placement assertions keep reading the role they were written against. No assertion is weakened and no case is removed.
• This is the last file in that class. A repository-wide sweep over every test that builds an openai-chat request and asserts a wire role found three: two were declared in fix(openai-chat): fold the developer role unless a destination records that it accepts it #5334, the Lab conformance vector in fix(lab): record that the Chat conformance destination accepts the developer role #5341, and this one. The remaining "developer" matches assert the internal representation before the adapter maps it, or accept either role, and are unaffected.

Verification

• The failing case names come from the hosted log for shard 3/4 of run 35511465069: T1, T5 and T6 of openai-chat dangling tool_calls hardening.
• tests/web-search/web-search.test.ts was checked and excluded deliberately: its capturing adapter pushes parsed.context.messages, which is the internal representation rather than the wire.
• Local suites, typecheck and builds were not run in this lane; exact-head hosted CI on this pull request is the execution evidence.

Checklist

• Scope stays focused and avoids unrelated cleanup.
• Docs or release notes were updated when needed.
• Security-sensitive changes were reviewed for secrets, auth, and unsafe defaults.

[coderabbitai]

Warning

Review limit reached

Next included review available in 2 minutes.

Check out review usage here.

View limit details

Limit details: You’ve used all 10 included reviews currently available.

You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository.

Learn how review limits work.

Review configuration:

⚙️ Run configuration

Configuration used: Repository: lidge-jun/opencodex/.coderabbit.yaml

Review profile: ASSERTIVE

Plan: Advanced

Run ID: c1b65274-66f7-4adc-ad3d-d86ea3e48b90

📥 Commits

Reviewing files that changed from the base of the PR and between 600075d and 9a81476.

📒 Files selected for processing (1)

• tests/adapters/openai/openai-chat-dangling-toolcalls.test.ts

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-09-20T13:13:45.851550Z	9a81476	PR opened

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" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[github-actions]

✅ Deterministic PR hygiene checks passed.