[Aditya Kumar (Aditya-XR)]

Motivation & Context

Message("user", "Hello") creates one text content per character, because Message.__init__ iterates contents and a str is itself a sequence of strings. The contents type hint accepts a bare string, the docstring says strings are converted to text content, and the repository's own hosted-workflows sample uses this form. With Chat Completions, each character is then sent as a separate user message.

Description & Review Guide

• What are the major changes?

  _parse_content_list wraps a bare string in a list before iterating, so it becomes a single text content, the same as Message("user", ["Hello"]). Message.__init__ already hands contents to this helper, so the constructor itself is unchanged:

   def _parse_content_list(contents_data: Sequence[Any]) -> list[Content]:
  +    if isinstance(contents_data, str):
  +        contents_data = [contents_data]
       contents: list[Content] = []

  The docstrings of Message.__init__ and _parse_content_list now say a single string is treated as one text item. New tests in test_types.py cover a bare string passed to Message positionally and by keyword, an empty string (one empty text content, as with [""]), and a bare string passed straight to _parse_content_list. They fail without the change: a 19-character string produces 19 contents, and "" produces none.

• What is the impact of these changes?

  For the repro in the linked issue, Message("user", "Write a slogan.") now has one text content, and Chat Completions receives one user message instead of 15. Lists, None and every other input are unchanged, and Message.__init__ is the helper's only caller. The sample and the 27 test call sites that pass a bare string now get the message they intended.

• What do you want reviewers to focus on?

  I kept the contents type hint as it is, since str already satisfies Sequence[str]; I can spell out str | Sequence[...] if you'd like it explicit. A single Content or dict passed without a list is still not accepted, as before. Those aren't sequences of items, so type checkers already flag them, and a single Content raises a TypeError.

Related Issue

Fixes #9326

AI Assistance

• No material AI assistance was used.
• This is an AI-assisted contribution. I reviewed, understood, and verified all submitted content and accept responsibility for it.

AI assistance details: I used Claude Code as an assistant to help find the bug and to draft the fix, the tests, the linked issue and this description. I did the final review myself: I read and understood every change, checked the repro and test results, and made the final decisions on what to submit.

Contribution Checklist

• The code builds clean without any errors or warnings
• All unit tests pass, and I have added new tests where possible
• The PR follows the Contribution Guidelines
• This PR links to an agreed issue with no competing open PR, or the Related Issue section documents a trivial-change or repository-automation exception.
• This is not a breaking change. If it is a breaking change, add the breaking change label (or add "[BREAKING]" to the title prefix, before or after any language prefix) — a workflow keeps the label and title prefix in sync automatically.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

[Eduard van Valkenburg (eavanvalkenburg)]

/review

[github-actions]

MAF Automated Review — Iteration 1

Result: No findings
Scope: full PR (1 commit(s)): 55e431ec6065
Model: gpt-5.6-sol

Overview

The change narrowly normalizes a bare string into one content item before the existing parser runs, while leaving None, lists, mixed content, metadata, and serialization paths unchanged. The new positional, keyword, and empty-string tests cover the corrected contract, and existing list and serialization tests constrain regressions. No publishable Critical, High, or Medium issue was established.

Reviewed the supplied pull-request change set across correctness, security/reliability, architecture, and failure behavior.
No publishable findings remained after source verification for this scope.

[Eduard van Valkenburg (eavanvalkenburg)]

Aditya Kumar (@Aditya-XR) The current PR template requires an AI Assistance disclosure. Please state whether material AI assistance was used and, if it was, briefly identify the assisted areas.

[Aditya Kumar (Aditya-XR)]

Thanks, I've added the AI Assistance section to the description. Yes, this was AI-assisted: I used Claude Code as an assistant to help find the bug and draft the fix, tests, issue and PR description. The final review was mine: I went through every change and the test results myself before submitting, and I'm happy to answer any questions about it.