Claude/add llm api support wc mv e

[pinglanyi]

Description

Replace this paragraph with a description of what this PR is changing or adding, and why. Consider including before/after screenshots.

List which issues are fixed by this PR. For larger changes, raising an issue first helps reduce redundant work.

Pre-launch Checklist

• I signed the CLA.
• I read the Contributors Guide.
• I read the Style Guide.
• I have added updates to the CHANGELOG.
• I updated/added relevant documentation.
• My code changes (if any) have tests.

If you need help, consider asking for advice on the discussion board.

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

[gemini-code-assist]

Code Review

This pull request introduces support for OpenAI-compatible LLM APIs across various sample agents, which is a valuable enhancement. The implementation is mostly sound, but there are several areas for improvement regarding code duplication, robustness, and adherence to Python style conventions. Specifically, the logic for selecting a default model could be more intuitive, and there's a recurring pattern of unsafe environment variable access. Furthermore, significant portions of configuration and model selection logic are duplicated across multiple agents, which will make future maintenance challenging. The repository's style guide also requires tests for new code changes, which appear to be missing for the new logic introduced.

[gemini-code-assist]

The logic for selecting the default model appears to prioritize Gemini if both OPENAI_API_KEY and GEMINI_API_KEY are set. Given that the documentation in .env.example recommends OpenAI, it would be more intuitive to prioritize OpenAI whenever its key is available.

Additionally, this block of model selection logic is duplicated across several sample agents (e.g., in contact_multiple_surfaces, orchestrator, restaurant_finder). This makes maintenance difficult. Consider refactoring this logic into a shared utility function to improve maintainability and ensure consistency.

Suggested change

	_default_model = (
	"openai/gpt-4o"
	if os.getenv("OPENAI_API_KEY") and not os.getenv("GEMINI_API_KEY")
	else "gemini/gemini-2.5-flash"
	)
	_default_model = (
	"openai/gpt-4o"
	if os.getenv("OPENAI_API_KEY")
	else "gemini/gemini-2.5-flash"
	)

[gemini-code-assist]

Accessing the environment variable with os.environ["OPENAI_BASE_URL"] can raise a KeyError if the variable is not set. Although the if condition checks for its existence, using os.getenv("OPENAI_BASE_URL") for the assignment is safer against future refactoring and more consistent with the check. This pattern is repeated across several __main__.py files in this PR.

Suggested change

	if os.getenv("OPENAI_BASE_URL") and not os.getenv("OPENAI_API_BASE"):
	os.environ["OPENAI_API_BASE"] = os.environ["OPENAI_BASE_URL"]
	if os.getenv("OPENAI_BASE_URL") and not os.getenv("OPENAI_API_BASE"):
	os.environ["OPENAI_API_BASE"] = os.getenv("OPENAI_BASE_URL")

[gemini-code-assist]

According to PEP 8, variable names should be in lowercase_with_underscores. The name LITELLM_MODEL suggests it's a constant, but its value is determined dynamically from environment variables. Renaming it to litellm_model would improve readability and adherence to Python's style guide. Please note that you will also need to update its usage on line 119. This naming issue is also present in other files (e.g., MODEL_ID in samples/personalized_learning/agent/agent.py).

Suggested change

	LITELLM_MODEL = os.getenv("AI_MODEL") or os.getenv("LITELLM_MODEL", _default_model)
	litellm_model = os.getenv("AI_MODEL") or os.getenv("LITELLM_MODEL", _default_model)