Skip to content

[bugfix] Fix bug with exploding agent runs due to broken conversion of content Parts to LiteLLM messages format #4303

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
alexkuzmik wants to merge 6 commits into
base: main
Choose a base branch
from
+41 −0
Open

[bugfix] Fix bug with exploding agent runs due to broken conversion of content Parts to LiteLLM messages format #4303

Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
ae90205
Make _part_has_payload respect function response
alexkuzmik Jan 28, 2026
04657c1
Add new test
alexkuzmik Jan 28, 2026
9b27ca1
Merge pull request #1 from alexkuzmik/alexkuzmik/fix-bug-with-ignored…
alexkuzmik Jan 28, 2026
0fe6e39
update test
alexkuzmik Jan 28, 2026
e513630
Merge branch 'alexkuzmik/fix-bug-with-ignored-tool-response-content-f…
alexkuzmik Jan 28, 2026
aba564c
Merge upstream/main and resolve test conflict
alexkuzmik Feb 4, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
41 changes: 41 additions & 0 deletions tests/unittests/models/test_litellm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1013,6 +1013,47 @@ def test_append_fallback_user_content_ignores_function_response_parts():
assert llm_request.contents[0].parts[0].text is None


@pytest.mark.asyncio
async def test_generate_content_async_no_fallback_for_function_response(
mock_acompletion, lite_llm_instance
):
"""Tests that no fallback message is added for a user message with a function response."""
llm_request = LlmRequest(
contents=[
types.Content(
role="user",
parts=[
types.Part.from_function_response(
name="test_function",
response={"result": "test_result"},
)
],
)
]
)

# Run generate_content_async which calls _append_fallback_user_content_if_missing
async for _ in lite_llm_instance.generate_content_async(llm_request):
pass

# Verify that the fallback message was NOT added to the llm_request
assert len(llm_request.contents) == 1
assert len(llm_request.contents[0].parts) == 1
assert llm_request.contents[0].parts[0].function_response is not None

# Verify that the message sent to litellm does not contain the fallback text
mock_acompletion.assert_called_once()
_, kwargs = mock_acompletion.call_args
user_messages = [
message for message in kwargs["messages"] if message["role"] == "user"
]
assert not any(
message.get("content")
== "Handle the requests as specified in the System Instruction."
for message in user_messages
)


litellm_append_user_content_test_cases = [
pytest.param(
LlmRequest(
Expand Down