phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-10-04 12:07:34 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Fix: Ensure that tool calls with no arguments get handled correctly #3560

Browse source 
When a model decides to use an MCP tool call that requires no arguments, it sets the arguments field to None. This causes validation errors because this field gets removed when being parsed by an openai compatible inference provider like vLLM
This PR ensures that, as soon as the tool call args are accumulated while streaming, we check to ensure no tool call function arguments are set to None - if they are we replace them with "{}"

Closes #3456

Added new unit test to verify that any tool calls with function arguments set to None get handled correctly

Signed-off-by: Jaideep Rao <jrao@redhat.com>
This commit is contained in:
Jaideep Rao 2025-09-26 08:17:33 -04:00
parent 6cce553c93
commit 4aa586d7af
3 changed files with 163 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

5
llama_stack/providers/inline/agents/meta_reference/responses/streaming.py
View file

@ -328,8 +328,11 @@ class StreamingResponseOrchestrator:
# Emit arguments.done events for completed tool calls (differentiate between MCP and function calls)
for tool_call_index in sorted(chat_response_tool_calls.keys()):
tool_call = chat_response_tool_calls[tool_call_index]
# Ensure that arguments, if sent back to the inference provider, are not None
tool_call.function.arguments = tool_call.function.arguments or "{}"
tool_call_item_id = tool_call_item_ids[tool_call_index]
final_arguments = chat_response_tool_calls[tool_call_index].function.arguments or ""
final_arguments = tool_call.function.arguments
tool_call_name = chat_response_tool_calls[tool_call_index].function.name
# Check if this is an MCP tool call