mirror/nanobot
1
0
Fork 0
mirror of https://github.com/HKUDS/nanobot.git synced 2026-04-30 06:45:55 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix(loop): add return_exceptions=True to parallel tool gather

Browse Source 
Without this flag, a BaseException (e.g. CancelledError from /stop)
in one tool would propagate immediately and discard results from the
other concurrent tools, corrupting the OpenAI message format.

With return_exceptions=True, all tool results are collected; any
exception is converted to an error string for the LLM.

Made-with: Cursor
This commit is contained in:
xzq.xu 2026-03-18 17:33:22 +08:00 committed by chengyongru
parent 6028b4828b
commit a2edee145f
1 changed files with 5 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
nanobot/agent/loop.py
View File

@ -261,12 +261,16 @@ class AgentLoop:
# Execute all tool calls concurrently — the LLM batches # Execute all tool calls concurrently — the LLM batches
# independent calls in a single response on purpose. # independent calls in a single response on purpose.
# return_exceptions=True ensures all results are collected
# even if one tool is cancelled or raises BaseException.
results = await asyncio.gather(*( results = await asyncio.gather(*(
self.tools.execute(tc.name, tc.arguments) self.tools.execute(tc.name, tc.arguments)
for tc in response.tool_calls for tc in response.tool_calls
)) ), return_exceptions=True)
for tool_call, result in zip(response.tool_calls, results): for tool_call, result in zip(response.tool_calls, results):
if isinstance(result, BaseException):
result = f"Error: {type(result).__name__}: {result}"
messages = self.context.add_tool_result( messages = self.context.add_tool_result(
messages, tool_call.id, tool_call.name, result messages, tool_call.id, tool_call.name, result
) )