Repository navigation
Conversation
…ing the loop agent_runner_loop called json.loads(tc.function.arguments) unguarded. With the text tool protocol, a model emitting `"arguments": "not json"` raised JSONDecodeError, and `"arguments": "[1,2,3]"` raised AttributeError on args['_index'] - both abort the whole run. parse_tool_call now maps non-object / unparsable arguments to the existing bad_json tool, so the error is fed back to the model for self-correction.
Xperiamol
force-pushed
the
fix/malformed-tool-args
branch
from
September 29, 2026 07:56
5c826d3 to
fc8d084
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
agent_runner_loopcallsjson.loads(tc.function.arguments)unguarded. With the text tool protocol, a model emitting"arguments": "not json"raisesJSONDecodeError, and"arguments": "[1,2,3]"raisesAttributeErroratargs['_index']- either aborts the whole run instead of reaching the existingbad_jsonself-correction path.parse_tool_callnow maps unparsable / non-object arguments tobad_json, so the error goes back to the model. Valid calls are unchanged.arguments"not json at all"JSONDecodeErrorbad_jsonfed back, run continues"[1,2,3]"AttributeErrorbad_jsonfed back, run continues{"script":"print(1)"}Net +6 lines.