Keep raw string for tool-call args that literal_eval to non-JSON types#37
Open
nightlessbaron wants to merge 2 commits into
Open
Keep raw string for tool-call args that literal_eval to non-JSON types#37nightlessbaron wants to merge 2 commits into
nightlessbaron wants to merge 2 commits into
Conversation
rmfan
requested changes
Jul 2, 2026
flukeskywalker
approved these changes
Jul 4, 2026
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
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
_deserialize_glm_valuefalls back toast.literal_evalfor arg values that aren't valid JSON, which can return types JSON can't represent. Seen in an RL run: the model wrote{480, 630}(meant[480, 630]), literal_eval turned it into a Python set, andjson.dumps(args)in_extract_ifmraisedTypeError— dropping every tool call in that turn.Accept the literal_eval result only if it round-trips through
json.dumps, otherwise keep the raw string, so a single malformed arg degrades to a string value instead of killing the whole parse. Added a regression test (fails before, passes after; full detector suite green).