fix(v1): trace Responses custom tools#2129
Open
hallerite wants to merge 1 commit into
Open
Conversation
hallerite
marked this pull request as ready for review
July 24, 2026 17:21
ApprovabilityVerdict: Approved Straightforward enhancement adding You can customize Macroscope's approvability policy. Learn more. |
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Parse Responses API
custom_tool_callandcustom_tool_call_outputitems into Verifiers' existing typedToolCallandToolMessagerepresentation.Codex uses custom tool items for tools such as
apply_patch. The Responses dialect previously recognized only function calls, so the patch call was absent from the assistant message and its result appeared as an empty user node in saved traces.Custom tool input is now retained as the tool-call arguments, the output is linked by call ID, and responses containing custom calls correctly report a tool-call finish reason.
Validation
uv run pytest tests/v1/test_graph.py(7 passed)uv run ruff check verifiers/v1/dialects/responses.py tests/v1/test_graph.pymarkdownlint-cli2,ruff check,ruff format, andtyhooksopenai/gpt-5.5, Prime sandbox, reward1.0apply_patchas a typed assistant tool call at node 4 and its linked tool result at node 5; the subsequent shell verification and final answer were also preserved