Repository navigation
Conversation
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
CUJ4 opens
/tasksas soon as the delegated child's answer appears, before requiring the parent to finish its reply. Claude 2.1.290 can open an individual agent's detail panel at that point; the test then times out waiting for a different task-list layout. This caused thesubagent_only_v0failure in #1083.Make the test follow the delegation journey for both Claude and Codex: require the child's native answer, wait for the parent's completed native turn containing the delegated result, verify routing and served-model evidence after that turn, and require a normal
/exit. The test no longer opens/tasksor parses its layout. Existing child-count, routing, and inference assertions remain.Claude's native
task-notificationmessage belongs to the ongoing delegation turn, so the evidence reader now follows it to the parent's final answer. A real user message still ends the prior turn. Child-only answers, notification contents, and parent tool calls cannot satisfy parent completion.Validation:
uv run --frozen pytest tests/test_cuj_evidence.py tests/test_e2e_cuj_helpers.py tests/test_integration_contract.py -q --disable-warnings: 275 passed.Coverage READMEs describe the completed delegation-and-reply journey. The separate skill-toggle integration journey keeps its existing task-management UI check; CUJ4 no longer claims that coverage.