fix(deliberation): stop passing --markdown/--limit to graph probe commands - #470
Merged
Merged
Conversation
…mands The scope planner appended --markdown --limit 8 to every graphtrail call, but only 'context' accepts those flags. callers, callees, and impact reject them and exit non-zero, so the planner always derived a single evidence scope and rejected every deliberation plan before dispatch (issue #442 experiment, route C: 8/8 pre-dispatch failures). Pass the markdown flags only on the context probe and add a regression test with a fake graphtrail binary that mirrors the real CLI's flag surface. Co-authored-by: Claude <noreply@anthropic.com>
|
Important Review skippedAuto reviews are limited based on label configuration. 🏷️ Required labels (at least one) (1)
Please check the settings in the CodeRabbit UI or the ⚙️ Run configurationConfiguration used: Repository: escoffier-labs/coderabbit/.coderabbit.yaml Review profile: CHILL Plan: Pro Plus Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
This was referenced Jul 24, 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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What
The deliberation scope planner appended
--markdown --limit 8to every graphtrail call, but onlycontextaccepts those flags.callers,callees, andimpactreject them and exit non-zero, so the planner always derived a single evidence scope and rejected every--deliberateplan before dispatch.This is the direct cause of route C's 0/10 in the #442 experiment: 8/8 cases failed pre-dispatch with
deliberation could not assemble enough grounded GraphTrail evidence scopesand route C never made a model call.Change
_graphtrail_markdownis now_graphtrail_scope_textand only passes--markdown --limit 8on the context probe.--markdown/--limiton the three probe commands). The existing tests mockedderive_evidence_scopeswholesale, which is why this was never caught.Verify
./scripts/verifypassed locally (Brigade verify run20260724-010851-work-verify-e2ec06, exit 0). The new test fails on main and passes here.Notes for review
This fixes the mode's entry path; it takes no position on whether the flag stays. That call is tracked in #442 and the experiment scoring there was produced with the mode broken, so route C remains unmeasured until a rerun.
Refs #442