[Radoub] chore: Challenge AI-generated issues in pre-warm and init-item - #2705
Open
LordOfMyatar wants to merge 1 commit into
Open
[Radoub] chore: Challenge AI-generated issues in pre-warm and init-item#2705LordOfMyatar wants to merge 1 commit into
LordOfMyatar wants to merge 1 commit into
Conversation
AI-filed tech debt carries fabricated line numbers and invented magnitudes; both commands now dispatch refuting Explore agents before planning or coding. init-item skips the pass when a pre-warm plan already records verification.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
AI-filed tech debt carries fabricated line numbers, invented magnitudes, and stale premises.
Pre-warming #2677 found three of five issues materially wrong, including one whose headline
perf claim collapsed once verified — the data it called uncached is cached.
/pre-warmPhase 1.4 and/init-itemPhase 1.5 now dispatch Explore agents instructed torefute rather than confirm, requiring quoted code at real line numbers and a per-claim verdict.
Wrong premises stop the work; wrong details get the issue body corrected first.
Scope is deliberately narrow: AI-generated tech debt and
/code-reviewoutput only. Humanenhancements are exempt, since a feature describes what should exist and has no premise to
falsify.
/init-itemskips the pass when the plan already records a Verification Summary, sothe challenge runs once per item.
Related Issues
Relates to #2677
Tests
Documentation-only change to
.claude/commands/— no code paths affected.🤖 Generated with Claude Code