Skip to content

fix(#537): add remediation consistency rule to correctness sub-agent - #538

Open
fullsend-ai-coder[bot] wants to merge 1 commit into
mainfrom
agent/537-remediation-consistency-rule
Open

fix(#537): add remediation consistency rule to correctness sub-agent#538
fullsend-ai-coder[bot] wants to merge 1 commit into
mainfrom
agent/537-remediation-consistency-rule

Conversation

@fullsend-ai-coder

Copy link
Copy Markdown
Contributor

Summary

  • Adds a "Remediation consistency" rule to the correctness sub-agent definition (skills/pr-review/sub-agents/correctness.md)
  • Prevents the agent from suggesting remediations that contradict its own severity analysis (e.g., suggesting a fallback to a state it described as "already broken")
  • Adds guidance for companion-PR scenarios to frame remediation around merge coordination rather than broken fallbacks

Context

On PR #527, the correctness sub-agent repeatedly raised an [api-contract] finding where it downgraded severity because a prior approach was "already broken," then simultaneously suggested falling back to that broken approach. This contradiction required 8+ human dismissals. The root cause was missing self-consistency guidance between severity rationale and remediation suggestions.

Changes

The new rule is placed under the "Runtime mechanism checklist" section, after the existing failure-paths bullet. This location is appropriate because api-contract findings are in the correctness sub-agent's domain, and the runtime mechanism checklist is where it evaluates cross-component contracts.

Testing

  • Verified the rule text matches the issue's proposed change
  • Confirmed placement is after the failure-paths bullet in the Runtime mechanism checklist
  • gitlint passed on the commit message
  • Secret scan passed

Closes #537

Post-script verification

  • Branch is not main/master (agent/537-remediation-consistency-rule)
  • Secret scan passed (gitleaks — 5e45ff31e2ccc1c69f0126f2f690aaef8e9712df..HEAD)
  • PR body secret scan passed (gitleaks — no-git)
  • Pre-commit hooks passed (authoritative run on runner)
  • Tests ran inside sandbox

Add guidance to the correctness sub-agent definition preventing
it from suggesting remediations that contradict its own analysis.
When a severity rationale acknowledges a prior state is broken,
the agent must not suggest falling back to that broken state.
When a companion PR provides the missing piece, the agent should
frame remediation around merge coordination instead.

The rule is placed under the "Runtime mechanism checklist" section
in correctness.md, after the failure-paths bullet, where the
agent evaluates cross-component contracts and API findings.

Note: pre-commit could not run in sandbox (network restriction).
The post-script runs it authoritatively on the runner.

Closes #537
@fullsend-ai-coder
fullsend-ai-coder Bot requested a review from a team as a code owner July 29, 2026 21:55
@fullsend-ai-coder fullsend-ai-coder Bot added the ready-for-review Triggers review agent dispatch label Jul 29, 2026
@fullsend-ai-review

fullsend-ai-review Bot commented Jul 29, 2026

Copy link
Copy Markdown

🤖 Finished Review · ✅ Success · Started 9:56 PM UTC · Completed 10:07 PM UTC
Commit: 76bc5d6 · View workflow run →

@fullsend-ai-review

Copy link
Copy Markdown

Review

Findings

Medium


Labels: PR modifies review agent sub-agent definition (skills/pr-review/sub-agents/correctness.md)

@fullsend-ai-review fullsend-ai-review Bot added requires-manual-review Review requires human judgment review-agent labels Jul 29, 2026
@github-actions

Copy link
Copy Markdown

This pull request has been automatically marked as stale because it has not had any activity in the last month. It will be closed in 2 weeks if no further activity occurs. Remove the stale label to reset the inactivity timer.

@github-actions github-actions Bot added the stale label Aug 29, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ready-for-review Triggers review agent dispatch requires-manual-review Review requires human judgment review-agent stale

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Correctness sub-agent should not suggest remediation that contradicts its own analysis

0 participants