fix(crosscheck): isolate verdict repair after provider failures - #333
Merged
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Intent
Harden Azure Crosscheck after two real consumer failures. Preserve exact-head binding, isolated Azure model/tool/verifier compartments, credentialless evidence verification, cleanup guarantees, and fail-closed semantics. Keep exactly one bounded structured-verdict repair. When an xhigh reviewer hits its output limit, the repair must not inherit the oversized transcript or repeat the same failure: use one fresh ephemeral minimal-reasoning attempt over the identical exact-head review packet. Preserve bounded, sanitized provider error diagnostics, internally consistent failed-run telemetry, executable regressions, and lightweight operator documentation. Do not use Bugbot; adversarially self-review, ship through a PR, merge, deploy immediately, and rerun Relvino PR kunchenguid#1514 until it returns an admitted verdict.
What Changed
Risk Assessment
✅ Low: The repair is well-bounded and source evidence supports the required fresh-session retry, fail-closed behavior, sanitized provider diagnostics, telemetry aggregation, and executable regression coverage.
Testing
Completed 1 recorded test check.
Pipeline
Updates from git push no-mistakes
✅ **intent** - passed
✅ No issues found.
✅ **Rebase** - passed
✅ No issues found.
🔧 **Review** - 1 issue found → auto-fixed ✅
bin/fm-crosscheck-pi-reviewer.py:176- The required “bounded, sanitized provider error diagnostics” contract is not met:" ".join(error_message.split())[:512]bounds length and normalizes whitespace, but preserves non-whitespace control characters such as ANSI escape sequences. A provider-suppliederrorMessagecan therefore inject terminal controls when the exception is printed. Sanitize to a safe printable representation before appending the bounded diagnostic.🔧 Fix: Sanitize bounded provider error diagnostics
✅ Re-checked - no issues remain.
if [ "${FM_AZURE_VALIDATION_CELL:-0}" = 1 ]; then exec "$FM_AZURE_VALIDATION_SHARD_BRIDGE" behavior --count "${FM_AZURE_VALIDATION_SHARD_COUNT:-8}"; else exec bin/fm-no-mistakes-test-command.sh; fi✅ **Document** - passed
✅ No issues found.
✅ **Lint** - passed
✅ No issues found.
✅ **Push** - passed
✅ No issues found.