Skip to content

Expand false-positive calibration corpus into CI-gated regressions #331

Open
Open
Expand false-positive calibration corpus into CI-gated regressions#331
Labels
area: detectorDetector/rule behaviordifficulty: mediumModerate task with multiple files or design choicesenhancementNew feature or requesttype: false-positiveSignal-quality false positive work

Description

@ColumbusLabs
ColumbusLabs
opened on Jul 21, 2026

Summary

examples/false-positives exists; turn it into a systematic per-rule regression corpus gated in CI.

Acceptance criteria

  • Structure for per-rule near-miss fixtures
  • CI job fails if known FP fixtures start alerting
  • Docs for adding new FP cases
  • Seed corpus for highest-noise rules

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    area: detectorDetector/rule behaviordifficulty: mediumModerate task with multiple files or design choicesenhancementNew feature or requesttype: false-positiveSignal-quality false positive work

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions