Context-aware reasoning
audytx looks at how your resources relate before it reports a finding. It runs 33 reasoning axes. Three of them:
DLQ identity
Looks at which queues act as dead-letter queues.
Lambda invocation graph
Looks at how each Lambda function is invoked.
Encryption variants
Looks at which encryption option a resource uses.
The reasoning is visible. A finding that the rest of
your code makes harmless does not silently disappear. It appears in the
PR comment under a collapsed 🧠 audytx reasoned about N
findings and chose not to flag them block, with the check ID, the
resource address and a short rationale. The reasoning is also written
into the SARIF upload as a suppressions[] entry with the
same justification. GitHub does not document acting on that field, so
whether the Security tab shows such an alert as closed is up to GitHub.