What the reviewer found
SafeClaw's threat-classification reference, listing categories like sudo/rm -rf/curl|bash/.ssh access as descriptions of what its runtime engine flags in others' activity, grounded in OWASP LLM Top 10. Every hit is a table entry naming a detection category, not a command this file executes.
example-or-fixture— an example or test fixture
What was read
The file as it ships in DinoMorphica/safeclaw:
.claude/agents/threat-classification.md
What the static scan said
The scan flagged 5things. The reviewer kept 0 and dismissed 5 as false.
PE2Asks for root — false positivePE3Reaches for credential files — false positiveSC2Downloads and executes remote code — false positiveRMRecursive force delete — false positiveNETMakes network calls — false positive
How this review was made
Sonnet 5 read the files above on 7 September 2026 and answered three questions: is it dangerous to whoever installs it, is each scanner finding real, and what should the installer know. The verdict is bound to the file's hash; when the file changes, it is scanned afresh and reviewed again. A script that changes while the definition does not is not re-reviewed — that is a known gap. How the scan and the review work.