What the reviewer found
A skill-marketplace hub (oc-hub) that fetches and imports skills from ClawHub/GitHub; the flagged text is the grep patterns and prompt-injection checklist its own mandatory security-scan subagent runs against imported files before install, and rm -rf only clears its own staging directory. The installer should know it fetches and writes third-party skill content, though a CLEAN scan result proceeds to install automatically without asking.
network— calls the vendor’s APIinstalls-software— installs software
What was read
The file as it ships in mfielding92/ClawedBack:
.claude/skills/oc-hub/SKILL.md
What the static scan said
The scan flagged 5things. The reviewer kept 0 and dismissed 5 as false.
P1Instruction-override phrasing — false positivePE3Reaches for credential files — false positiveAS3Enumerates other installed skills — 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.