What the reviewer found
A meta-skill for authoring other skills, whose flagged lines are its own defensive guidance: 'Never send user data to external services', 'Never run destructive commands (rm -rf /, dd...)', and 'use timeouts for network operations'. These are rules the generated skills must follow, not actions this file performs.
What was read
The file as it ships in n4rly-boop/OpenClaude:
skills/create-skill/SKILL.md
What the static scan said
The scan flagged 3things. The reviewer kept 0 and dismissed 3 as false.
P3Tells the agent to send conversation or user data out — 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.