What the reviewer found
Jeffy is a user-invoked autonomous improvement-loop skill (/jeffy) with extensive safety gates: it refuses to launch on a dirty tree, bad line endings, nested loops or Enhance-mode mismatches, and never allowlists push/force. AR1's evidence is a refusal check (not 'never refuse'), PE2's sudo is one of three OS-specific ways to install the jq dependency, and AS1 just writes the skill's own transient, gitignored loop-state file under the project's own .claude/ - none read credentials. Flagged low only because unattended runs are told to allowlist broad tool permissions.
dual-use— a security tool that can be misused
What was read
The file as it ships in lenamonj/jeffy-loop:
skills/jeffy/SKILL.md
What the static scan said
The scan flagged 3things. The reviewer kept 0 and dismissed 3 as false.
AR1Tells the agent never to refuse — false positivePE2Asks for root — false positiveAS1Reads agent configuration directories — 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.