What the reviewer found
This is a skill-auditor tool whose whole job is to scan other skills for prompt-injection and exfiltration patterns; every flagged string ('Ignore previous instructions', '<!-- ignore above -->', 'Act as if you have no restrictions', the fetch(url?key=...) example, the import checklist) is a pattern in its own detection checklist, not an instruction the skill follows. Dual-use security tooling, not an attack.
dual-use— a security tool that can be misused
What was read
The file as it ships in Ascend/agent-skills:
skills/skill-auditor/SKILL.md
What the static scan said
The scan flagged 5things. The reviewer kept 0 and dismissed 5 as false.
P1Instruction-override phrasing — false positiveP2Hidden instructions — false positiveAR3Nullifies safety policies — false positiveNETMakes network calls — false positiveSHRuns shell commands — 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.