What the reviewer found
All are quoted example injection/jailbreak payloads for testing target LLM apps, not instructions to this agent.
What was read
The file as it ships in H-mmer/pentest-agents:
providers/cursor/.cursor/skills/pentest-agents-hunt-llm-ai/SKILL.md
What the static scan said
The scan flagged 9things. The reviewer kept 0 and dismissed 9 as false.
P1Instruction-override phrasing — false positiveP2Hidden instructions — false positiveAR3Nullifies safety policies — false positiveE1Sends data to an external URL — false positivePE3Reaches for credential files — false positiveSC3Encoded or obfuscated payload — false positiveSSRF1Cloud metadata endpoint — false positiveNETMakes network calls — false positiveSHRuns shell commands — false positive
How this review was made
Sonnet 5 read the files above on 6 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.