What the reviewer found
A bug-bounty/pentest playbook for security analysts hunting prompt-injection vulnerabilities in OTHER people's LLM applications. The 'ignore previous instructions', HTML-comment exfil, and ~/.ssh/id_rsa lines are labeled PoC payload examples describing what an attacker's injected content would look like, not instructions this file is trying to slip past the reviewing agent.
dual-use— a security tool that can be misusedprompt-injection— tries to steer the agentexample-or-fixture— an example or test fixture
What was read
The file as it ships in MingyiSecLab/Mingyi-Atlas:
src/skills/standard/analyst/prompt-injection/SKILL.md
What the static scan said
The scan flagged 3things. The reviewer kept 0 and dismissed 3 as false.
P1Instruction-override phrasing — false positiveP2Hidden instructions — false positivePE3Reaches for credential files — 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.