Caveat verdict
gangjing
Contrarian code/architecture review skill that provides adversarial critique and red-teaming; despite executionSinkDetected=true, the content explicitly restricts script execution to only when the user explicitly requests it, and all operations are scoped to review and critique.
⚠ Flagged for review — coarse, uncorroborated signal, not a confirmed exploit. Review the config yourself before installing.
Automated static analysis — not a human review. Caveat flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.
Findings (10)
Recursive delete from root or home — destructive command
references/attack-patterns.md · code · rm -rf /
Accesses sensitive system files
references/attack-patterns.md · code · /etc/passwd
Reads /proc/self/environ — dumps all environment variables
references/attack-patterns.md · code · /proc/self/environ
<script> tag in markdown — potential code injection
references/attack-patterns.md · code · <script>
Unicode homoglyph detected — uses lookalike characters to evade pattern matching
references/attack-patterns.md · code
Dynamic __import__('os') — Python OS command execution
references/attack-patterns.md · code · __import__('os')
Pipe to python — executes piped content as Python code
SKILL.md · prose · downgraded · | Python
Instructs covert action — may act without user awareness
references/attack-patterns.md · code · silently
References child_process — can spawn system processes
templates/attack-engine-kit.md · code · child_process
Popular HTTP library — network access
templates/attack-engine-kit.md · code · got
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class D). Final tier capped at Caution — cannot be lifted by any downgrade, example-payload opt-in, or allowlist.
Permissions & capabilities
No declared permissions — minimal attack surface.
Is this flag fair?
Thanks — recorded.