Caveat verdict
guardian-angel
Ethics and virtue-based AI conscience framework; all content is defensive guidance for safe agent behavior with no suspicious execution or exfiltration patterns.
⚠ 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 (6)
Prompt injection — tries to override agent instructions
references/prompt-injection-defense.md · code · IGNORE ALL PREVIOUS INSTRUCTIONS
Recursive delete from root or home — destructive command
PLUGIN-SPEC.md · code · rm -rf /
References sudo — requests elevated privileges
PLUGIN-SPEC.md · code · sudo
Possible prompt injection — attempts to redefine agent identity
references/prompt-injection-defense.md · code · You are now
Instructs covert action — may act without user awareness
drafts/edge-cases-consolidated.md · prose · downgraded · quietly
Popular HTTP library — network access
references/category-rubric.md · prose · downgraded · Got
Permissions & capabilities
No declared permissions — minimal attack surface.
Is this flag fair?
Thanks — recorded.