Caveat verdict
policy-engine
Security governance plugin enforcing allowlists and deny-patterns on tool calls via before_tool_call hooks; executionSinkDetected reflects pattern-matching on commands which is the stated purpose.
⚠ 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 (8)
Accesses sensitive system files
SKILL.md · prose · downgraded · /etc/passwd
Pipe-to-shell pattern (curl | sh) — supply chain attack vector
DESIGN.md · prose · downgraded · curl ... | bash
Pipe to bash — executes piped content as shell commands
DESIGN.md · prose · downgraded · | bash
Pipe-to-shell pattern (wget | sh)
src/patterns.ts · prose · downgraded · wget -O- | sh
Pipe to sh — executes piped content as shell commands
src/patterns.ts · prose · downgraded · | sh
Recursive delete from root or home — destructive command
tests/engine.test.ts · prose · downgraded · rm -rf /
Accesses .ssh directory
tests/engine.test.ts · prose · downgraded · .ssh/
Base64 encoding/decoding
DESIGN.md · prose · downgraded · Base64-encode
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class A/F). 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.