ClawAudit verdict
l76-core-arch
Architecture template and documentation skill for AgentSkills development showing best practices and code patterns; process_exec capability is illustrated with normal read/write/exec patterns for development use.
⚠ Flagged for review — coarse, uncorroborated signal, not a confirmed exploit. Review the config yourself before installing.
Automated static analysis — not a human review. ClawAudit flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.
Findings (5)
Possible hardcoded credential
references/troubleshooting.md · code · apiKey: 'your-key-here
Uses exec() — may execute shell commands
SKILL.md · code · exec(
Pipe to bash — executes piped content as shell commands
PUBLISH-SUMMARY.md · prose · downgraded · | Bash
Instructs covert action — may act without user awareness
references/troubleshooting.md · code · silently
References agent memory files
MEMORY_ITEMS.md · prose · downgraded · MEMORY.md
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class A/B). Final tier capped at Caution — cannot be lifted by any downgrade, example-payload opt-in, or allowlist.
Permissions & capabilities
Requires 1 system binary.
process_exec Is this flag fair?
Thanks — recorded.