ClawAudit verdict
sentryskills
Security middleware running preflight, runtime, and output checks; processes data locally with no exfiltration.
⚠ 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.
Permission integrity
file_read+write
Findings (8)
Recursive delete from root or home — destructive command
install/codex_install.md · prose · downgraded · rm -rf ~
Pipe to bash — executes piped content as shell commands
shared/scripts/predictive_analysis.py · prose · downgraded · |bash
Instructs covert action — may act without user awareness
shared/scripts/predictive_analysis.py · prose · downgraded · silently
References sudo — requests elevated privileges
shared/scripts/predictive_analysis.py · prose · downgraded · sudo
setuid — privilege escalation mechanism
shared/scripts/predictive_analysis.py · prose · downgraded · setuid
subprocess execution — runs system commands from Python
shared/scripts/run_local_benchmark_template.py · prose · downgraded · subprocess.check_call(
Popular HTTP library — network access
shared/references/runtime_policy.phase2.json · prose · downgraded · axios
Python os.environ.get — reads environment variable
shared/scripts/logging_config.py · prose · downgraded · os.environ.get(
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class A). 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.
file_write Is this flag fair?
Thanks — recorded.