ClawAudit verdict
otc-confirmation
One-time confirmation code security mechanism delivering codes via SMTP and requiring out-of-band confirmation before executing sensitive operations; designed to prevent unauthorized execution.
⚠ 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 (8)
Prompt injection — tries to override agent instructions
ai-devops-agent-security-pack/examples/devops_workflow.md · code · Ignore previous instructions
Recursive delete from root or home — destructive command
ai-devops-agent-security-pack/03_permission_guard.md · code · rm -rf /
Accesses .ssh directory
ai-devops-agent-security-pack/examples/devops_workflow.md · code · .ssh/
subprocess execution — runs system commands from Python
ai-devops-agent-security-pack/03_permission_guard.md · code · subprocess.run(
References sudo — requests elevated privileges
ai-devops-agent-security-pack/03_permission_guard.md · code · sudo
Base64 encoding/decoding
ai-devops-agent-security-pack/code_examples/audit_logger.py · prose · downgraded · Base64-encode
Python os.environ.get — reads environment variable
ai-devops-agent-security-pack/code_examples/confirmation_service.py · prose · downgraded · os.environ.get(
Sets world-executable permissions
scripts/generate_code.sh · prose · downgraded · chmod 700
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class A/B/D). Final tier capped at Caution — cannot be lifted by any downgrade, example-payload opt-in, or allowlist.
Permissions & capabilities
Requires 3 environment variables.
Is this flag fair?
Thanks — recorded.