ClawAudit verdict
devops-agent
DevOps automation skill with explicit safety constraints requiring user confirmation for destructive operations, dry-run mode, rollback support, and prohibitions on logging secrets; behavior matches stated operational purpose.
⚠ 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
network_out
package_install
Findings (12)
Possible hardcoded credential
SKILL.md · code · token: '<BOT_TOKEN>
References sudo — requests elevated privileges
SKILL.md · code · sudo
Pipe to python — executes piped content as Python code
SKILL.md · prose · downgraded · | Python
HTTP request to bare IP address — common in malicious payloads
SKILL.md · code · http://127.0.0.1
apt-get install — installs system packages
SKILL.md · code · apt-get install
Pipe to bash — executes piped content as shell commands
README.md · prose · downgraded · | Bash
Pipe-to-shell pattern (curl | sh) — supply chain attack vector
scripts/preflight-check.sh · prose · downgraded · curl -fsSL https://get.docker.com | sh
Pipe to sh — executes piped content as shell commands
scripts/preflight-check.sh · prose · downgraded · | sh
Opens WebSocket connection
SKILL.md · code · WebSocket
References webhook/callback URL
SKILL.md · code · WEBHOOK_URL
References SSH/GPG private keys
scripts/backup-generator.sh · prose · downgraded · GPG_KEY
yum install — installs system packages
scripts/preflight-check.sh · prose · downgraded · yum install
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class A/B/C). 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.
network_innetwork_outpackage_install Is this flag fair?
Thanks — recorded.