ClawAudit verdict
codex-agent
Project manager skill for orchestrating OpenAI Codex CLI via tmux; network_in is used to receive async notifications, and all operations are transparently described as managing Codex task 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 (6)
HTTP request to bare IP address — common in malicious payloads
references/codex-cli-reference.md · code · http://127.0.0.1
Uses exec() — may execute shell commands
SKILL.md · prose · downgraded · exec(
Python os.environ.get — reads environment variable
INSTALL.md · code · os.environ.get(
Accesses shell history/config
INSTALL.md · prose · downgraded · ~/.zshrc
Accesses OpenClaw config/secrets directly
INSTALL.md · prose · downgraded · ~/.openclaw/openclaw.json
Opens WebSocket connection
knowledge/features.md · prose · downgraded · WebSocket
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class A/B/D/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.
network_in Is this flag fair?
Thanks — recorded.