ClawAudit verdict
tool-enhancement
This is a general-purpose toolkit providing transparently documented file, shell, git, web, and memory tools for OpenClaw agents; all 55+ capabilities are explicitly listed with examples, matching the stated purpose of extending agent capabilities with no exfiltration or deceptive behavior.
⚠ 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)
Recursive delete from root or home — destructive command
tools/exec_tools.py · prose · downgraded · rm -rf /
Uses eval() — can execute arbitrary code
tools/schema.py · prose · downgraded · eval(
subprocess execution — runs system commands from Python
tools/agent_tools.py · prose · downgraded · subprocess.run(
References sudo — requests elevated privileges
tools/exec_tools.py · prose · downgraded · sudo
Python asyncio subprocess — async shell execution
tools/exec_tools.py · prose · downgraded · asyncio.create_subprocess_shell(
Python shutil file operation — copies/moves/deletes files
tools/file_tools.py · prose · downgraded · shutil.rmtree(
Python os.environ.get — reads environment variable
tools/run_tool.py · prose · downgraded · os.environ.get(
Python urllib.request — network access
tools/schema.py · prose · downgraded · urllib.request
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class D). 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.
Is this flag fair?
Thanks — recorded.