ClawAudit verdict
LLM Stream Guard
llm-loop-breaker
Stream entropy guard that patches global.fetch to detect and abort hallucinating LLM streams, combined with a host resource watchdog daemon; while powerful, the global.fetch patching is its core documented purpose and all behavior serves the stated gateway protection goal.
⚠ 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 (5)
Pipe to python — executes piped content as Python code
SKILL.md · prose · downgraded · | Python
apt-get install — installs system packages
deploy.sh · prose · downgraded · apt-get install
Dynamic import() — loads module at runtime
deploy.sh · prose · downgraded · import("
subprocess execution — runs system commands from Python
host-resource-watchdog.py · prose · downgraded · subprocess.check_output(
Python os.environ.get — reads environment variable
host-resource-watchdog.py · prose · downgraded · os.environ.get(
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class B/D). Final tier capped at Caution — cannot be lifted by any downgrade, example-payload opt-in, or allowlist.
Permissions & capabilities
Requires 1 environment variable. Requires 3 system binaries.
Is this flag fair?
Thanks — recorded.