ClawAudit verdict
lobsterai-report-agent
Multi-agent document generation system that writes feasibility reports via parallel sub-agents; process_exec is used to run local Python scripts for document assembly, behavior matches stated 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.
Findings (5)
Uses eval() — can execute arbitrary code
references/bug_fix_guide.md · code · eval(
os.system/popen — direct OS command execution
parallel_tracker.py · prose · downgraded · os.system(
Popular HTTP library — network access
references/bug_fix_guide.md · code · got
subprocess execution — runs system commands from Python
src/config.py · prose · downgraded · subprocess.run(
Python os.environ.get — reads environment variable
notify.py · prose · downgraded · os.environ.get(
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.
process_exec Is this flag fair?
Thanks — recorded.