ClawAudit verdict
aicwos
Local copywriting data pipeline that reads from and writes to local database via CLI scripts, with no external network calls or credential exfiltration evident in the content.
⚠ 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 (4)
Pipe to python — executes piped content as Python code
SKILL.md · code · | python
Python shutil file operation — copies/moves/deletes files
scripts/core/lecturer_service.py · prose · downgraded · shutil.rmtree(
Blob URL — may embed executable content
scripts/core/semantic_model.py · prose · downgraded · blob:
Python urllib.request — network access
scripts/knowledge_sync.py · prose · downgraded · urllib.request
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class A). 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.