ClawAudit verdict
xuanzang-skill
A motivational/role-play persona skill that adjusts the agent's communication style based on config; reads local reference files and has no network access, exfiltration, or malicious 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 (4)
Long base64 string (100+ chars) — likely obfuscated payload
README.md · frontmatter · ErFb4xpPbc08Rvlqa6EMHlcL1BA+iql+uUndzdXor2EWnkaw3Nr/n6nsYfQ1GzTRCxhHIfyJRiwh45IA
subprocess execution — runs system commands from Python
scripts/harness-engine.py · prose · downgraded · subprocess.run(
subprocess with shell=True — command injection vector
scripts/harness-engine.py · prose · downgraded · subprocess.run(
cmd, shell=True
Popular HTTP library — network access
references/role-shawujing-pro.md · prose · downgraded · axios
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.