ClawAudit verdict
dmp-persona-insight-5-2-1-clean-1
dmp-persona-insight
DMP user persona analytics skill that calls the disclosed Mingdata open platform API using user-supplied AK/SK credentials; credential_access is for the user's own API keys to a named third-party service.
⚠ 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.
Permission integrity
credential_access
Findings (4)
Possible hardcoded credential
references/api-integration.md · code · api_key = "sk-xxx-xxx-xxx
Accesses OpenClaw config/secrets directly
references/api-integration.md · code · ~/.openclaw/openclaw.json
POSTs data to external URL
references/api-integration.md · code · .post(
"https://
Python os.environ.get — reads environment variable
references/api-integration.md · code · os.environ.get(
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class B). 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.
credential_access Is this flag fair?
Thanks — recorded.