openclaw-health-check
Pass
Audited by Gen Agent Trust Hub on Apr 8, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes common system utilities (such as
df,du,find,grep, andawk) and local service managers (pm2,openclaw) to gather operational metrics and service status information from the local environment. - [REMOTE_CODE_EXECUTION]: Although automated scans flagged a pipe to
python3, the implementation uses a static command-line string to parse JSON data from a local health endpoint (http://localhost:18789/health). This approach is used for data processing rather than executing remote script content. - [SAFE]: The skill accesses local session log files in
~/.openclaw/to calculate cost estimates and activity levels. This data ingestion is focused on generating the diagnostic report and does not involve exfiltration to external domains.
Audit Metadata