mrt-rerouter-skill

Pass

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill interacts exclusively with official Singapore government API endpoints (data.gov.sg and onemap.gov.sg) to retrieve transit and environmental data.
  • [COMMAND_EXECUTION]: The skill includes a CLI tool (mrt_rerouter.py) for processing transit queries. This tool uses standard argument parsing and performs intended transit calculation logic.
  • [EXTERNAL_DOWNLOADS]: The skill fetches public datasets (CSV/GeoJSON) from official government sources via a multi-step initiate-and-poll mechanism. These downloads are limited to data for transit and weather analysis.
  • [REMOTE_CODE_EXECUTION]: The unit tests utilize importlib for dynamic script loading to facilitate testing. This usage is restricted to the local development and testing environment and does not involve remote code execution from untrusted sources.
  • [SAFE]: Caching logic stores API responses in a local directory (~/.hermes/cache/) to minimize network load, following standard performance practices.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 25, 2026, 04:52 AM
Security Audit — agent-trust-hub — mrt-rerouter-skill