google-maps-search
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a bash script (
search_google_maps.sh) to handle API interactions. The script performs extensive validation on all user-supplied inputs, including type checking (integer/float/string) and range enforcement for parameters like zoom level, coordinates, and ratings. - [DATA_EXFILTRATION]: The skill communicates exclusively with the Frevana API (
ai-factory.frevana.comby default). It uses temporary files for payload processing and ensures they are deleted upon completion using a shell trap. Network operations are restricted to the primary API purpose. - [CREDENTIALS_UNSAFE]: The skill follows secure practices for credential management. It accepts a bearer token (
FREVANA_TOKEN) from environment variables or a CLI flag, and provides an interactive prompt for secure entry if the token is missing in an interactive session. No credentials are hardcoded. - [DYNAMIC_EXECUTION]: The skill uses Python for JSON serialization and validation. The Python code is passed via stdin to a
python3process, ensuring that the script logic is static and not influenced by untrusted external data other than through environment variables that are rigorously sanitized by the wrapper shell script.
Audit Metadata