research-grounding
Pass
Audited by Gen Agent Trust Hub on May 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides instructions and a script for verifying source credibility. It identifies necessary tools and environment variables without performing any sensitive operations or exfiltration.
- [COMMAND_EXECUTION]: Executes a local shell script (
check_prerequisites.sh) to verify system environment and tool installation. This script only performs read-only checks for thegeminibinary and theGOOGLE_API_KEYenvironment variable. - [EXTERNAL_DOWNLOADS]: References the
gemini-clitool from the official Google Gemini GitHub repository for automated grounding features. This is a reference to a well-known service and does not involve automated downloads or execution of untrusted code. - [DATA_EXFILTRATION]: While the script checks for the presence of a
GOOGLE_API_KEY, it specifically masks the value (showing only the first 8 characters) and does not transmit the key to any external location.
Audit Metadata