gws-shared
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates interaction with Google Workspace APIs where the agent ingests untrusted data from external sources. \n
- Ingestion points: Content retrieved via gws CLI commands such as email messages, document content, and spreadsheet data (e.g., in SKILL.md under Usage). \n
- Boundary markers: The instructions do not specify explicit delimiters or instructions to ignore embedded commands when processing content from these external sources. \n
- Capability inventory: The skill provides capabilities to read and write files (using --output and --upload flags) and perform network operations via the CLI. \n
- Sanitization: While the CLI supports a --sanitize flag, the instructions do not mandate its use for all untrusted data ingestion. \n- [EXTERNAL_DOWNLOADS]: The skill references the official GitHub repository for the Google Workspace CLI (googleworkspace/cli) for installation and issue tracking.
Audit Metadata