gws-sheets-read

Pass

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the gws command-line tool, which is a binary provided by the vendor for interacting with Google Workspace.- [PROMPT_INJECTION]: The skill presents a risk of indirect prompt injection by ingesting data from external spreadsheets into the agent's working context.
  • Ingestion points: Data is fetched from spreadsheets via the gws sheets +read command in SKILL.md.
  • Boundary markers: The skill documentation does not provide delimiters or instructions to the agent to ignore or isolate commands found within the retrieved spreadsheet data.
  • Capability inventory: While the reading tool itself is restricted, the agent context typically includes other capabilities (e.g., file system access, shell execution) that could be manipulated by instructions embedded in the spreadsheet.
  • Sanitization: No validation or sanitization of the spreadsheet content is performed before it is presented to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 25, 2026, 08:24 AM
Security Audit — agent-trust-hub — gws-sheets-read