google-sheets-automation

Pass

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on executing local Python scripts (scripts/auth.py and scripts/sheets.py) to perform all operations, including authentication and spreadsheet manipulation.
  • [EXTERNAL_DOWNLOADS]: The documentation mentions that tokens are automatically refreshed using a remote "Google's cloud function," which introduces a dependency on an external network resource for maintaining authentication state.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes untrusted data from external Google Sheets and presents it to the agent context.
  • Ingestion points: Data enters the agent context through scripts/sheets.py via commands like get-text, get-range, and find (documented in SKILL.md).
  • Boundary markers: There are no documented boundary markers or instructions to the agent to ignore embedded commands within the spreadsheet data.
  • Capability inventory: The skill includes write capabilities such as update-range, append-rows, and batch-update (documented in SKILL.md), which could be abused if an attacker injects instructions into a spreadsheet the agent reads.
  • Sanitization: No sanitization or validation of the spreadsheet content is mentioned before the data is processed or returned to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 14, 2026, 09:52 AM