recipe-log-deal-update
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read data from a Google Sheets spreadsheet using the
gws sheets +readcommand. This creates a vulnerability surface where instructions hidden within the spreadsheet could influence the agent's subsequent actions. - Ingestion points: Data enters the context via
gws sheets +readin Step 2 ofSKILL.md. - Boundary markers: The instructions do not define clear delimiters or warnings to treat the read data as untrusted content.
- Capability inventory: The skill has the capability to write back to the spreadsheet using
gws sheets +appendin Step 3. - Sanitization: There are no explicit instructions to sanitize or validate the content retrieved from the spreadsheet before the agent continues its workflow.
Audit Metadata