recipe-find-large-files
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a command-line tool named 'gws' to interact with Google Workspace Drive. It provides specific command patterns for listing files with metadata and deleting files by ID.
- [INDIRECT_PROMPT_INJECTION]: The skill involves processing external data in the form of Google Drive file lists. This creates an attack surface where a maliciously named file could attempt to influence the agent's behavior when it parses the tool's output.
- Ingestion points: The output of the
gws drive files listcommand specified inSKILL.md. - Boundary markers: No explicit delimiters or instructions to ignore embedded content are included in the recipe.
- Capability inventory: The skill utilizes the
gws drive files deletecapability. - Sanitization: The recipe does not specify any sanitization or validation steps for the filenames before they are reviewed by the agent.
Audit Metadata