gws-drive
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on the 'gws' binary to perform operations via the shell. This tool is required for interacting with the Google Workspace APIs as described in the metadata.
- [PROMPT_INJECTION]: Risk of indirect prompt injection exists because the skill ingests untrusted data and has exploitable capabilities.
- Ingestion points: Files contents (
files.get,files.download), comments (comments.list), and replies (replies.list) are fetched from Google Drive (SKILL.md). - Boundary markers: None identified. There are no instructions for the agent to ignore embedded commands within the fetched content.
- Capability inventory: The skill can create files (
files.create), update content (files.update), and manage sharing permissions (permissions.create,permissions.delete) (SKILL.md). - Sanitization: No sanitization or validation of the retrieved content is mentioned before it is processed by the agent.
Audit Metadata