The Agent Skills Directory

Prompt Injection (LOW): The skill is vulnerable to indirect prompt injection by processing untrusted data from Google Drive. 1. Ingestion points: Data enters the agent via scripts/drive.py search, list, and find-folder commands. 2. Boundary markers: None detected to distinguish between system instructions and file data. 3. Capability inventory: The skill allows high-impact actions including upload, move, and trash (delete) operations. 4. Sanitization: No evidence of data sanitization or instruction filtering is mentioned in the documentation.
Command Execution (SAFE): The skill executes its own internal Python scripts (scripts/auth.py, scripts/drive.py) to perform its functions.
Credentials Security (SAFE): The skill follows security best practices by using the system keyring for token storage instead of plain-text files.

google-drive