organize-screenshots
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use shell commands such as
ls,mkdir,mv, andwcfor local file management. It correctly specifies escaping spaces in file paths to prevent shell parsing errors. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted filenames from the user's Desktop, creating a surface for potential injection.
- Ingestion points: Filenames found via
lscommand execution on the Desktop. - Boundary markers: Not explicitly defined for the filename data.
- Capability inventory: File system manipulation (
mkdir,mv) and command execution. - Sanitization: The skill provides specific instructions to escape special characters like spaces in filenames, mitigating common injection risks.
Audit Metadata