scan-large-images
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a bundled shell script (
scanner.sh) or Python script (scanner.py) to traverse the project directory and invoke system tools likesips(macOS) for image analysis. - [COMMAND_EXECUTION]: Uses the
playcommand (SoX utility) to generate an audible notification upon completion of the scanning task. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data in the form of local filenames and image metadata, which could potentially contain malicious instructions intended to influence the agent's report generation.
- Ingestion points:
os.walkinscanner.pyandfindcommand inscanner.sh(line 117). - Boundary markers: No explicit delimiters or instructions are provided to ensure the agent treats filesystem data purely as non-executable text.
- Capability inventory: The agent has access to shell environments and Python execution with filesystem read permissions.
- Sanitization: Standard string formatting is used for reporting; however, no specific sanitization for prompt injection is implemented.
Audit Metadata