favicon
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell commands (
magick,cp) using a user-supplied argument ($1) for the image path. Although the argument is double-quoted in the instructions, there is a risk of command injection if the input contains shell metacharacters and the agent does not perform strict sanitization. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from untrusted local project files, creating a potential attack surface for indirect prompt injection.
- Ingestion points: Reads the
namefield fromsite.webmanifest,package.json, andconfig/application.rb(Step 3). - Boundary markers: None. The skill does not use delimiters or instructions to ignore potential commands within these data sources.
- Capability inventory: The skill has the capability to execute shell commands via
magickand write/update files across the project directory (Steps 5, 6, and 7). - Sanitization: The skill only performs title-case conversion for the extracted app name but lacks explicit escaping or validation before interpolating this data into manifest files or HTML tags.
Audit Metadata