image

Pass

Audited by Gen Agent Trust Hub on Aug 9, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides practical examples of shell commands for image optimization using common utilities such as cwebp, mogrify (part of the ImageMagick suite), and jpegoptim. It also includes a curl command used for inspecting image assets on a website. These tools are appropriate and standard for the stated purpose of visual asset production.
  • [PROMPT_INJECTION]: The skill instructions include a step to read local project files (e.g., .agents/product-marketing.md) to gather context. While reading external data can present a surface for indirect prompt injection, this is a standard design pattern for personalizing agent responses based on existing project documentation. The skill treats this as a source of instruction rather than untrusted data, which is typical for development-oriented agents.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 9, 2026, 08:50 AM