kmp-imagevector-generator

Pass

Audited by Gen Agent Trust Hub on Oct 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides guidelines for fetching SVG assets from the official Tailwind Labs Heroicons GitHub repository. This process utilizes a well-known service to retrieve legitimate, open-source design resources.
  • [COMMAND_EXECUTION]: The script scripts/convert_image_to_imagevector.py uses the subprocess.run method to execute the potrace utility for monochrome image tracing. The implementation is secure, using a list of arguments rather than a shell string and operating on temporary files to prevent command injection.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (SVGs and images). It addresses this attack surface through a deterministic generation pipeline and explicit instructions (Rule 2) that forbid the agent from reading the generated Kotlin output. This prevents malformed path data from being ingested back into the agent's reasoning context.
  • [DYNAMIC_EXECUTION]: The generator script optionally imports and utilizes external libraries such as vtracer and picosvg at runtime to handle complex image processing tasks like vector tracing and stroke-to-fill normalization. These are standard Python packages used for their intended functional purposes within the toolchain.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 12, 2026, 09:39 PM