cowart-image-gen

Pass

Audited by Gen Agent Trust Hub on Aug 29, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill operates as an instructional set for the agent to use predefined Cowart Model Context Protocol (MCP) tools (e.g., get_cowart_selection, insert_cowart_image). These tools reside in the platform environment, not within the skill itself.
  • [SAFE]: The workflow involves reading and writing canvas state to a local project directory (projectDir) and managing assets within a structured project folder (canvas/pages/.../assets/). This is standard functionality for the intended use case.
  • [SAFE]: The skill references an internal imagegen capability for generating bitmaps and includes logic to resolve local file paths using $CODEX_HOME. This relies on the agent's native environment rather than external remote code execution.
  • [SAFE]: There are no detected obfuscation techniques, credential exposures, or attempts to bypass safety guidelines. Instructions focus on aspect ratio calculations, tldraw shape metadata, and asset management.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 29, 2026, 07:00 PM
Security Audit — agent-trust-hub — cowart-image-gen