anthropic-style-diagram
Pass
Audited by Gen Agent Trust Hub on Jul 28, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill depends on
playwright, a well-known and trusted browser automation library used here for rendering SVG content. This dependency is standard for the skill's primary purpose. - [COMMAND_EXECUTION]: The
scripts/render.mjsfile executes thenpm root -gcommand to programmatically locate the Playwright installation within the execution environment. This is a common pattern for environment discovery. - [PROMPT_INJECTION]: The skill includes comprehensive safety instructions in
SKILL.mdthat explicitly direct the agent to avoid including executable elements such as<script>tags,onclickhandlers, or external links in the generated SVG files, effectively mitigating risks associated with untrusted markup. - [SAFE]: Analysis of all scripts and instructional content confirmed that no sensitive information is accessed or exfiltrated, and no persistence mechanisms are established.
Audit Metadata