xhs-md2img
Warn
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- EXTERNAL_DOWNLOADS (MEDIUM): The skill uses
npx -yto run scripts. This command can download and execute packages from the npm registry without manual verification, posing a potential supply chain risk. - COMMAND_EXECUTION (LOW): The skill executes external browser processes through Playwright to render content.
- DATA_EXFILTRATION (LOW): The logic for converting relative image paths to
file://URLs allows for potential local data exposure. A malicious markdown file could be crafted to render sensitive system files into the output images. - PROMPT_INJECTION (LOW): The skill lacks sanitization for the input Markdown, making it vulnerable to indirect prompt injection. Ingestion point: The
<markdown-file>CLI argument. Boundary markers: None present. Capability inventory: npx execution, file system read/write, and browser spawning. Sanitization: None mentioned in rendering logic.
Audit Metadata