agency-socials

Pass

Audited by Gen Agent Trust Hub on Jun 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Uses AppleScript via osascript to save images from the system clipboard to the user's desktop directory for inclusion in generated assets.
  • [COMMAND_EXECUTION]: Executes a Node.js script using the Playwright library to render HTML templates into image files.
  • [EXTERNAL_DOWNLOADS]: Fetches typography assets from trusted and well-known services including Google Fonts and the JSDelivr CDN.
  • [PROMPT_INJECTION]: The skill ingests untrusted user input (speaker names, headlines) and interpolates it into HTML templates. This creates a surface for indirect prompt injection, as no boundary markers or sanitization steps are defined for the process. The skill possesses capabilities for file-writing and shell execution which could be influenced by malicious input during the rendering stage.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 12, 2026, 10:20 PM
Security Audit — agent-trust-hub — agency-socials