baoyu-post-to-x

Pass

Audited by Gen Agent Trust Hub on Mar 25, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill provides instructions to install the Bun runtime via a remote shell script from bun.sh, which is the standard installation method for this well-known service.
  • [COMMAND_EXECUTION]: The skill executes system-level commands to spawn Google Chrome with debugging enabled and runs platform-specific utilities (Swift, AppleScript, PowerShell, xdotool) to manage the clipboard and simulate user input. These actions are required for the skill to interact with the browser's interface.
  • [EXTERNAL_DOWNLOADS]: The scripts/md-to-html.ts component includes functionality to download remote images from URLs provided in Markdown content using standard Node.js networking modules.
  • [PROMPT_INJECTION]: The skill processes external Markdown files and media content, creating a surface for potential indirect prompt injection attacks. Ingestion points: scripts/md-to-html.ts (Markdown file), scripts/x-browser.ts (image paths), scripts/x-video.ts (video path). Boundary markers: None present. Capability inventory: Full browser automation via Chrome DevTools Protocol, execution of system scripts for clipboard management, and local file system writes. Sanitization: Uses escapeHtml for code blocks and leverages established Markdown parsing libraries.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 25, 2026, 03:07 PM
Security Audit — agent-trust-hub — baoyu-post-to-x