linkedin-post

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/linkedin-post.js uses child_process.execSync to start the browser using the clawdbot CLI tool. It also uses CDP Runtime.evaluate to execute JavaScript for browser automation tasks.
  • [EXTERNAL_DOWNLOADS]: The skill requires the ws npm package for WebSocket-based communication with the browser's DevTools Protocol.
  • [PROMPT_INJECTION]: The skill handles user-provided post content which is injected into an active browser session, creating an indirect prompt injection surface.
  • Ingestion points: The --text argument in scripts/linkedin-post.js accepts arbitrary user input.
  • Boundary markers: No specific delimiters or instructions to ignore embedded content are provided in the agent instructions.
  • Capability inventory: The skill can execute shell commands (execSync), perform network requests (fetch), and execute scripts in a browser context (Runtime.evaluate).
  • Sanitization: The script performs HTML entity encoding and JavaScript string escaping on user input before injecting it into the LinkedIn editor.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 02:44 PM
Security Audit — agent-trust-hub — linkedin-post