growth-automation

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The Customer Service Bot (src/automation/customerService.js) monitors mentions and DMs for incoming messages and auto-responds using templates, which introduces a vulnerability surface for processing untrusted external instructions. * Ingestion points: X/Twitter DMs and mentions. * Boundary markers: None described in the documentation to distinguish user messages from system instructions. * Capability inventory: The skill can post tweets, send DMs, follow/unfollow users, and rotate account sessions via actions.js and multiAccount.js. * Sanitization: No sanitization or validation logic is specified for the content of incoming messages before processing.
  • [DATA_EXFILTRATION]: The Link Scraper and Session Logger tools are designed to extract and export account-related data, such as all links shared by a user and detailed logs of all automated actions, into JSON or CSV files.
  • [DYNAMIC_EXECUTION]: The skill requires users to manually execute large JavaScript files (e.g., actions.js, core.js) in the browser console. This execution method grants the code full access to the user's authenticated session, cookies, and sensitive account functions on the target platform.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 08:37 AM