gramio-pick-username

Pass

Audited by Gen Agent Trust Hub on Apr 20, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The agent is instructed to execute a bundled Node.js script (check-usernames.mjs) to verify username availability. This script is transparent, contains no external dependencies, and is used for its intended purpose.
  • [DATA_EXFILTRATION]: Performs network requests to t.me (Telegram's well-known domain) to scrape public metadata. No sensitive user information or environment variables are transmitted during these checks.
  • [SAFE]: The skill processes external HTML data from Telegram profile pages, creating an indirect prompt injection surface. The risk is minimized as follows:
  • Ingestion points: Data is sourced from public t.me URLs.
  • Boundary markers: The utility returns structured JSON, providing a clear boundary between external data and the agent's context.
  • Capability inventory: Uses Bash for local script execution with standard permissions.
  • Sanitization: The scraping logic uses specific regular expressions to extract only the necessary metadata (buttons and titles), filtering out irrelevant or potentially malicious page content.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 20, 2026, 11:47 AM