develop-web-game

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill mentions checking for npx and installing the @playwright/mcp package. This is a well-known package from a trusted organization used for browser automation and model context protocol integration.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external web pages via browser console logs and JSON state output, which represents a potential surface for indirect prompt injection.\n
  • Ingestion points: window.render_game_to_text() output and console error trackers in scripts/web_game_playwright_client.js.\n
  • Boundary markers: None; the agent is instructed to process the raw JSON and logs directly.\n
  • Capability inventory: The skill uses fs.writeFileSync to save artifacts and executes a Node.js script to manage browser interactions.\n
  • Sanitization: None; data is captured from the browser environment and persisted directly to local files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 12:17 AM
Security Audit — agent-trust-hub — develop-web-game