playwright-network-analyzer

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill instructs capturing and analyzing "all HTTP requests" and request/response payloads (including authentication flows) with no guidance to redact headers or tokens, so the LLM would likely receive and could output sensitive values (cookies, Bearer tokens, API keys) verbatim.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The skill's SKILL.md workflow explicitly navigates to arbitrary web pages (Step 1 "Navigate to: https://app.example.com/dashboard") and captures/reads network requests and console messages via browser_network_requests and read_console_messages (Step 4 and Notes), meaning it ingests untrusted public third‑party content (external pages, CDNs, APIs) and uses those responses to drive analysis and recommendations.