lighthouse

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly accepts external HTTPS URLs (SKILL.md Step 1) and runs Lighthouse via npx @lhci/cli which fetches and audits arbitrary public web pages (Step 4), then parses the generated .lighthouseci/lhr-*.json results (Step 5) and uses those audit findings to drive recommendations and optional code changes (Steps 6–7), so untrusted third-party page content can influence the agent's analysis and actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill runs "npx @lhci/cli" at runtime (which fetches and executes code from the npm registry), so it relies on remotely fetched/executed package code as a required dependency.