playwright-oauth-code-capture

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). This skill instructs capturing and logging authorization codes and shows embedding client_id/client_secret and the captured code directly into requests/code (including console output), which requires handling secret values verbatim and risks exfiltration.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). This content explicitly teaches how to intercept OAuth authorization codes in a browser automation context (capturing and exchanging auth codes), which is a direct credential-theft technique with high potential for abuse even if it has legitimate uses.

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

• Third-party content exposure detected (high risk: 0.80). The skill navigates to public OAuth provider pages (e.g., https://accounts.google.com/...) and intercepts the redirect URL from localhost, ingesting third-party redirect data (the captured URL/code) which is parsed and used to drive token exchange and subsequent actions.