trust-but-verify

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly asks the agent to request local/test credentials (email/password), save them to project memory (reference_local_auth.md), and use them to fill login forms (i.e., include the secret values verbatim in browser-fill actions and saved files), which requires the LLM to handle and output secrets directly.