a6-plugin-limit-count

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt includes configuration and CLI examples that embed plaintext Redis passwords (e.g., "redis_password": "secret") and shows JSON/YAML snippets that an agent would plausibly fill with real credentials, meaning the LLM would be expected to output secret values verbatim if provided—an exfiltration risk.