a6-plugin-wolf-rbac

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The documentation includes concrete examples that embed passwords and rbac tokens verbatim in request bodies and Authorization headers (curl examples and token format), which would require an LLM to handle or output secret values directly if used as-is, creating an exfiltration risk.

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

• Third-party content exposure detected (medium risk: 0.65). At runtime the plugin calls the (external) Wolf RBAC server and ingests its response JSON (e.g., user_info/nickname and injected headers) into the request/LLM-handling context; Wolf server outputs are outsider-authored free text from outside the operating user’s control.