researching

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill requires reading files fully and producing research documents with concrete code references and snippets (file:line, detailed findings), which could force the model to reproduce any secrets found in the codebase verbatim, creating an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.70). The skill's Research Process (Step 3) explicitly instructs spawning a "web-search-researcher" agent "for external documentation," which means it can fetch and ingest open/public third-party web content that could influence research decisions and subsequent actions.