films-search

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the agent to extract and return pan "extractCode" values (share passwords) verbatim in the JSON/response, which requires handling and outputting secret/password-like values.

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

• Third-party content exposure detected (high risk: 0.85). At runtime, the skill uses the web-search skill to fetch public web search result snippets and then passes extracted page URLs into deep_extract.py, which uses cloudscraper to fetch outsider-authored HTML from those arbitrary third-party pages; the script then reads that fetched HTML as text and extracts title, quality, and extractCode from it into the JSON returned to the agent (LLM context).