competitor-skill

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the agent to prompt the user for their Patsnap API key and then write that user-provided key verbatim into a .env file, requiring the LLM to handle and output a secret value.

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

• Third-party content exposure detected (medium risk: 0.30). source: scripts/fetch_competitor_report.py runtime path main() → search(raw_query, limit) calls Patsnap /search/patent/query-search-patent with query_text=raw_query derived directly from outsider user input, then ingests untrusted free-text fields from the responses (e.g., patent titles/summaries/fig URLs and literature abstracts/AI summaries) into render_report() and render_html.py for output.