hr-network-analyst

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly ingests and analyzes public, user-generated third‑party content—e.g., Semantic Scholar/arXiv API calls, GitHub API, Twitter/X and Reddit analysis, conference website scraping, and LinkedIn/Proxycurl scraping/enrichment—and the agent is expected to read and interpret that content as part of its network reconstruction workflow, creating a clear vector for indirect prompt injection.