findclient

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

• Third-party content exposure detected (high risk: 0.85). The skill performs runtime web searches and extracts free-form text from outsider-authored sources (e.g., LinkedIn pages, Glassdoor quotes, industry directories, government registries, conference sponsor pages) in Phase 3, which the agent then uses to generate the prospect report and thus feeds that outsider text into the LLM context for scoring/summarization.