sales-xpoz
Warn
Audited by Snyk on Jun 14, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.85). Outsider free text can enter the LLM context via MCP tool calls that return scraped social content (e.g.,
getTwitterPostsByKeywords/getTwitterPostComments/ Reddit post/comment bodies) authored by third parties, which the agent then reads to answer.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata