lead-generation

Warn

Audited by Snyk on Mar 4, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly instructs the agent to fetch and read user-generated content from public social platforms (Twitter/Instagram/Reddit) via Xpoz API calls like xpoz.getTwitterPostsByKeywords and xpoz.getRedditPostsByKeywords and then use those posts to score leads and craft outreach, exposing the agent to untrusted third-party content that could contain indirect prompt injections.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 4, 2026, 07:49 PM