using-heavy-mcps

Warn

Audited by Snyk on Feb 16, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs running mcporter calls against third-party MCPs (e.g., Sanity Developer.query_documents and Brain.vault) and then pasting the filtered mcporter + jq output into chat (see "One-liners in chat" and the Sanity/Brain examples), so the agent will ingest external/user-generated content from those sources.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 16, 2026, 10:58 PM