NYC

mteb-leaderboard

Warn

Audited by Snyk on Feb 15, 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 explicitly instructs the agent to fetch and read live leaderboard data from public third‑party sources such as HuggingFace Spaces, GitHub repositories, blog posts, and community discussions—untrusted, user-generated content that the agent must ingest as part of its workflow, creating a clear avenue for indirect prompt injection.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 15, 2026, 09:25 PM