deepagents-guide

Warn

Audited by Snyk on Feb 24, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 0.70). The SKILL.md includes a "researchSubagent" example that lists a tool named "internetSearch", indicating the agent may fetch and read open web/public content (third‑party webpages) as part of its workflow, which could allow indirect prompt injection.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 24, 2026, 10:16 PM