dspy-prompting

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: 0.80). The skill's examples include a ReAct search_tool labeled "Search Wikipedia" and a MultiHopQA that calls dspy.Retrieve and then concatenates retrieved "passages" for the model to read, which clearly ingests public third‑party content (Wikipedia/public web) into the agent's workflow.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 15, 2026, 09:26 PM