deepsearch-service

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: 0.90). The engine posts user-generated queries to an external WebSearch endpoint (self.websearch_url -> POST /search) and ingests returned page URLs and main_content into collected_info which is summarized and fed into LLM prompts (QUERY_DECOMPOSITION_PROMPT, SUFFICIENCY_CHECK_PROMPT, REPORT_SYNTHESIS_PROMPT), so it clearly consumes untrusted public web content that the agent reads and interprets.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 16, 2026, 12:23 AM