autoresearch-fleet

Warn

Audited by Snyk on Apr 17, 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 orchestrator (scripts/launch.sh → generated orchestrator.sh) explicitly enables plateau-triggered web search—injecting "You MUST search the web before coding" into the agent prompt and passing codex flag web_search="live" or adding claude "--tools default"—so the agent will fetch and act on open/public web content (search results) which can materially influence commits and subsequent tool use.

Issues (1)

W011
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 17, 2026, 11:08 AM
Issues
1