auto

Warn

Audited by Snyk on Feb 28, 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.md (and scripts/auto.sh) explicitly states that enabling the marker activates PostToolUse hooks that track WebSearch/WebFetch/Read/Grep and write those external research sources to claude_usage/sources.md, which means the agent will fetch and ingest untrusted public web content as part of its workflow and that content could influence subsequent actions.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 28, 2026, 11:04 PM