hive-mind-advanced

Warn

Audited by Snyk on Mar 3, 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 skill explicitly integrates with GitHub (e.g., "npx claude-flow hive-mind spawn 'Analyze repo quality' --objective 'owner/repo'" and PR review examples) which requires fetching and interpreting arbitrary public repository/PR content that can influence decisions and subsequent actions, exposing the agent to untrusted third-party input.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 3, 2026, 05:44 AM