gap-finder

Warn

Audited by Snyk on Mar 9, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (high risk: 1.00). The skill's mandatory workflow and VALIDATION.md explicitly require searching and ingesting open web sources (Google, GitHub, npm/App Store/ProductHunt, Stack Overflow, Reddit, forum posts) and recording/using those URLs and posts to kill or approve ideas, so the agent will read and act on untrusted third-party, user-generated content as part of its decision loop.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 9, 2026, 10:16 PM