continuous-improvement
Warn
Audited by Snyk on Apr 8, 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's SKILL.md explicitly describes automated scans of public sources (GitHub Search API, npm Registry, and community sources like X/Twitter, Hacker News, Reddit, Discord) that write proposal JSON files which the evaluator and security agents read and act on, so untrusted, user-generated third‑party content can materially influence approvals and implementation.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata