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
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 8, 2026, 02:29 PM
Issues
1
Security Audit — snyk — continuous-improvement