answeroverflow

Warn

Audited by Snyk on Mar 24, 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.md explicitly instructs the agent to search and fetch public, user-generated Discord conversations from AnswerOverflow (e.g., https://www.answeroverflow.com/m/ and MCP tools like get_thread_messages), so the agent will ingest untrusted third‑party content that can influence subsequent decisions.

Issues (1)

W011
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 24, 2026, 12:24 AM
Issues
1