skills/rheinmir/setup/agent-reach/Snyk

agent-reach

Warn

Audited by Snyk on Sep 15, 2026

Risk Level: MEDIUM

Full Analysis

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

Third-party content exposure detected (medium risk: 0.30). The skill instructs the agent to search and read social media feeds, GitHub issues/repositories, Reddit, and web pages, which exposes the agent to outsider-authored free text via active search and ingestion.

Issues (1)

W011

MEDIUM

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

Audit Metadata

Risk Level

MEDIUM

Analyzed

Sep 15, 2026, 04:42 PM

Issues

1