skills/reason-machines/ai-agent-skills/agent-reach-internet-access/Snyk

agent-reach-internet-access

Warn

Audited by Snyk on Oct 8, 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 workflow involves reading GitHub issues and social feeds (Twitter, Reddit, Weibo, V2EX), which allows outsider-authored free text to be ingested into the agent.

Issues (1)

W011

MEDIUM

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

Audit Metadata

Risk Level

MEDIUM

Analyzed

Oct 8, 2026, 11:38 PM

Issues

1