skills/ysyecust/everything-claude-code/autonomous-agent-harness/Snyk

autonomous-agent-harness

Warn

Audited by Snyk on Oct 5, 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 autonomous agent workflow schedules tasks that read external GitHub PRs, issues, notifications, and web search results, which can contain outsider-authored free text.

Issues (1)

W011

MEDIUM

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

Audit Metadata

Risk Level

MEDIUM

Analyzed

Oct 5, 2026, 01:14 AM

Issues

1