opencode
Warn
Audited by Snyk on Mar 1, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.80). Yes — SKILL.md explicitly documents GitHub integration (e.g., "opencode pr 123" / "Fetch and checkout a PR") and OAuth/MCP login flows, which cause the agent to fetch and ingest untrusted, user-generated content (public PRs/web pages) that can directly influence subsequent tool actions.
Audit Metadata