sentry-instrument

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

• Third-party content exposure detected (medium risk: 0.30). In SKILL.md Step 4 and prerequisite handling, the workflow treats MCP-returned event/issue data as untrusted and surfaces MCP issue title/error message/URL, so runtime ingestion includes outsider-authored free text originating from users’ app inputs that become Sentry event/issue fields.