sentry

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s runtime path calls the Sentry REST API via the bundled scripts/sentry_api.py, and the LLM ingests the returned issue/event JSON fields (e.g., titles, messages, release/culprit/url) which are outsider-authored content from Sentry projects, creating an indirect prompt-injection risk.