sentry-debug-issue

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

• Third-party content exposure detected (medium risk: 0.30). In “Sentry — Debug an Issue”, the agent uses the Sentry MCP tools search_issues / issue-details (via search_sentry_tools→execute_sentry_tool) to pull full issue/event context and then ingests attacker-controlled fields like exception messages, breadcrumbs, request bodies, and stack frames into the LLM at runtime.