sentry-debug-issue

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

• Third-party content exposure detected (high risk: 0.85). SKILL.md instructs the agent to use authenticated Sentry MCP tools like search_issues / execute_sentry_tool to pull full issue/event context (exception messages, breadcrumbs, request bodies, stack frames, etc.), which are explicitly “attacker-controllable” and thus are outsider-authored free-text that will be inserted into the agent’s LLM context via the MCP tool outputs at runtime.