sentry-automation

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

• Third-party content exposure detected (high risk: 0.75). The required workflow calls Rube MCP tools to fetch Sentry issue/event details (outsider-authored free text from Sentry users/stack traces) which the agent then reads into its LLM context via the tool responses (e.g., issue details, event details with stack traces).

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs the client to add and contact the external MCP server https://rube.app/mcp at runtime (used with RUBE_SEARCH_TOOLS to fetch current tool schemas), which can directly control the agent's available tools and behavior.