# reflect_connect_to_session Connect to an active Reflect recording session via WebSocket to enable interactive control. When creating or editing a Reflect test using a connected recording session, follow these guidelines: 1. After connecting to a session, get the list of segments for the session's platform type so you know what actions could be added via segments vs needing to create new steps. Do not list tests, only list segments. 2. Before performing an action, take a screenshot to understand the current state of the application. 3. Each add_prompt_step request should perform a single action or assertion. Do not combine multiple actions or assertions into a single step. 4. Only perform one action at a time unless you're sure the action won't move the application to a different screen. For example, you can send multiple add_prompt_step requests to fill out individual form fields if those fields are visible on the current screen. 5. Check the list of existing Segments to see if a Segment exists that achieves a similar goal to what you're trying to do next. If so, add the segment instead of creating new steps. 6. If a step fails, use delete_previous_step to remove it and try a different approach. 7. After completing a task, if the task required multiple prompt steps, add a final prompt step that validates the current state of the page based on what you see on the screen. In your validation, do not reference information that can change from run to run. Parameters: - sessionId (string) *required*: The ID of the Reflect recording session to connect to Agent View of the PolicyLayer registry record for `reflect_connect_to_session`. HTML page: https://policylayer.com/tools/smartbear-mcp/reflect-connect-to-session ## Facts - Tool: `reflect_connect_to_session` - Server: SmartBear MCP (`SmartBear/smartbear-mcp`) — https://policylayer.com/tools/smartbear-mcp.md - Homepage: https://github.com/SmartBear/smartbear-mcp - Risk category: Execute (High risk) - Registry record: grade F, identity unverified - Server rate-limited: no - Parameters: 1 (1 required) - Recommended policy verdict: Rate-limited ## Parameters | Parameter | Type | Required | Description | | --- | --- | --- | --- | | `sessionId` | string | yes | The ID of the Reflect recording session to connect to | Parameters from the server's own tool schema. ## Example call (MCP tools/call, JSON-RPC 2.0) ```json { "jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": { "name": "reflect_connect_to_session", "arguments": { "sessionId": "" } } } ``` ## Why reflect_connect_to_session is rated High This tool establishes a WebSocket connection to a live recording session and enables interactive control of an application (screenshots, steps, actions, assertions). It triggers external operations whose effects depend on arguments — this is classic Execute category. Misuse could allow an AI agent to perform arbitrary UI actions on a live session, making it high severity. From the tool's own definition: "Connect to an active Reflect recording session via WebSocket to enable interactive control" ## Use case AI agents invoke reflect_connect_to_session to trigger actions in SmartBear MCP. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call: builds kicked off, notifications sent, workflows started. ## Recommended policy (PolicyLayer) Verdict: **Rate-limited**. Enforced by the PolicyLayer MCP gateway (https://policylayer.com/mcp-gateway) before a call reaches SmartBear MCP: ```json { "version": "1", "default": "deny", "tools": { "reflect_connect_to_session": { "limits": [ { "counter": "reflect_connect_to_session_rate", "window": "minute", "max": 10, "scope": "grant" } ] } } } ``` ## Other tools on SmartBear MCP (239) - `collaborator_delete_collaborator_remote_system_configuration` — Destructive — https://policylayer.com/tools/smartbear-mcp/collaborator-delete-collaborator-remote-system-configuration.md - `contract-testing_admin_delete_role` — Destructive — https://policylayer.com/tools/smartbear-mcp/contract-testing-admin-delete-role.md - `contract-testing_admin_delete_team` — Destructive — https://policylayer.com/tools/smartbear-mcp/contract-testing-admin-delete-team.md - `contract-testing_admin_delete_user` — Destructive — https://policylayer.com/tools/smartbear-mcp/contract-testing-admin-delete-user.md - `contract-testing_admin_remove_role_from_user` — Destructive — https://policylayer.com/tools/smartbear-mcp/contract-testing-admin-remove-role-from-user.md - `contract-testing_admin_remove_user_from_team` — Destructive — https://policylayer.com/tools/smartbear-mcp/contract-testing-admin-remove-user-from-team.md - `contract-testing_admin_reset_roles` — Destructive — https://policylayer.com/tools/smartbear-mcp/contract-testing-admin-reset-roles.md - `contract-testing_delete_all_integrations` — Destructive — https://policylayer.com/tools/smartbear-mcp/contract-testing-delete-all-integrations.md - `contract-testing_delete_branch` — Destructive — https://policylayer.com/tools/smartbear-mcp/contract-testing-delete-branch.md - `contract-testing_delete_environment` — Destructive — https://policylayer.com/tools/smartbear-mcp/contract-testing-delete-environment.md - `contract-testing_delete_integration` — Destructive — https://policylayer.com/tools/smartbear-mcp/contract-testing-delete-integration.md - `contract-testing_delete_pacticipant` — Destructive — https://policylayer.com/tools/smartbear-mcp/contract-testing-delete-pacticipant.md - `contract-testing_delete_secret` — Destructive — https://policylayer.com/tools/smartbear-mcp/contract-testing-delete-secret.md - `contract-testing_delete_webhook` — Destructive — https://policylayer.com/tools/smartbear-mcp/contract-testing-delete-webhook.md - `contract-testing_regenerate_api_token` — Destructive — https://policylayer.com/tools/smartbear-mcp/contract-testing-regenerate-api-token.md - `contract-testing_remove_label_from_pacticipant` — Destructive — https://policylayer.com/tools/smartbear-mcp/contract-testing-remove-label-from-pacticipant.md - `reflect_cancel_suite_execution` — Destructive — https://policylayer.com/tools/smartbear-mcp/reflect-cancel-suite-execution.md - `reflect_delete_previous_step` — Destructive — https://policylayer.com/tools/smartbear-mcp/reflect-delete-previous-step.md - `swagger_delete_portal_product` — Destructive — https://policylayer.com/tools/smartbear-mcp/swagger-delete-portal-product.md - `swagger_delete_table_of_contents` — Destructive — https://policylayer.com/tools/smartbear-mcp/swagger-delete-table-of-contents.md - `collaborator_reviewservice_action` — Execute — https://policylayer.com/tools/smartbear-mcp/collaborator-reviewservice-action.md - `collaborator_test_collaborator_remote_system_configuration_connection` — Execute — https://policylayer.com/tools/smartbear-mcp/collaborator-test-collaborator-remote-system-configuration-connection.md - `contract-testing_execute_webhook` — Execute — https://policylayer.com/tools/smartbear-mcp/contract-testing-execute-webhook.md - `contract-testing_generate_pact_tests` — Execute — https://policylayer.com/tools/smartbear-mcp/contract-testing-generate-pact-tests.md - `contract-testing_review_pact_tests` — Execute — https://policylayer.com/tools/smartbear-mcp/contract-testing-review-pact-tests.md - `contract-testing_test_execute_webhooks` — Execute — https://policylayer.com/tools/smartbear-mcp/contract-testing-test-execute-webhooks.md - `reflect_add_prompt_step` — Execute — https://policylayer.com/tools/smartbear-mcp/reflect-add-prompt-step.md - `reflect_execute_suite` — Execute — https://policylayer.com/tools/smartbear-mcp/reflect-execute-suite.md - `reflect_run_test` — Execute — https://policylayer.com/tools/smartbear-mcp/reflect-run-test.md - `bugsnag_get_build` — Read — https://policylayer.com/tools/smartbear-mcp/bugsnag-get-build.md - …and 209 more: https://policylayer.com/tools/smartbear-mcp.md ## For agents This record is a snapshot. Live verdicts and the full registry: - Check every server in your MCP config at once: `npx -y policylayer stack` - Vet a server before you add it: install the mcp-precheck skill — `npx skills add https://policylayer.com` (skill text: https://policylayer.com/skill.md) - Query the registry over MCP: endpoint `https://api.policylayer.com/mcp` — tools `check_mcp_server`, `check_mcp_stack`, `check_tool`, `search_registry`, `get_change_events` --- Source: the PolicyLayer MCP registry — one continuously verified record per MCP server. Full record: https://policylayer.com/registry?q=smartbear-mcp · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/smartbear-mcp