Nexus MCP Server

The Nexus MCP Server provides an interface for working with your Autify Nexus remote workspace through the Model Context Protocol (MCP). By connecting from an MCP-compatible AI client such as Claude Desktop, Claude Code, or Cursor, you can create and edit scenarios, run test plans, and analyze execution results using natural language.

What is the Model Context Protocol (MCP)?

The Model Context Protocol is an open standard that lets AI assistants interact safely with external services and data sources. An MCP-compatible client and an MCP server (such as the one described in this article) communicate through a standardized protocol, enabling the AI to perform a wide range of operations on your behalf.

What you can do with the Nexus MCP Server

Once an AI client is connected to the Nexus MCP Server, you can perform operations such as the following:

Create, search, view, update, archive, and restore scenarios
Add, modify, reorder, and delete steps within a scenario
Create, view, update, run, cancel, archive, and restore test plans
Search execution results and inspect step-level details, including error information, screenshots, and DOM snapshots
List the labels and directories in a workspace

Available tools

The Nexus MCP Server exposes the following tools (operations that can be invoked from an AI client).

Workspaces

Tool	Description
`list_workspaces`	Lists the workspaces you have access to.
`read_directories`	Lists the directories in a workspace.
`read_labels`	Lists the labels in a workspace.

Scenarios

Tool	Description
`create_scenarios`	Creates multiple scenarios in bulk from natural-language steps.
`create_scenario`	Creates a single scenario with metadata such as labels, directories, and data tables.
`search_scenarios`	Searches for scenarios by text.
`read_scenarios`	Retrieves the details of a scenario by UUID.
`update_scenarios`	Updates scenario metadata such as labels, directories, and data.
`archive_scenarios` / `restore_scenarios`	Archives or restores scenarios.

Scenario steps

Tool	Description
`edit_scenario_steps`	Adds, modifies, or deletes steps within a scenario.
`move_scenario_steps`	Reorders steps within a scenario.
`delete_scenario_steps`	Removes steps from a scenario.

Test plans

Tool	Description
`list_test_plans`	Lists test plans (with optional filtering by name).
`read_test_plan`	Retrieves a test plan's configuration and target scenarios by UUID.
`create_test_plan`	Creates a test plan, including execution settings and schedules.
`update_test_plan`	Updates a test plan.
`execute_test_plan`	Runs a test plan.
`cancel_test_plan_runs`	Cancels test plan runs that are in progress.
`archive_test_plans` / `restore_test_plans`	Archives or restores test plans.

Execution results

Tool	Description
`search_execution_results`	Searches execution results by criteria such as status, platform, or browser.
`read_execution_results`	Retrieves the full execution results, including per-scenario results and step details.
`read_step_result`	Retrieves error details, screenshots, and browser logs for a specific step.
`read_step_result_dom_snapshot`	Retrieves the DOM snapshot (HTML) of a step before and after execution.
`archive_execution_results` / `restore_execution_results`	Archives or restores execution results.

Connecting to the Nexus MCP Server

Documentation Index