EvalView

Instructions

Implementation Reference

• evalview/mcp_server.py:429-430 (handler)

  The handler logic for 'list_tests', which maps the MCP tool call to the 'evalview golden list' CLI command.

  elif name == "list_tests":
      cmd = ["evalview", "golden", "list"]

• evalview/mcp_server.py:147-157 (schema)

  The schema definition for the 'list_tests' tool, defining its name, description, and input parameters.

  {
      "name": "list_tests",
      "description": (
          "List all available golden baselines in this EvalView project. "
          "Shows test names, variant counts, and when each baseline was last updated."
      ),
      "inputSchema": {
          "type": "object",
          "properties": {},
      },
  },

• evalview/commands/mcp_cmd.py:279-280 (registration)

  A comment in the code acknowledging 'list_tests' as an exposed MCP tool.

  Exposes run_check, run_snapshot, and list_tests as MCP tools so you can
  run regression checks inline without switching to a terminal.