Teradata MCP Server

Instructions

Implementation Reference

• src/teradata_mcp_server/tools/qlty/qlty_tools.py:182-217 (handler)

  The handler function that implements the core logic for the qlty_univariateStatistics tool. It executes a Teradata query using TD_UnivariateStatistics to compute all univariate statistics for the specified column in the table.

  def handle_qlty_univariateStatistics(
      conn: TeradataConnection,
      database_name: str | None,
      table_name: str,
      column_name: str,
      *args,
      **kwargs
  ):
      """
      Get the univariate statistics for a table.
  
      Arguments:
        database_name - name of the database
        table_name - table name to analyze
        column_name - column name to analyze
  
      Returns:
        ResponseType: formatted response with query results + metadata
      """
      logger.debug(f"Tool: handle_qlty_univariateStatistics: Args: table_name: {database_name}.{table_name}, column_name: {column_name}")
  
      if database_name is not None:
              table_name = f"{database_name}.{table_name}"
      with conn.cursor() as cur:
          rows = cur.execute(f"select * from TD_UnivariateStatistics ( on {table_name} as InputTable using TargetColumns ('{column_name}') Stats('ALL')) as dt ORDER BY 1,2")
          data = rows_to_json(cur.description, rows.fetchall())
          metadata = {
              "tool_name": "qlty_univariateStatistics",
              "database_name": database_name,
              "table_name": table_name,
              "column_name": column_name,
              "stats_calculated": ["ALL"],
              "rows": len(data)
          }
          logger.debug(f"Tool: handle_qlty_univariateStatistics: Metadata: {metadata}")
          return create_response(data, metadata)

• src/teradata_mcp_server/app.py:273-281 (registration)

  The dynamic registration code that discovers and registers all handle_* functions (including handle_qlty_univariateStatistics as 'qlty_univariateStatistics') from loaded tool modules as MCP tools, inferring schema from function signatures and using docstrings for descriptions.

  for name, func in all_functions.items():
      if not (inspect.isfunction(func) and name.startswith("handle_")):
          continue
      tool_name = name[len("handle_"):]
      if not any(re.match(p, tool_name) for p in config.get('tool', [])):
          continue
      wrapped = make_tool_wrapper(func)
      mcp.tool(name=tool_name, description=wrapped.__doc__)(wrapped)
      logger.info(f"Created tool: {tool_name}")

• src/teradata_mcp_server/tools/module_loader.py:21-25 (helper)

  ModuleLoader maps the 'qlty' prefix to the qlty tools module, enabling lazy loading of the qlty_univariateStatistics handler when qlty tools are required by the profile.

  MODULE_MAP = {
      'base': 'teradata_mcp_server.tools.base',
      'dba': 'teradata_mcp_server.tools.dba',
      'fs': 'teradata_mcp_server.tools.fs',
      'qlty': 'teradata_mcp_server.tools.qlty',

• src/teradata_mcp_server/tools/qlty/__init__.py:1-2 (registration)

  The qlty package __init__ imports all functions from qlty_tools.py, making handle_qlty_univariateStatistics available for dynamic discovery and registration.

  from .qlty_resources import *
  from .qlty_tools import *