Teradata MCP Server

Instructions

Implementation Reference

• src/teradata_mcp_server/tools/base/base_tools.py:248-283 (handler)

  The handler function that implements the core logic of the base_tablePreview tool. It executes 'SELECT TOP 5 * FROM table' to fetch a sample of data, extracts column information, builds metadata, and returns a formatted response using create_response.

  def handle_base_tablePreview(conn: TeradataConnection, table_name: str, database_name: str | None = None, *args, **kwargs):
      """
      This function returns data sample and inferred structure from a database table or view via SQLAlchemy, bind parameters if provided (prepared SQL), and return the fully rendered SQL (with literals) in metadata.
  
      Arguments:
        table_name - table or view name
        database_name - Database name
  
      Returns:
        ResponseType: formatted response with query results + metadata
      """
      logger.debug(f"Tool: handle_base_tablePreview: Args: tablename: {table_name}, databasename: {database_name}")
  
      if database_name is not None:
          table_name = f"{database_name}.{table_name}"
      with conn.cursor() as cur:
          cur.execute(f'select top 5 * from {table_name}')
          columns = cur.description
          sample = rows_to_json(cur.description, cur.fetchall())
  
          metadata = {
              "tool_name": "base_tablePreview",
              "database": database_name,
              "table_name": table_name,
              "columns": [
                  {
                      "name": c[0],
                      "type": c[1].__name__ if hasattr(c[1], '__name__') else str(c[1]),
                      "length": c[3]
                  }
                  for c in columns
              ],
              "sample_size": len(sample)
          }
          logger.debug(f"Tool: handle_base_tablePreview: metadata: {metadata}")
          return create_response(sample, metadata)

• src/teradata_mcp_server/app.py:272-281 (registration)

  Dynamic registration loop that discovers all 'handle_*' functions from loaded tool modules and registers them as MCP tools. The tool 'base_tablePreview' is registered from 'handle_base_tablePreview' by stripping the 'handle_' prefix and using the function's docstring as description.

  all_functions = module_loader.get_all_functions()
  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}")