Law Scrapper MCP

Instructions

Implementation Reference

• app.py:368-372 (registration)

  Registration of the 'get_publisher_details' tool using the @app.tool decorator, specifying name, description, and tags.

  @app.tool(
      name="get_publisher_details",
      description="Get detailed information about a specific legal publisher including act counts and publication timeline.",
      tags={"metadata", "publishers", "reference", "sources"}
  )

• app.py:373-414 (handler)

  The handler function 'get_publisher_info' that executes the tool logic: takes publisher code, fetches detailed info (acts count, years, etc.) from Sejm API endpoint https://api.sejm.gov.pl/eli/acts/{publisher}, handles errors by returning empty dict.

  def get_publisher_info(
      publisher: Annotated[str, "Publisher code (DU for Dziennik Ustaw, MP for Monitor Polski)"]
  ) -> dict:
      """Fetches detailed information about a specific legal act publisher.
  
      Provides comprehensive metadata about a legal publisher including publication
      statistics, available years, and document counts. This information is useful
      for understanding the scope and coverage of different legal gazettes.
  
      Args:
          publisher: Publisher code (e.g., 'DU' for Dziennik Ustaw, 'MP' for Monitor Polski).
  
      Returns:
          dict: Detailed information about the publisher containing code, name, shortName,
                actsCount, and years array. Returns empty dict if publisher not found or error.
  
      Examples:
          User asks: "Tell me about DU publisher":
              Parameters: publisher = 'DU'
              Returns: {'code': 'DU', 'name': 'Dziennik Ustaw', 'shortName': 'Dz.U.', 'actsCount': 96086, 'years': [1918, 1919, ...]}
          User asks: "What is the Dziennik Ustaw?":
              Parameters: publisher = 'DU'
          User asks: "Give me details about Monitor Polski":
              Parameters: publisher = 'MP'
          User asks: "How many acts are in DU?":
              Parameters: publisher = 'DU'
          User asks: "What years are covered by MP?":
              Parameters: publisher = 'MP'
      """
      logger.debug(f"get_publisher_info called with publisher: {publisher}")
      try:
          url = f"https://api.sejm.gov.pl/eli/acts/{publisher}"
          logger.debug(f"Making GET request to: {url}")
          response = requests.get(url, headers={"Accept": "application/json"})
          response.raise_for_status()
  
          data = response.json()
          logger.info(f"get_publisher_info retrieved details for publisher: {publisher}")
          return data
      except Exception as e:
          logger.error(f"Error: {e}")
          return {}

• app.py:374-374 (schema)

  Input schema definition using Annotated type hint for the 'publisher' parameter (string, e.g., 'DU' or 'MP').

  publisher: Annotated[str, "Publisher code (DU for Dziennik Ustaw, MP for Monitor Polski)"]