Graphistry MCP

Instructions

Set node color encoding for a graph using Graphistry's encode_point_color API.

Args:
    graph_id (str): The ID of the graph to modify (from visualize_graph).
    column (str): The node column to use for color encoding (e.g., 'type', 'score').
    categorical_mapping (dict, optional): Map of category values to color codes. Example: {'mac': '#F99', 'macbook': '#99F'}. If not provided, Graphistry will auto-assign colors.
    default_mapping (str, optional): Color code to use for values not in categorical_mapping. Example: 'silver'.
    as_continuous (bool, optional): If True, treat the column as continuous and use a gradient palette. Example: True for numeric columns like 'score'.

Returns:
    dict: { 'graph_id': ..., 'url': ... } with the updated visualization URL.

Example:
    encode_point_color(graph_id, column='type', categorical_mapping={'mac': '#F99', 'macbook': '#99F'}, default_mapping='silver')

Implementation Reference

• src/graphistry_mcp_server/server.py:333-369 (handler)

  The main handler function for the 'encode_point_color' tool, decorated with @mcp.tool() for registration in the FastMCP server. It modifies the graph's node colors based on a specified column and mappings using Graphistry's encode_point_color method.

  @mcp.tool()
  async def encode_point_color(
      graph_id: str,
      column: str,
      categorical_mapping: Optional[dict] = None,
      default_mapping: Optional[str] = None,
      as_continuous: Optional[bool] = False
  ) -> Dict[str, Any]:
      """
      Set node color encoding for a graph using Graphistry's encode_point_color API.
  
      Args:
          graph_id (str): The ID of the graph to modify (from visualize_graph).
          column (str): The node column to use for color encoding (e.g., 'type', 'score').
          categorical_mapping (dict, optional): Map of category values to color codes. Example: {'mac': '#F99', 'macbook': '#99F'}. If not provided, Graphistry will auto-assign colors.
          default_mapping (str, optional): Color code to use for values not in categorical_mapping. Example: 'silver'.
          as_continuous (bool, optional): If True, treat the column as continuous and use a gradient palette. Example: True for numeric columns like 'score'.
  
      Returns:
          dict: { 'graph_id': ..., 'url': ... } with the updated visualization URL.
  
      Example:
          encode_point_color(graph_id, column='type', categorical_mapping={'mac': '#F99', 'macbook': '#99F'}, default_mapping='silver')
      """
      if graph_id not in graph_cache:
          raise ValueError(f"Graph not found: {graph_id}")
      g = graph_cache[graph_id]["graph"]
      kwargs = {"column": column}
      if categorical_mapping:
          kwargs["categorical_mapping"] = categorical_mapping
      if default_mapping:
          kwargs["default_mapping"] = default_mapping
      if as_continuous:
          kwargs["as_continuous"] = as_continuous
      g = g.encode_point_color(**kwargs)
      graph_cache[graph_id]["graph"] = g
      return {"graph_id": graph_id, "url": g.plot(render=False)}