convert_cell_type

Change a Jupyter notebook cell's type between code, markdown, or raw formats to modify content behavior and presentation.

Instructions

Convert a cell from one type to another

Input Schema

TableJSON Schema

Name	Required	Description
`notebook_path`	Yes	Absolute path to the Jupyter notebook file
`cell_index`	Yes	Zero-based index of the cell
`new_type`	Yes	New cell type

Implementation Reference

src/jupyter-handler.js:269-311 (handler)

Main handler function that performs the cell type conversion by modifying the notebook JSON, handling type-specific properties like execution_count and outputs, and persisting the changes.

async convertCellType(notebookPath, cellIndex, newType) {
  const notebook = await this.readNotebook(notebookPath);
  this.validateCellIndex(notebook.cells, cellIndex);
  this.validateCellType(newType);
  
  const cell = notebook.cells[cellIndex];
  const oldType = cell.cell_type;
  
  if (oldType === newType) {
    return {
      content: [
        {
          type: "text",
          text: `Cell ${cellIndex} is already of type '${newType}'`
        }
      ]
    };
  }
  
  // Convert cell type
  cell.cell_type = newType;
  
  // Handle type-specific properties
  if (newType === 'code') {
    cell.execution_count = null;
    cell.outputs = [];
  } else {
    // Remove code-specific properties for non-code cells
    delete cell.execution_count;
    delete cell.outputs;
  }
  
  await this.writeNotebook(notebookPath, notebook);
  
  return {
    content: [
      {
        type: "text",
        text: `Successfully converted cell ${cellIndex} from '${oldType}' to '${newType}'`
      }
    ]
  };
}

src/index.js:166-184 (schema)

Input schema defining the parameters for the convert_cell_type tool.

inputSchema: {
  type: "object",
  properties: {
    notebook_path: {
      type: "string",
      description: "Absolute path to the Jupyter notebook file"
    },
    cell_index: {
      type: "integer",
      description: "Zero-based index of the cell"
    },
    new_type: {
      type: "string",
      enum: ["code", "markdown", "raw"],
      description: "New cell type"
    }
  },
  required: ["notebook_path", "cell_index", "new_type"]
}

src/index.js:163-185 (registration)

Tool registration in the ListTools response, providing name, description, and schema.

{
  name: "convert_cell_type",
  description: "Convert a cell from one type to another",
  inputSchema: {
    type: "object",
    properties: {
      notebook_path: {
        type: "string",
        description: "Absolute path to the Jupyter notebook file"
      },
      cell_index: {
        type: "integer",
        description: "Zero-based index of the cell"
      },
      new_type: {
        type: "string",
        enum: ["code", "markdown", "raw"],
        description: "New cell type"
      }
    },
    required: ["notebook_path", "cell_index", "new_type"]
  }
},

src/index.js:362-367 (registration)

Dispatch/routing in CallToolRequest handler to invoke the convertCellType method.

case "convert_cell_type":
  return await this.jupyterHandler.convertCellType(
    args.notebook_path,
    args.cell_index,
    args.new_type
  );

src/jupyter-handler.js:89-93 (helper)

Helper method to validate the new cell type against supported types.

validateCellType(cellType) {
  if (!this.supportedCellTypes.includes(cellType)) {
    throw new Error(`Invalid cell type '${cellType}'. Supported types: ${this.supportedCellTypes.join(', ')}`);
  }
}

MCP Jupyter Server