Skip to main content
Glama
powerdrillai

Powerdrill MCP Server

Official
by powerdrillai

mcp_powerdrill_list_datasets

Retrieve available datasets from Powerdrill for AI data analysis, with options to filter by name, limit results, and paginate through large collections.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
limitNoMaximum number of datasets to return
pageNumberNoThe page number to start listing (default: 1)
pageSizeNoThe number of items on a single page (default: 10)
searchNoSearch for datasets by name

Implementation Reference

  • The primary handler function for the MCP tool 'mcp_powerdrill_list_datasets'. It extracts parameters, initializes the PowerdrillClient, fetches datasets from the API, applies optional limit, formats the result as JSON text content, and handles errors appropriately.
    async (args, extra) => {
      try {
        const { limit, pageNumber, pageSize, search } = args;
    
        // Initialize Powerdrill client
        const client = new (await import('./utils/powerdrillClient.js')).PowerdrillClient();
    
        // Fetch datasets
        const response = await client.listDatasets({
          pageNumber,
          pageSize,
          search
        });
    
        // Check if response is valid
        if (!response) {
          throw new Error(`Empty response received from API`);
        }
    
        if (response.code !== 0) {
          throw new Error(`API returned error code: ${response.code}, message: ${response.message || 'No message'}`);
        }
    
        if (!response.data || !response.data.records) {
          throw new Error(`Invalid API response format: missing data.records property`);
        }
    
        // Apply limit if provided
        let datasets = response.data.records || [];
        if (limit && limit > 0) {
          datasets = datasets.slice(0, limit);
        }
    
        if (datasets.length === 0) {
          return {
            content: [
              {
                type: "text",
                text: JSON.stringify({
                  message: "No datasets found",
                  datasets: []
                })
              }
            ]
          };
        }
    
        // Format the response as MCP content
        const result = {
          count: datasets.length,
          total: response.data.total_items || datasets.length,
          datasets: datasets.map((dataset: Dataset) => ({
            id: dataset.id,
            name: dataset.name,
            description: dataset.description || ''
          }))
        };
    
        // Return the formatted response
        return {
          content: [
            {
              type: "text",
              text: JSON.stringify(result, null, 2)
            }
          ]
        };
      } catch (error: any) {
        console.error(`Error listing datasets: ${error.message}`);
        console.error(error.stack);
    
        // Return error response
        return {
          content: [
            {
              type: "text",
              text: JSON.stringify({
                error: `Error listing datasets: ${error.message}`,
                errorType: error.name,
                errorStack: process.env.NODE_ENV === 'development' ? error.stack : undefined
              }, null, 2)
            }
          ],
          isError: true
        };
      }
    }
  • Zod schema defining the input parameters for the tool: limit, pageNumber, pageSize, and search.
      limit: z.number().optional().describe('Maximum number of datasets to return'),
      pageNumber: z.number().optional().describe('The page number to start listing (default: 1)'),
      pageSize: z.number().optional().describe('The number of items on a single page (default: 10)'),
      search: z.string().optional().describe('Search for datasets by name')
    },
  • src/index.ts:36-130 (registration)
    Registration of the 'mcp_powerdrill_list_datasets' tool with the MCP server using server.tool(name, schema, handler).
      'mcp_powerdrill_list_datasets',
      {
        limit: z.number().optional().describe('Maximum number of datasets to return'),
        pageNumber: z.number().optional().describe('The page number to start listing (default: 1)'),
        pageSize: z.number().optional().describe('The number of items on a single page (default: 10)'),
        search: z.string().optional().describe('Search for datasets by name')
      },
      async (args, extra) => {
        try {
          const { limit, pageNumber, pageSize, search } = args;
    
          // Initialize Powerdrill client
          const client = new (await import('./utils/powerdrillClient.js')).PowerdrillClient();
    
          // Fetch datasets
          const response = await client.listDatasets({
            pageNumber,
            pageSize,
            search
          });
    
          // Check if response is valid
          if (!response) {
            throw new Error(`Empty response received from API`);
          }
    
          if (response.code !== 0) {
            throw new Error(`API returned error code: ${response.code}, message: ${response.message || 'No message'}`);
          }
    
          if (!response.data || !response.data.records) {
            throw new Error(`Invalid API response format: missing data.records property`);
          }
    
          // Apply limit if provided
          let datasets = response.data.records || [];
          if (limit && limit > 0) {
            datasets = datasets.slice(0, limit);
          }
    
          if (datasets.length === 0) {
            return {
              content: [
                {
                  type: "text",
                  text: JSON.stringify({
                    message: "No datasets found",
                    datasets: []
                  })
                }
              ]
            };
          }
    
          // Format the response as MCP content
          const result = {
            count: datasets.length,
            total: response.data.total_items || datasets.length,
            datasets: datasets.map((dataset: Dataset) => ({
              id: dataset.id,
              name: dataset.name,
              description: dataset.description || ''
            }))
          };
    
          // Return the formatted response
          return {
            content: [
              {
                type: "text",
                text: JSON.stringify(result, null, 2)
              }
            ]
          };
        } catch (error: any) {
          console.error(`Error listing datasets: ${error.message}`);
          console.error(error.stack);
    
          // Return error response
          return {
            content: [
              {
                type: "text",
                text: JSON.stringify({
                  error: `Error listing datasets: ${error.message}`,
                  errorType: error.name,
                  errorStack: process.env.NODE_ENV === 'development' ? error.stack : undefined
                }, null, 2)
              }
            ],
            isError: true
          };
        }
      }
    );
  • Helper method in PowerdrillClient class that makes the HTTP GET request to the Powerdrill API endpoint `/datasets` with pagination and search query parameters, handles timeouts and errors.
    async listDatasets(options: { 
      pageNumber?: number;
      pageSize?: number;
      search?: string;
      timeout?: number; 
    } = {}) {
      const timeout = options.timeout || 30000; // Default 30 second timeout
    
      try {
        // Build query parameters
        let queryParams = `user_id=${this.config.userId}`;
        
        if (options.pageNumber) {
          queryParams += `&page_number=${options.pageNumber}`;
        }
        
        if (options.pageSize) {
          queryParams += `&page_size=${options.pageSize}`;
        }
        
        if (options.search) {
          queryParams += `&search=${encodeURIComponent(options.search)}`;
        }
    
        const response = await this.client.get(`/datasets?${queryParams}`, {
          timeout: timeout,
          validateStatus: (status) => status >= 200 && status < 500 // Don't throw on 4xx errors
        });
    
        // Handle HTTP errors manually
        if (response.status >= 400) {
          throw new Error(`HTTP error ${response.status}: ${response.statusText || 'Unknown error'}`);
        }
    
        // Validate response structure
        if (!response.data) {
          throw new Error('Invalid API response: missing data');
        }
    
        return response.data;
      } catch (error: any) {
        if (error.code === 'ECONNABORTED') {
          console.error(`PowerdrillClient: Request timed out after ${timeout}ms`);
          throw new Error(`Request timed out after ${timeout}ms`);
        }
    
        if (error.response) {
          // The request was made and the server responded with a status code
          console.error(`PowerdrillClient: HTTP error ${error.response.status}`, error.response.data);
          throw new Error(`API error: ${error.response.status} ${error.response.statusText}`);
        } else if (error.request) {
          // The request was made but no response was received
          console.error('PowerdrillClient: No response received', error.request);
          throw new Error('No response received from API');
        }
    
        console.error('PowerdrillClient: Error listing datasets:', error.message);
        throw error;
      }
    }

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/powerdrillai/powerdrill-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server