rodcar/pnda-mcp
If you are the rightful owner of pnda-mcp and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
PNDA-MCP is a Model Context Protocol (MCP) server designed for Peru's National Open Data Platform, facilitating AI agents in discovering and analyzing public datasets.
PNDA-MCP serves as a Model Context Protocol (MCP) server for Peru's National Open Data Platform, known as Plataforma Nacional de Datos Abiertos (PNDA). The platform, datosabiertos.gob.pe, hosts a wealth of datasets that are invaluable for data analysis. However, navigating and retrieving the most pertinent data can be challenging for AI agents. PNDA-MCP addresses this by offering tools and prompts that enable AI agents or any MCP client, such as VS Code or Claude Desktop, to efficiently search for and access dataset metadata and associated data files. The primary objective is to empower data scientist agents or code agents to automatically discover and analyze public datasets. The repository also includes an ETL pipeline that extracts, transforms, and indexes dataset titles, ensuring that the data is readily accessible and up-to-date. By leveraging semantic search and text vector embeddings, PNDA-MCP simplifies the process of finding relevant datasets, making it an essential tool for data-driven decision-making.
Features
- Facilitates AI agents in discovering and analyzing public datasets.
- Provides tools and prompts for efficient dataset search and access.
- Includes an ETL pipeline for data extraction, transformation, and indexing.
- Supports semantic search with text vector embeddings.
- Integrates with MCP clients like VS Code and Claude Desktop.
Tools
dataset_search
Search for relevant datasets from the PNDA Peru. 'query' is the search text, 'top_k' limits the number of results returned (max 25).
dataset_details
Get dataset details including title, metadata, and resources. Returns complete resource information: direct download URLs, file names, sizes, creation dates, MIME types, formats, states, and descriptions.