An MCP server over a real, sourced dataset. Clientvo: the monthly searches, read from search (a month), served as an agent-callable dataset with its source and licence on every answer.

Clientvo publishes monthly searches at https://clientvo.com/tools. This package serves that dataset to any Model Context Protocol client — Claude, or any agent — so an assistant can answer questions from it and cite the source.

Install

npx @ellul-estate/clientvo-com-clientvo-tools-mcp

Or add it to an MCP client config as a stdio server running npx @ellul-estate/clientvo-com-clientvo-tools-mcp.

Tools

tool	what it answers
`dataset_columns`	the schema: columns, which are numeric, row count, provenance
`dataset_search`	rows matching a text query
`dataset_row`	the row(s) where a column equals a value
`dataset_compare`	rows side by side, for "X vs Y"
`dataset_stats`	count, min, max, mean, median, sum of a numeric column
`dataset_top`	the highest or lowest rows by a numeric column
`dataset_provenance`	the source, computed date, licence and citation

The full dataset is also offered as a CSV resource.

Data & citation

Source: the customers you enter, cited under the tool it appears in. Every answer carries the dataset page and licence. The data is free to cite with a link to https://clientvo.com/tools.

Licence: https://clientvo.com/terms.

Published by Clientvo · https://clientvo.com