MCP Analyst
MCP Analyst is an MCP server that empowers claude to analyze local CSV or Parquet files.
What is MCP Analyst?
MCP Analyst is a server designed to empower Claude to analyze local CSV or Parquet files efficiently.
How to use MCP Analyst?
To use MCP Analyst, install the required software (uv) and configure the Claude settings to include the MCP server. You can then run the server with your dataset without uploading the full file.
Key features of MCP Analyst?
- Analyzes large datasets locally without uploading them.
- Supports CSV and Parquet file formats.
- Optimizes cost by processing data locally.
Use cases of MCP Analyst?
- Analyzing large datasets that exceed the context window.
- Performing data analysis without the need for cloud storage.
- Integrating with Claude for enhanced data processing capabilities.
FAQ from MCP Analyst?
- What file formats does MCP Analyst support?
MCP Analyst supports both CSV and Parquet file formats for data analysis.
- Do I need to upload my files to use MCP Analyst?
No, MCP Analyst allows you to analyze files locally, which helps in optimizing costs and maintaining data privacy.
- How do I install the required software?
You can install uv using Homebrew on Mac or Winget on Windows as per the installation instructions provided.
MCP Analyst
MCP Analyst is an MCP server that empowers claude to analyze local CSV or Parquet files.
Use this server when your dataset size is bigger than the context window or you don't want to upload full file to optimize the cost.
Installation
Install uv
uv is required to run the MCP server.
Mac
brew install uv
Windows
winget install --id=astral-sh.uv -e
Add servers in MCP
To use the server in Claude you would need to update the Claude config
On MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json On Windows: %APPDATA%/Claude/claude_desktop_config.json
{
"mcpServers": {
"analyst": {
"command": "uvx",
"args": [
"mcp-analyst",
"--file_location",
"<replace_this_with_path_to_csv_or_parquet_files_on_your_machine>"
]
}
}
}
How do I provide more than one CSV file?
The server supports the using globs in the file location. So for example if your directory data has more than one csv file then you can give file-location as <parent_path_to_data>/data/*.csv