This is the public-facing repository for MCP Analytics — it contains documentation, examples, and listing assets. The API server code is maintained in a private repository.
Contributions welcome in the following areas:
Open a bug report. Please include:
- Your MCP client (Claude Desktop, Cursor, VS Code, etc.)
- Your data source (Shopify, Stripe, CSV, etc.)
- What you asked the AI to do and what went wrong
- Any error messages from the AI response
The more specific, the faster we can reproduce and fix it.
Open a feature request. Most valuable requests:
- New data connectors (e.g., HubSpot, Salesforce, Amazon Seller)
- New analysis types for specific use cases
- Improvements to report output or interpretation
MCP Analytics has an autonomous module builder that produces new validated analysis modules from a specification — usually within a day. If you have a specific analytical need, describe it in a feature request with:
- The question you're trying to answer
- What data you have (column names and types)
- What output you'd expect
Well-specified module requests get prioritized and built quickly.
Docs live in this repo (docs/, examples/) and at mcpanalytics.ai/docs. For doc improvements, open a PR directly or file an issue describing what's unclear.
The examples/ folder contains real-world usage examples. PRs adding new examples for specific use cases (Shopify revenue analysis, Stripe churn prediction, etc.) are welcome.
This project follows the Contributor Covenant. Be respectful.
Do not report security vulnerabilities in public issues. See SECURITY.md.