{
  "$schema": "https://static.modelcontextprotocol.io/schemas/mcp-server-card/v1.json",
  "version": "1.0",
  "protocolVersion": "2025-03-26",
  "serverInfo": {
    "name": "datafold",
    "title": "Datafold MCP Server",
    "version": "1.0.0"
  },
  "description": "Connect AI agents to your data stack. Datafold automates data engineering with cross-database data comparison, column-level lineage, data quality monitoring, and SQL querying — all accessible via MCP.",
  "iconUrl": "https://www.datafold.com/images/favicon.png",
  "documentationUrl": "https://docs.datafold.com/datafold-mcp",
  "transport": {
    "type": "streamable-http",
    "endpoint": "https://app.datafold.com/mcp/"
  },
  "authentication": {
    "required": true,
    "schemes": ["bearer"]
  },
  "capabilities": {},
  "instructions": "Use Datafold to compare data across databases, trace column-level lineage for impact analysis, manage data quality monitors, run SQL queries against connected warehouses, and validate data migrations. Authenticate with a Datafold API key as a Bearer token. Full tool schemas are available after connection via tools/list.",
  "tools": "dynamic",
  "_meta": {
    "use_cases": [
      {
        "title": "Cross-Database Data Comparison",
        "description": "Compare tables row-by-row and column-by-column across same or different databases. Identify added, removed, and modified rows, schema differences, and column-level match statistics. Supports composite primary keys, sampling for large tables, and tolerance thresholds."
      },
      {
        "title": "Column-Level Data Lineage",
        "description": "Trace data flow at the column level across your entire data stack — Snowflake, Databricks, BigQuery, PostgreSQL, PowerBI, Tableau, and more. Understand upstream sources and downstream consumers for impact analysis, root cause debugging, and safe refactoring."
      },
      {
        "title": "Data Quality Monitoring",
        "description": "Create and manage data quality monitors as code (YAML). Track metric anomalies, schema changes, data freshness, and custom SQL rules. Trigger checks on demand and review run history with alerts via Slack, email, or PagerDuty."
      },
      {
        "title": "SQL Querying & Data Exploration",
        "description": "Run SQL queries against any connected data warehouse. Discover available data sources, search for tables by name, and inspect column schemas — all without leaving your agent environment."
      },
      {
        "title": "Data Migration Validation",
        "description": "Validate data parity between source and target systems during platform migrations. Compare datasets progressively — primary keys first, then critical columns, then full comparison — with sampling for large-scale validation."
      },
      {
        "title": "CI/CD Data Testing",
        "description": "Catch data quality regressions before deployment by comparing production data against staging. Integrate with dbt, Airflow, and git workflows to surface silent data issues that schema tests miss."
      }
    ]
  }
}