Datasentry
Local-first MCP server for data-quality discovery, evidence, drift, and controlled repair.
DataSentry automatically detects data-quality problems, explains the evidence, and turns fixes into a previewable, verifiable, reversible workflow. The deterministic core runs locally, while LLMs are optional and AI may propose repairs but humans approve state-changing actions. DataSentry scans files, databases, and cloud objects with evidence-driven detectors.
Repair is deliberately conservative: propose → preview → apply to copy → verify → rollback, and the original file is never overwritten. The same evidence is exposed through multiple interfaces: CLI/TUI, REST, Web, MCP, JSON, Markdown, HTML, JUnit, and SARIF. Inputs include CSV, Parquet, JSONL, XLSX, DuckDB, SQLite, PostgreSQL, MySQL, and cloud objects.
The MCP stdio server exposes DataSentry's SDK to MCP-capable agents while preserving approval and repair boundaries.
Summary drafted from the project's own website. Every sentence is backed by text on that page and was reviewed before publishing.