Browse
→ Data & Storage
→ DataSentry MCP Server
DataSentry MCP Server
Local-first AI copilot for data quality with MCP, REST, CLI and web interfaces. Scans CSV, Parquet, JSONL, XLSX, DuckDB, SQLite, PostgreSQL, MySQL and cloud-object data; applies 39 evidence-driven detectors; scores data across six quality dimensions; supports drift detection and AI-suggested repairs gated by human approval. Public GitHub repository created August 9, 2026.
MCP unverified
Integration
| Transport | stdio |
| Auth | none |
| Endpoint | https://github.com/Jackxiaozhiren/datasentry |
| Install | pip install datasentry-ai |
Use Cases
| 01 | Run local data-quality scans from an AI assistant and receive evidence-backed issue reports with samples, ratios, confidence and six-dimension quality scores |
| 02 | Use MCP to let agents inspect detected data issues, propose repair rules and preview changes while requiring human approval before any repair is applied |
| 03 | Monitor schema, row-count, score and issue-distribution drift across repeated scans in data engineering or analytics workflows |
Tags
data-quality data-observability etl duckdb local-first data-cleaning ai-copilot python
Machine-readable: /api/servers.json
· JSON-LD schema embedded in <head>