Skip to main content
Glama

Related Servers

Alternatives to dataranger

No user-submitted related servers found.

    Related Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      Enables autonomous data quality inspection and repair workflows. It scans DuckDB warehouses for anomalies, generates and verifies fixes in a dry-run copy, then applies them after validation, with full audit logging.
      MIT
    • A
      license
      B
      quality
      D
      maintenance
      Agentic data quality MCP server — runs structured validation rules against warehouses (DuckDB, BigQuery, Athena, Databricks, Postgres), diagnoses failures with LLM root cause analysis, and proposes SQL remediations. Full audit trail of every AI decision.
      6
      15 PyPI
      5
      Apache 2.0
    • A
      license
      B
      quality
      B
      maintenance
      Local-first MCP server for data quality that finds suspicious data, explains findings with evidence, tracks drift, and supports human-approved, reversible repair workflows. Deterministic by default, with AI optional.
      25
      253 PyPI
      2
      Apache 2.0
    • A
      license
      Not graded
      quality
      D
      maintenance
      Enables AI agents to safely work with SQLite databases by enforcing read/write separation, dry-run writes with confirmation, automatic backups, and an audit trail.
      MIT
    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables agents to discover and inspect registered local CSV/Parquet datasets through typed, bounded MCP tool calls while a human CLI controls imports, plan approval, and exports. Repairs are constrained to structured operations with atomic application and reproducible receipts, so data changes require explicit human sign-off.
      MIT
    • A
      license
      A
      quality
      C
      maintenance
      Enables a language model to run four deterministic data checks it cannot argue with — column-contract validation with an explicit quarantine list, arithmetic reconciliation of opening/movements/closing, distinct entity counting against row counts, and duplicate detection returned for human review — each answering with structured findings and row/column locators. Failures come back as results with reasons, so the model reports exactly which rows and columns are broken rather than asserting the data is clean.
      4
      14 PyPI
      MIT

    TDQS

    A4.7/5.0

    Scored across 6 tools

    Disambiguation5/5

    The four patrol_* tools are unambiguously partitioned by data source (dbt model, DuckDB, files, Snowflake), and each description explicitly names the siblings and says when to prefer them. get_incident is a clearly separate read-only evidence lookup, and run_demo is clearly a synthetic-data harness. No two tools could plausibly be selected for the same task.

    Naming Consistency5/5

    All six names are snake_case verb_noun: get_incident, patrol_dbt, patrol_duckdb, patrol_files, patrol_snowflake, run_demo. The shared patrol_ prefix correctly groups the four source variants, and get_/run_ follow the same predictable pattern.

    Tool Count5/5

    Six tools is tight and every one earns its place: four source-specific patrols, one evidence retrieval, one demo. The variant split is driven by real environmental differences (credentials, manifest, file formats) rather than padding, and nothing feels redundant.

    Completeness4/5

    The core lifecycle is covered: run a patrol on any supported source, get the summary, drill into a single incident, and validate the workflow on demo data. The main gap is that fixes are only emitted as fixed_version.csv for review with no tool to inspect or apply them, and there is no way to list prior patrol runs or their outputs folders.

    Maintenance

    ActivityMaintained
    ResponsivenessNo recent community issues