Skip to main content
Glama

Related Servers

Alternatives to AI-ML-Lab-MCP

  • A
    license
    A
    quality
    B
    maintenance
    Enables Claude and other MCP-compatible agents to process documents, extract structured data, detect PII, and export LLM-ready datasets through natural language tool calls.
    8
    71 PyPI
    1
    MIT

Related Servers

  • A
    license
    A
    quality
    B
    maintenance
    Enables Claude to inspect, transform, project, cluster, tensorize, slice, and plot high-dimensional datasets server-side while returning only bounded summaries, embeddings, metrics, and images to the conversation.
    20
    MIT
  • A
    license
    Not graded
    quality
    B
    maintenance
    Enables MCP clients to load and inspect datasets, run read-only SQL with DuckDB semantics, ask natural-language questions that are translated into SQL and summarized, and generate chart specs. It works with any MCP client and optionally uses an OpenAI-compatible model for analysis and visualization tools.
    740 npm
    MIT
  • A
    license
    Not graded
    quality
    C
    maintenance
    Search, preview, and analyze datasets from 20+ platforms and millions of datasets via a single MCP connector. Works with Claude instantly
    7 npm
    29
    MIT
  • F
    license
    Not graded
    quality
    D
    maintenance
    Lets Claude and other MCP-compatible clients securely connect to Syntitan, an AI-Ready Data Platform, through natural language for data management and analytics.
    -
  • A
    license
    Not graded
    quality
    C
    maintenance
    Enables Claude to query client and document data in plain English with cited answers, enforce per-client access, safely file incoming documents, and record audited actions through MCP tools.
    MIT
  • F
    license
    Not graded
    quality
    D
    maintenance
    Enables LLM clients to load, clean, visualize, and analyze tabular datasets, and to generate multi-format analytic reports through natural language instructions.
    -

TDQS

B3.3/5.0

Scored across 97 tools

Disambiguation3/5

There are several overlapping families: get_dataset_metadata/dataset_info/dataset_shape/list_columns/column_types all report dataset structure; describe_dataset/analyze_distribution/analyze_categorical overlap with value_counts/unique_values; normalize_features/scale_features/standardize_features do largely the same job; fill_missing_mean/median/mode overlap with impute_missing and handle_missing_values. Descriptions do mitigate this by cross-referencing each other (e.g. select_features explicitly distinguishes itself from feature_selection), but the near-identical pair feature_selection vs select_features and the three scaling tools will still cause misselection.

Naming Consistency4/5

Everything is snake_case with no camelCase mixing, and most names are verb_noun (load_csv, detect_outliers, train_test_split, compare_models). Deviations are minor: visualization and inspection tools are bare nouns (histogram, heatmap, dataset_info, regression_metrics) and verb prefixes vary between analyze_/find_/suggest_. The feature_selection / select_features pair is the one genuinely confusing naming collision.

Tool Count1/5

97 tools is far past the 50+ extreme-mismatch threshold, and many are trivially collapsible: eight separate plot tools could be one plot tool with a kind parameter, twelve model trainers could be one with an algorithm parameter, and three scaling tools could be one with a method argument. The domain is broad, but the surface is bloated well beyond what an agent can reliably navigate.

Completeness4/5

The pipeline is unusually complete: load/export, EDA, cleaning, missing-value handling, outlier treatment, splitting, encoding/scaling/imputation, feature engineering and selection, a dozen regressors and classifiers, metric and diagnostic plots, plus model comparison, benchmarking and ranking. Remaining gaps are hyperparameter search/tuning (the docs say 'tune the winner afterwards' but no tuning tool exists), no tool to score new/unseen data with a trained model, and no model persistence or unsupervised methods.

Maintenance

ActivityMaintained
ResponsivenessNo recent community issues