io.github.CSOAI-ORG/data-science-ai-mcp

Rank features by estimated importance for a prediction task. Args: features: Feature metadata as [{"name": "age", "type": "numeric", "cardinality": 50, "missing_pct": 2, "correlation_with_target": 0.65, "variance": 150}] target_type: ML task type (classification, regression) method: Importance method (statistical, permutation, tree_based) Behavior: This tool is read-only and stateless — it produces analysis output without modifying any external systems, databases, or files. Safe to call repeatedly with identical inputs (idempotent). Free tier: 10/day rate limit. Pro tier: unlimited. No authentication required for basic usage. When to use: Use this tool when you need structured analysis or classification of inputs against established frameworks or standards. When NOT to use: Not suitable for real-time production decision-making without human review of results.

Compare ML models across metrics. Returns composite scores, speed comparisons, and recommendations for production vs accuracy. Args: models: Model results as [{"name": "XGBoost", "accuracy": 0.92, "precision": 0.90, "recall": 0.88, "f1": 0.89, "training_time_sec": 120, "inference_ms": 5, "parameters": 50000}] task_type: ML task (classification, regression, ranking) Behavior: This tool generates structured output without modifying external systems. Output is deterministic for identical inputs. No side effects. Free tier: 10/day rate limit. Pro tier: unlimited. No authentication required for basic usage. When to use: Use this tool when you need structured analysis or classification of inputs against established frameworks or standards. When NOT to use: Not suitable for real-time production decision-making without human review of results.

Profile a dataset: completeness, quality issues, type distribution, and per-column statistics. Args: columns: Column metadata as [{"name": "age", "type": "numeric", "non_null_count": 950, "unique_count": 80, "min": 18, "max": 90, "mean": 35.2, "std": 12.1}] row_count: Total number of rows sample_values: Optional sample values per column as {"col_name": [val1, val2, ...]} Behavior: This tool is read-only and stateless — it produces analysis output without modifying any external systems, databases, or files. Safe to call repeatedly with identical inputs (idempotent). Free tier: 10/day rate limit. Pro tier: unlimited. No authentication required for basic usage. When to use: Use this tool when you need structured analysis or classification of inputs against established frameworks or standards. When NOT to use: Not suitable for real-time production decision-making without human review of results.

Compute pairwise Pearson correlations between variables. Flags strong correlations and multicollinearity warnings. Args: variables: Variables with values as [{"name": "height", "values": [170, 175, 160, ...]}, {"name": "weight", "values": [70, 80, 55, ...]}] Behavior: This tool is read-only and stateless — it produces analysis output without modifying any external systems, databases, or files. Safe to call repeatedly with identical inputs (idempotent). Free tier: 10/day rate limit. Pro tier: unlimited. No authentication required for basic usage. When to use: Use this tool when you need structured analysis or classification of inputs against established frameworks or standards. When NOT to use: Not suitable for real-time production decision-making without human review of results.

Recommend visualizations based on data characteristics and analysis goal. Returns chart types, code hints, and library suggestions. Args: data_description: Dataset info as {"columns": [{"name": "x", "type": "numeric", "cardinality": 50}], "row_count": 1000, "goal": "explore"}. Goals: explore, comparison, composition, distribution, relationship Behavior: This tool is read-only and stateless — it produces analysis output without modifying any external systems, databases, or files. Safe to call repeatedly with identical inputs (idempotent). Free tier: 10/day rate limit. Pro tier: unlimited. No authentication required for basic usage. When to use: Use this tool when you need structured analysis or classification of inputs against established frameworks or standards. When NOT to use: Not suitable for real-time production decision-making without human review of results.