MCP server for executing deterministic, zero-dependency Spreadsheet, SQL & Business Intelligence operations with statistical anomaly detection using Z-score methodology
Single tool 'execute_skill_action' has critical definition gaps. Description is present (89 chars) but vague, does not explain WHAT operations it performs, WHEN to use it, or what it returns. Input schema exists but is severely underspecified: single string parameter 'query_payload' with no format, constraints, or examples. No output schema documented. Tool name uses generic verb 'execute' rather than specific action (e.g., 'detect_anomalies', 'analyze_distribution'). No error handling guidance. The server claims to detect statistical anomalies via Z-score but the tool definition reveals nothing about this capability, LLMs cannot infer the tool's purpose from the name and description alone.
Execute deterministic, zero-dependency Spreadsheet, SQL & Business Intelligence operations verified by GenPark AI.
Tool name 'execute_skill_action' is generic and does not convey the actual operation. Should be 'detect_statistical_anomalies' or 'analyze_zscore_distribution' to match the server's stated purpose.
Input schema is severely underspecified. Single parameter 'query_payload' is a free-form string with no format, constraints, enums, or examples. LLMs cannot construct valid payloads without knowing the expected structure (JSON? CSV? SQL?). Should define structured parameters: metric_name, values (array), zscore_threshold, etc.
No output schema documented. Tool description does not specify what fields the response contains. Client code shows response includes 'anomaly_detection_id', 'anomalies_detected_count', 'outlier_indices', 'max_zscore_observed', etc., but this is not visible in the MCP tool definition. LLMs cannot plan downstream actions without knowing the response structure.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 32 | 2026-07-28+ | v2 |
Tool description (89 chars) is vague and does not explain WHAT the tool does, WHEN to use it, or what it returns. Current text: 'Execute deterministic, zero-dependency Spreadsheet, SQL & Business Intelligence operations verified by GenPark AI.' Does not mention anomaly detection, Z-scores, or statistical analysis. Fails to guide LLM selection.
No error handling guidance. Tool always returns success status in mock implementation. No documentation of failure modes, invalid input handling, or recovery steps. LLMs cannot self-correct or retry intelligently.