A Google Sheets MCP server for data manipulation and analysis
Server has 8 tools with basic descriptions and parameter schemas, but significant gaps prevent production use. All tools have descriptions (10-50 chars, below the 50-200 char baseline for LLM optimization), and input schemas are present with type definitions. However, output schemas are completely undocumented, LLMs cannot plan downstream calls or extract results. Parameter descriptions are minimal (e.g., 'Dictionary with column names and values' lacks format/constraint guidance). No error handling guidance, no recovery paths, and no distinction between retryable vs. fatal errors. Tool naming is verb-first (good), but composition issues exist: add_row, edit_row, delete_row, add_sheet_column, rename_sheet_column, and transform_sheet_column all operate on sheets but lack clear differentiation in descriptions. No tool annotations (readOnlyHint, destructiveHint, idempotentHint) despite clear risk levels (READ_ONLY, WRITE, DESTRUCTIVE). Security: no evidence of secret injection for Google auth credentials, likely exposed as parameters or environment variables without isolation.
Add a new row to the Google Sheet
Add a new calculated column to the Google Sheet
Delete a row from the Google Sheet
Edit an existing row in the Google Sheet
Fetch data from a Google Sheet and return as JSON
Rename a column in the Google Sheet
Transform values in a column using various operations
No output schemas documented. LLMs cannot infer what fields are returned, breaking downstream tool chaining and forcing discovery detours.
Descriptions are 30-50 characters, well below the 50-200 char baseline for LLM optimization. Lack WHAT/WHEN/WHY context. E.g., 'Update a Google Sheet with modified data' does not explain when to use this vs. add_row or edit_row.
No error handling guidance. Tools raise exceptions without recovery hints. E.g., 'Error updating sheet: {e}' tells LLM nothing about retry eligibility, user-fixable issues, or next steps.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 40 | <=2025-11-25 | v2 |
Update a Google Sheet with modified data
No tool annotations despite clear risk levels. delete_row is DESTRUCTIVE but lacks destructiveHint. fetch_google_sheet is READ_ONLY but lacks readOnlyHint. Agents cannot reason about side effects.
Parameter descriptions lack format/constraint guidance. 'row_data' is described as 'Dictionary with column names and values', no mention of required fields, data types, or validation rules. 'transformation' accepts 'uppercase', 'lowercase', etc., but no enum constraint or validation.