MCP server for ODBC-based database connectivity with specialized support for NetSuite via ODBC
14 tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. Most tools have 1-2 parameters with minimal descriptions (10-30 chars). No tool annotations (readOnlyHint/destructiveHint). Output schemas are not documented, LLMs cannot predict return structure. Error handling and recovery guidance absent. Naming is consistent (netsuite_* prefix) but generic descriptions like 'Configuration tool for the server' lack actionable context. Tools 10-12 (write operations) lack confirmation/dry-run patterns. Average description length ~40 chars, well below the 194-char baseline for A+ tools.
Insert a new row into a table
Configuration tool for the server
Count rows in a table
Execute INSERT, UPDATE, DELETE, or other non-query SQL statements
Execute SELECT SQL statements and return results
Execute a stored procedure or function
Get columns for a specific table
Get foreign key information for a table
No output schemas documented. LLMs cannot predict return structure, field names, or types. Breaks tool chaining and forces agents to guess at response format.
Write operations (execute_nonquery, execute_routine, add_row) lack confirmation/dry-run patterns. Agents can execute destructive SQL without safeguards.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 43 | 2026-07-28+ | v2 |
Get indexes for a specific table
Get primary key information for a table
Get stored procedures and functions
Get list of tables from the database
Instructions for using the NetSuite ODBC server
Get the database server version
Parameter descriptions are minimal (10-30 chars). 'Name of the schema' and 'Name of the table' lack context on format, constraints, or when to use. Below 194-char baseline for A+ tools.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint). LLMs cannot distinguish safe reads from destructive writes without explicit hints.
No error handling or recovery guidance. Tools return raw errors without actionable next steps. LLMs cannot self-correct on invalid SQL, missing tables, or permission failures.