Loading ToolBench

Reading the current corpus…

mysql_mcp_server MCP server · F · ToolBench

mysql_mcp_server

A Model Context Protocol (MCP) server that enables secure interaction with MySQL databases. This server allows AI assistants to list tables, read data, and execute SQL queries through a controlled interface, making database exploration and analysis safer and more structured.

<=2025-11-25STDIO1 tools

43/100

Definition Quality48

Protocol Readiness47

Supportability33

Spec posture

Inferred effective spec <=2025-11-25·current baseline 2026-07-28

Static source inference · medium confidence · detected: Logging

Deprecated protocol patterns detected

Summary

Single tool 'execute_sql' has a basic schema and minimal description. The tool name is clear (verb_noun pattern), but the description is only 44 characters, below the 50-200 char baseline for LLM-optimized descriptions. The input schema is present with proper JSON Schema structure, but lacks parameter descriptions beyond a bare string type. No output schema is documented. The tool performs a destructive operation (WRITE risk) but has no confirmation pattern, dry-run capability, or error recovery guidance. No tool annotations (readOnlyHint, destructiveHint, idempotentHint) are present despite the WRITE risk classification.

Tools (1)

execute_sqlwriteauthsource verified48/100

Execute an SQL query on the MySQL server

Issues (5)

HIGHtool-description ↗

Tool description is only 44 characters ('Execute an SQL query on the MySQL server'). LLM-optimized descriptions should be 50 - 200 chars and include WHEN to use it, prerequisites, and what it returns. Current description lacks context for tool selection.

execute_sql

HIGHtool ↗

No output schema documented. Callers cannot predict the structure of results (CSV format, JSON, error format). LLMs need to know what fields to expect for downstream reasoning.

execute_sql

CRITICALconfirmation-request ↗

Destructive tool (WRITE risk) lacks confirmation pattern or dry-run capability. Agents can execute DELETE, DROP, or UPDATE without safeguards. No error recovery guidance if a query fails mid-transaction.

execute_sql

HIGHtool ↗

No tool annotations (destructiveHint, readOnlyHint, idempotentHint). The tool is marked WRITE risk but the schema does not declare this formally, forcing LLMs to infer intent from the name alone.

execute_sql

Recommendations

Expand tool description to 80 - 150 chars: 'Execute arbitrary SQL queries on the MySQL database. Use for SELECT, INSERT, UPDATE, DELETE, and DDL operations. Returns CSV-formatted results or error details. WARNING: Destructive operations are irreversible.'
Add parameter description for 'query': 'The SQL statement to execute (SELECT, INSERT, UPDATE, DELETE, CREATE, DROP, ALTER, etc.). Max 10,000 chars. Avoid TRUNCATE without confirmation.'
Document output schema: 'Returns TextContent with type=text. For SELECT/SHOW: CSV format with headers. For INSERT/UPDATE/DELETE: row count and affected rows. For errors: error message and recovery hint.'
Add tool annotations to the Tool object: destructiveHint=true (since WRITE risk), idempotentHint=false (since repeated executions may have side effects).
Implement a confirmation pattern for destructive queries: add a 'dry_run' boolean parameter (default false) or require explicit confirmation via a separate 'confirm_destructive_query' tool before executing DELETE/DROP/TRUNCATE.

Spec posture evidence

Inferred effective spec: <=2025-11-25.

Relies on Logging (deprecated) - log to stderr or use OpenTelemetry

Score history

Overall score trend

First recorded score · v2 rubric

43/100

Scored	Grade	Overall	Spec posture	Rubric
2026-09-23	F	43	<=2025-11-25	v2