Universal Database MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| DRYRUN | No | Enable dry-run mode (queries are not executed) | false |
| MYSQL_URI | No | MySQL connection URI (e.g., mysql://user:pass@host:3306/db) | |
| DUCKDB_PATH | No | Path to DuckDB database file or :memory: | |
| SQLITE_PATH | No | Path to SQLite database file | |
| POSTGRES_URI | No | PostgreSQL connection URI (e.g., postgresql://user:pass@host:5432/db) | |
| QUERY_TIMEOUT | No | Query timeout in seconds | 30 |
| ENABLE_LOGGING | No | Enable query logging to stderr | true |
| MYSQL_READONLY | No | Set to false to allow writes | true |
| RATE_LIMIT_RPM | No | Rate limit in requests per minute | 60 |
| DUCKDB_READONLY | No | Set to false to allow writes | true |
| MAX_RESULT_ROWS | No | Maximum number of rows returned in a query result | 1000 |
| SQLITE_READONLY | No | Set to false to allow writes | true |
| ALLOW_DESTRUCTIVE | No | Set to true to allow INSERT/UPDATE/DELETE/DROP | false |
| POSTGRES_READONLY | No | Set to false to allow writes (not recommended) | true |
| WHITELISTED_TABLES | No | Comma-separated list of table names that queries can access (optional) |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| logging | {} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| queryA | Execute a SQL query against the database. Supports parameterized queries for injection safety. In read-only mode, destructive operations (INSERT, UPDATE, DELETE, DROP, etc.) are blocked. |
| schemaA | Get database schema information: tables, columns, types, and constraints. |
| explainA | Get query execution plan without running the query. |
| healthA | Check database connection health and get server version info. |
| query_historyA | Get recent query execution history (most recent first). |
| snapshot_schemaA | Take a snapshot of the current database schema for later drift detection. |
| schema_diffA | Compare the current schema against the last snapshot taken with snapshot_schema. |
| list_databasesA | List all configured databases and their connection status. Returns: JSON with list of databases including key, type, database name, and connection status. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 8 tools
Each tool targets a distinct operation: schema introspection, query execution, plan explanation, health checking, history, schema snapshot/diff, and database listing. The schema-related trio (schema, snapshot_schema, schema_diff) and health/list_databases have adjacent concerns, but descriptions clarify boundaries.
Tool names mix styles: noun-only (schema, health), verb-only (query, explain), and snake_case compounds (query_history, snapshot_schema, schema_diff, list_databases). All are readable and lowercase, but there is no consistent verb_noun pattern.
Eight tools is well-scoped for a universal database server, covering introspection, execution, planning, health, history, and schema drift. Each tool earns its place without redundancy.
The set covers schema reading, arbitrary SQL execution, query plans, health checks, history, and schema drift detection, which covers most database workflows. Missing dedicated transaction control or database administration tools, though query can handle many of those operations.