MHTL Warehouse MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| WAREHOUSE_DB_PATH | No | Absolute path to the DuckDB file. If unset, defaults to ../data/mock.duckdb relative to dist/. |
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
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_tablesA | List all tables and views in the warehouse with their row counts. Use this first to discover what data is available. |
| describe_tableA | Return the column names and data types for a given table or view. Use this to learn the schema before writing a query. |
| queryA | Execute a read-only SELECT/WITH SQL query against the DuckDB warehouse and return the rows. Tables live in the 'main' schema (e.g. dms_customer, dms_contract, dms_contact_note, los_loan_app). A LIMIT is applied automatically if you omit one. Use ? placeholders with the |
| find_customerA | Search customers by (partial) first name, last name, or customer id. Returns matching customers with their internal oid — pass that oid (or the name) to customer_insights. |
| customer_insightsA | Given a customer name or id, return a consolidated insight profile: identity, all contracts, delinquency/NPL summary, outstanding & overdue balances, recent collection contact notes, and loan applications. This is the tool for requests like 'get the insights of the customer Max'. |
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 5 tools
Each tool has a clearly distinct purpose: find_customer for searching customers, customer_insights for consolidated profile, list_tables and describe_table for schema discovery, query for arbitrary SQL. No overlap.
Four tools follow a consistent verb_noun pattern (find_customer, customer_insights, describe_table, list_tables). 'query' deviates as a single verb, but it's a common term and doesn't create confusion.
Five tools is well-scoped for a warehouse MCP server. It provides essential operations (schema discovery, customer lookup, SQL query) without unnecessary bloat.
Covers the main use cases: discovering schema, finding customers, getting insights, and querying. Minor gaps like a dedicated contract search tool, but SQL can cover those needs.