Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
GENIE_HOSTNoAlternate host for Genie (defaults to DATABRICKS_HOST)
GENIE_TOKENNoAlternate token for Genie (defaults to DATABRICKS_TOKEN)
DATABRICKS_HOSTYesDatabricks workspace URL, e.g. https://your-workspace.cloud.databricks.com
DATABRICKS_TOKENYesDatabricks personal access token
DATABRICKS_SCHEMANoSchema name (default: ecommerce_genie_ontology)
DATABRICKS_CATALOGNoCatalog name (default: ecommerce_genie_ontology)ecommerce_genie_ontology
DATABRICKS_OLTP_SCHEMANoOLTP schema name
DATABRICKS_WAREHOUSE_IDNoOptional SQL warehouse ID used for queries

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

CapabilityDetails
tools
{
  "listChanged": false
}
prompts
{
  "listChanged": false
}
resources
{
  "subscribe": false,
  "listChanged": false
}
experimental
{}

Tools

Functions exposed to the LLM to take actions

NameDescription
list_fraud_casesA

List the 15 fraud detection cases. Each case is a named SQL evidence pack.

list_fraud_agentsA

List the 10 Databricks Genie fraud specialists that share OLTP + dims/facts.

run_fraud_caseB

Run one fraud case (ids 01-15). Returns at most 50 evidence rows.

run_fraud_agent_casesC

Run every evidence pack owned by one of the 10 fraud specialists.

generate_historical_oltpC

Write customer, address, order, line, shipment, and entity_link tables.

generate_realtime_ordersC

Append 100-10000 new OLTP orders. year_window is latest | last_2 | last_3 | all.

etl_star_historicalC

Rebuild star-schema dims and fact_sales from OLTP (overwrite).

etl_star_cdcB

Apply Delta change feed from OLTP customer_order into fact_sales (append).

query_datasetA

Run a SELECT against dims, facts, or OLTP. LIMIT 50. Never dumps full tables.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

TDQS

B3.4/5.0

Scored across 9 tools

Disambiguation4/5

Most tools have clearly distinct roles: generation, ETL, fraud case listing/execution, and querying. The main ambiguity is between run_fraud_case and run_fraud_agent_cases, but their descriptions clarify single-case versus agent-owned case batches.

Naming Consistency5/5

All tool names follow a consistent snake_case verb_noun pattern, with recognizable prefixes like run_, list_, generate_, etl_star_, and query_. This makes the tool surface predictable and easy to navigate.

Tool Count5/5

Nine tools is a well-scoped size for this server's purpose. Each tool covers a distinct phase of the workflow—data generation, ETL, fraud case evaluation, and querying—without unnecessary redundancy.

Completeness4/5

The core lifecycle of generating OLTP data, building star-schema artifacts, running fraud cases, and querying results is well covered. Minor gaps include the lack of explicit schema/reset management or direct row-level CRUD, but agents can work around these using query_dataset and generation tools.

Maintenance

ActivityMaintained
ResponsivenessNo issues