Skip to main content
Glama
Sign In
Lagaam

Lagaam

Official
by lagaam-ai
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Remote

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
TRINO_HOSTNoTrino coordinator host.localhost
TRINO_PORTNoTrino coordinator port.8080
TRINO_USERNoTrino user.lagaam
LAGAAM_ENGINENoSet to pinot to use the native Pinot adapter.trino
LAGAAM_MAX_ROWSNoScanned-row estimate budget per query.
LAGAAM_AUDIT_LOGNoAudit JSONL file path.stderr
LAGAAM_AGENT_NAMENoIdentity stamped on the audit trail.anonymous
LAGAAM_METADATA_TTLNoMetadata cache TTL, seconds.300
LAGAAM_QUERY_TIMEOUTNoWall-clock seconds per query.300
LAGAAM_ALLOWED_TABLESNoComma list of catalog.schema.table grants. Required unless LAGAAM_ALLOW_ALL_TABLES is set to true.
LAGAAM_MAX_SCAN_BYTESNoScan-bytes budget per query, pre-execution.53687091200
LAGAAM_ALLOW_ALL_TABLESNoSet to true to run with no grant at all.
LAGAAM_MAX_RETURNED_ROWSNoRows returned to the agent per query; max 100000.1000
LAGAAM_MAX_INTERMEDIATE_ROWSNoRows the engine would build at its widest step.50000000

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_catalogsA

List every catalog, schema, and table you are allowed to query.

Call this first to ground yourself before describing tables or writing SQL — table names you have not seen here are guesses.

describe_tableA

Get the exact columns and types of one table. Read-only.

    Always describe a table before querying it; column names you have
    not seen here are guesses. It reads the engine's metadata and
    statistics, never the table's rows. A table outside your grant is
    refused before the engine is asked; a missing one errors with a
    pointer to list_catalogs. Answers are cached, 5 minutes by default,
    so a new column can lag. row_estimate is the engine's row-count
    statistic, null when it has none or can't be trusted (views, tables
    without stats, Pinot tables with a realtime part). Column comments
    are null when unset, and always on Pinot.
query_dataA

Run a read-only SELECT and get the rows back.

    Write a single SELECT in the engine's dialect. The query is checked
    for safety, priced against your budget, and executed with a row cap —
    so name the columns you need (no SELECT *), and add WHERE filters to
    keep the scan small. At most 1,000 rows come back by default (a
    larger LIMIT is lowered to the cap), and `truncated` marks a result
    the cap cut off. A query that touches any table outside your grant
    is refused before it runs. If it is rejected, the message says what
    to fix. Describe the tables first so column and table names are exact.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A4.4/5.0

Scored across 3 tools

Disambiguation5/5

Each tool targets a distinct step in the metadata-to-query workflow: catalog discovery, table introspection, and query execution. The descriptions explicitly state the intended ordering and cross-reference each other, leaving no realistic chance of misselection.

Naming Consistency5/5

All three names follow a uniform snake_case verb_noun pattern (list_catalogs, describe_table, query_data). The verbs are accurate and non-overlapping.

Tool Count4/5

Three tools is lean but each is essential to the read-only query lifecycle, so nothing is redundant. It sits at the low end of the comfortable range, with no room for auxiliary operations like budget or explain previews.

Completeness4/5

Discover-schema, describe-table, and run-SELECT cover the core read-only workflow end to end, including safety refusals and row caps. Minor gaps exist — no way to preview query cost/budget or search schema objects by name — but agents can work around them.

Maintenance

ActivityActive
ResponsivenessLimited data

Resources