Query Language

Marmot provides a query language for searching and filtering assets in your catalog. The language supports free-text search, field-specific filters, comparison operators, and boolean logic.

Free-Text Search

Free-text queries search across asset names, descriptions, and metadata:

user orders

For more precise results, use field filters.

Field Filters

Field filters allow you to query specific attributes of assets using prefixes:

Asset Type (@type)

Filter assets by their type:

@type: "table"
@type: "topic"
@type: "bucket"

Provider (@provider)

Filter assets by their provider or platform:

@provider: "kafka"
@provider: "postgres"
@provider: "s3"

Asset Name (@name)

Filter by asset name:

@name: "users"
@name contains "order"
@name: "customer*"

Kind (@kind)

Filter by the type of resource in Marmot:

@kind: "asset"
@kind: "glossary"
@kind: "team"

Custom Metadata (@metadata.*)

Query any custom metadata field you've added:

@metadata.team: "data-platform"
@metadata.environment: "production"
@metadata.owner: "alice@company.com"
@metadata.region: "eu-west-1"

You can access nested metadata using dot notation:

@metadata.config.retention: "7d"
@metadata.tags.compliance: "pii"

Operators

Marmot supports various operators for precise filtering:

Operator	Description	Example
: or =	Exact match	@type: "table"
contains	Substring match	@name contains "customer"
!=	Not equal	@metadata.environment != "test"
>	Greater than	@metadata.partitions > 10
<	Less than	@metadata.rows < 1000000
>=	Greater than or equal	@metadata.replicas >= 3
<=	Less than or equal	@metadata.retention <= 30

Wildcards

Use * for wildcard matching:

@name: "customer*"           # Starts with "customer"
@name: "*-prod"              # Ends with "-prod"
@metadata.team: "*platform*" # Contains "platform"

Range Queries

Query numeric ranges efficiently:

@metadata.partitions range [5 TO 20]
@metadata.size range [1000 TO 50000]

Boolean Logic

Combine filters with AND, OR, and NOT for complex queries:

# All Kafka topics in production
@type: "topic" AND @provider: "kafka" AND @metadata.environment: "production"

# High-priority or critical assets
@metadata.priority: "high" OR @metadata.criticality: "critical"

# Production assets excluding test data
@metadata.environment: "production" AND NOT @name contains "test"

Complex Queries with Parentheses

Use parentheses to control query logic:

(@type: "table" OR @type: "view") AND @provider: "postgres"

@metadata.team: "platform" AND (@metadata.status: "active" OR @metadata.priority: "high")

(@provider: "kafka" OR @provider: "sqs") AND NOT @metadata.environment: "*test*"

Examples

Find all Kafka topics:

@type: "topic" AND @provider: "kafka"

Find assets owned by a specific team:

@metadata.team: "data-platform"

Find Kafka topics with more than 10 partitions:

@type: "topic" AND @provider: "kafka" AND @metadata.partitions > 10

Find non-production tables:

@type: "table" AND @metadata.environment != "production"

Combine free-text with filters:

order @type: "table" OR @name contains "order"

Complex boolean query:

@metadata.criticality: "high" AND (@metadata.status: "degraded" OR @metadata.alerts > 0)

Query Patterns

Iterative refinement: Start with broad queries and add filters to narrow results:

# Start with
kafka

# Refine to
@provider: "kafka" AND @metadata.environment: "production"

# Further refine
@provider: "kafka" AND @metadata.environment: "production" AND @metadata.partitions > 5

Wildcard exploration: Use wildcards when exact names are unknown:

@name: "*customer*" AND @type: "table"

Hybrid queries: Combine free-text with field filters:

payment @type: "topic" AND @metadata.team: "payments"