RDF4J MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| RDF4J_MCP_MAX_LIMIT | No | Maximum query LIMIT | 10000 |
| RDF4J_MCP_DEFAULT_LIMIT | No | Default query LIMIT | 100 |
| RDF4J_MCP_QUERY_TIMEOUT | No | Query timeout (seconds) | 30 |
| RDF4J_MCP_RDF4J_SERVER_URL | No | RDF4J server URL | http://localhost:8080/rdf4j-server |
| RDF4J_MCP_DEFAULT_REPOSITORY | No | Default repository ID |
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": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| sparql_selectA | Execute a SPARQL SELECT query and return results as JSON. Use this for queries that return tabular data with variable bindings. |
| sparql_constructA | Execute a SPARQL CONSTRUCT or DESCRIBE query, return Turtle.Use this for queries that return RDF triples/graphs. |
| sparql_askA | Execute a SPARQL ASK query and return a boolean result. Use this for yes/no questions about the data. |
| describe_resourceA | Get all triples about a resource (subject or object). Returns Turtle format with human-readable summary. |
| search_classesA | Find classes in the ontology. Can filter by name pattern. Returns class IRIs with labels and comments. |
| search_propertiesA | Find properties in the ontology. Can filter by pattern, domain, or range. Returns property IRIs with labels, domains, and ranges. |
| find_instancesA | Find instances of a class. Returns instance IRIs with labels. |
| get_schema_summaryB | Get an overview of the ontology schema including statistics, main classes, properties, and namespaces. |
| list_repositoriesA | List all available RDF repositories. Returns repository IDs, titles, and access permissions. |
| get_namespacesA | Get namespace prefix mappings for a repository. Returns prefix-namespace pairs for use in SPARQL queries. |
| get_statisticsC | Get statistics about a repository including triple counts, class counts, property counts, and more. |
| select_repositoryA | Select a repository to use as the default for subsequent operations. This persists until another repository is selected. |
| get_current_repositoryA | Get the currently selected default repository ID. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| explore_knowledge_graph | Guided exploration of a knowledge graph. Provides context about the schema and suggests exploration paths. |
| write_sparql_query | Help write a SPARQL query from a natural language description. Provides schema context to assist with query construction. |
| explain_ontology | Explain the structure and meaning of an ontology or schema. Describes classes, properties, and their relationships. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| Repository List | List of all available RDF repositories |
TDQS
Scored across 13 tools
Most tools have clearly distinct purposes (e.g., sparql_select vs. sparql_ask vs. sparql_construct). However, describe_resource overlaps significantly with sparql_construct as both return triples, which could cause misselection.
All tool names follow a consistent snake_case verb_noun pattern (e.g., list_repositories, get_statistics, select_repository). The sparql_* tools use a predictable sparql_<query_type> convention, maintaining overall consistency.
13 tools is well-scoped for an RDF4J server, covering querying, schema exploration, and repository management without feeling bloated or sparse.
The tool set provides strong coverage for querying, schema discovery, and repository selection. Minor gaps include no SPARQL UPDATE/insert/delete operations and no explicit repository creation, but these are not critical for a read-oriented RDF explorer.