Press Monitor Wikidata MCP Server β Usage Guide
The Wikidata MCP Server mirrors the REST API with JSON-RPC tools designed for AI agents and automations. Tool names align with the OpenAPI operationId fields.
Base URL
https://mcp.pressmonitor.com/wikidata/v1
Authentication
All MCP calls require:
Authorization: Bearer YOUR_API_KEY
MCP Configuration Example (mcp.json)
{ "mcpServers": { "wikidata": { "command": "npx", "args": [ "-y", "mcp-remote", "https://mcp.pressmonitor.com/wikidata/v1", "--header", "Authorization: Bearer ${WIKIDATA_API_TOKEN}" ], "env": { "WIKIDATA_API_TOKEN": "" } } } }
List Available Tools
curl -s -X POST "https://mcp.pressmonitor.com/wikidata/v1" -H "Content-Type: application/json" -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'
Tool Names & Arguments (mapped to REST)
Below tools correspond to all API paths:
-
search_entities_by_label β
/search/by-label
Args:query_text(required),lang_code,limit,offset -
search_entities_by_description β
/search/by-description
Args:query_text(required),lang_code,limit,offset -
search_entities_by_claim_value β
/search/by-claim
Args:property(required),value,mode(exact|prefix|fulltext),lang_code,limit,offset -
suggest_entities β
/suggest-entities
Args:query_text(required),lang_code,limit,offset -
search_entities_by_type β
/search/by-type
Args:type(Q-id, required),query_text(optional),lang_code,limit,offset -
search_entities_by_property β
/find/by-property
Args:property(required),value,qualifier,rank,lang_code,limit,offset -
search_entities_by_date_range β
/search/temporal
Args:property,from,to,value,lang_code,limit,offset -
get_entity_by_id β
/entities/{id}
Args:id(Q-id, required),lang_code -
get_entity_claims β
/entities/{id}/claims
Args:id(Q-id, required),lang_code,limit,offset -
get_properties β
/properties
Args: (none)
Invoke a Tool - Example: Web Search
search_entities_by_label
curl -s -X POST "https://mcp.pressmonitor.com/wikidata/v1" -H "Content-Type: application/json" -H "Authorization: Bearer YOUR_API_KEY" -d '{ "jsonrpc":"2.0", "id": 1, "method":"tools/call", "params":{ "name":"search_entities_by_label", "arguments": { "query_text": "Paris", "lang_code":"en", "limit": 10 } } }'
search_entities_by_claim_value
curl -s -X POST "https://mcp.pressmonitor.com/wikidata/v1" -H "Content-Type: application/json" -H "Authorization: Bearer YOUR_API_KEY" -d '{ "jsonrpc":"2.0", "id": 2, "method":"tools/call", "params":{ "name":"search_entities_by_claim_value", "arguments": { "property":"P17", "value":"Q142", "mode":"exact", "limit": 20 } } }'
get_entity_by_id
curl -s -X POST "https://mcp.pressmonitor.com/wikidata/v1" -H "Content-Type: application/json" -H "Authorization: Bearer YOUR_API_KEY" -d '{ "jsonrpc":"2.0", "id": 3, "method":"tools/call", "params":{ "name":"get_entity_by_id", "arguments": { "id":"Q90", "lang_code":"en" } } }'
Result Envelope
MCP responses embed tool outputs in a JSON-RPC envelope. A typical success looks like:
{ "jsonrpc": "2.0", "id": 1, "result": { "content": [{ "type": "json", "json": { "data": [ /* results */ ], "meta": { /* search/claim metadata, totals, took, etc. */ } } }] } }
Best Practices
- Begin with label and description tools for discovery; refine with claim / property tools for graph-precise results.
- Keep
limitbetween 10β50 for interactive usage; increase for batch workflows. - Cache frequent
/entities/{id}and/propertiescalls to minimize latency and cost.
Support
Docs: https://www.pressmonitor.com/en/docs/wikidata-api-mcp
Email: support@pressmonitor.com