• IDENTIFIERS
  • PROPERTIES
  • MATCHES
  • HUMAN

How to find specific entities with your AI

Connect Wikidata to your agent to bridge the gap between vague descriptions and exact database IDs. Your AI handles the heavy lifting of semantic matching.

Ask AI about this page

Short answer

How can I find Wikidata items without an ID?

Your AI uses semantic similarity to match your natural language descriptions to the closest Wikidata items and properties. It translates your intent into precise identifiers through vector searches. You get exact IDs and property keys ready for further queries.

Where your data lands

The outcomes of a search

The AI maps your text to the Wikidata structure.

IDENTIFIERS

Exact Q-codes

The AI finds the specific Q-numbers for items you describe. You get the precise IDs needed for structured data tasks.

PROPERTIES

Property P-codes

It identifies the correct property IDs for relationships. This lets you query specific attributes like 'date of birth' or 'place of origin'.

MATCHES

Similarity scores

The agent calculates how closely a concept matches a known entity. You see how confident the AI is in its match.

HUMAN

Manual verification

You review the suggested matches to ensure the semantic similarity aligns with your intent. This prevents errors from ambiguous descriptions.

The workflow

What your AI does when a search starts.

The agent moves from vague concepts to structured Wikidata data.

  1. Semantic item search

    The AI takes your description and looks for the most relevant items in the Wikidata database. It finds entities that match the meaning of your words.

    search_items_vector

  2. Property matching

    If you are looking for a specific attribute or relationship, the AI scans the property index. It finds the correct P-code for the concept you mentioned.

    search_properties_vector

  3. Confidence check

    The agent compares the vector of your request against the found items. This ensures the match is actually relevant to your prompt.

    get_similarity_score

  4. Final identification

    The AI presents the specific IDs and names. You receive the exact identifiers required for your next data operation.

    search_items_vector

Try it

Copy these to start.

Use these prompts to begin your search.

Starting points

Swap the bracketed text for the actual concepts or entities you are looking for.

Wikidata Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_W84JNbFgMiFDqgSOAzXVMSCHncfEanhMXyqzOlEz/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Wikidata capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "wikidata-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_W84JNbFgMiFDqgSOAzXVMSCHncfEanhMXyqzOlEz/mcp"
    }
  }
}
Copy into chat04

  • Find the Wikidata ID for [the concept of a specific type of architecture].

  • Search for the property that describes [the relationship between a person and their birthplace].

  • I'm looking for [a specific historical figure]; find their Wikidata Q-code using semantic similarity.

  • What is the property ID for [the concept of population density] in Wikidata?

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect Wikidata once, then ask.

Click the link to connect the MCP. Your credentials stay encrypted, and you can immediately start querying Wikidata through your preferred AI client.

Connect Wikidata to your AI

FAQ

How this task behaves

  • 01

    Can the AI edit Wikidata entries?

    No. This connector only allows the AI to search and retrieve items and properties.

  • 02

    What if the AI can't find the exact item?

    The AI will return the closest semantic matches based on the similarity score. You will need to manually verify if the match is correct.

  • 03

    Does it work with exact IDs?

    Yes, but its main strength is finding items when you only have a description or a concept.

  • 04

    Can the AI delete properties?

    No. The tools provided are strictly for searching and scoring similarity.

  • 05

    How does it handle typos?

    Because it uses vector search, it focuses on the meaning of your words rather than exact spelling, which helps with typos.

  • More questions about Wikidata? See everything the Wikidata Connector can do

Connect Wikidata to Claude, Cursor, ChatGPT & more