Nor is built by the people who use it. To tend this commons well, we look at what parts of the archive get visited, what people search for, and where the site is slow. Nothing is sold, nothing is shared with advertisers.
If you agree, we’ll store analytics cookies. You can change your mind any time. What we store.
Nor / MCP
The Nor Collection publishes a Model Context Protocol server so that MCP-compatible AI assistants can query the archive directly — searching entries, retrieving full records, and mapping designer networks without leaving your research session. No authentication required for public data.
Five tools are available to any connected assistant. You can search the archive by keyword or phrase and get back matching entries with titles, dates, designers, and categories. You can run a semantic search — describe a visual quality or concept in plain language and retrieve works that match. Given any archive entry, you can find visually similar works drawn from across the 14,000-entry collection. You can pull the full record for a single entry: title, description, practitioners, clients, categories, tags, and image. And you can map a designer’s network — their works and the collaborators they appear alongside. These tools are designed for research use: tracing influences, surfacing connections, and building context around specific works or practitioners.
The MCP endpoint is https://norcollection.ca/api/mcp. In Claude, go to Settings → Integrations and add a new MCP server using the configuration below. Most other MCP-compatible assistants follow the same pattern — consult your assistant’s documentation for where to add remote server entries.
{
"mcpServers": {
"nor-collection": {
"url": "https://norcollection.ca/api/mcp"
}
}
}The MCP server provides read access only. It cannot create, modify, or delete records, and it has no access to unpublished or restricted content. What your assistant can see through MCP is exactly what any visitor to the public site can see — the same published entries, the same practitioner profiles, the same image records. There is no elevated access, no authentication bypass, and no exposure of internal data.
MCP is designed for research and exploration within an AI assistant. If you are writing code, building an integration, or need structured programmatic access to the archive’s data, the GraphQL API is the right tool. It offers full query flexibility across the same underlying dataset.
API documentation →