Expose Semantic Search through MCP

Ideas & Features
1

Fibery already has AI Search/Semantic Search inside the app, and it is one of the most valuable features for using Fibery as a knowledge base or second brain.

However, this semantic search capability is currently not exposed through the MCP Server. The MCP search tool seems to rely on full-text search only.

It would be extremely useful if external AI agents like ChatGPT, Claude, Cursor, etc. could access Fibery’s semantic search index directly through MCP.

A tool like semantic_search could return the most relevant entities based on meaning, with optional filters by database/space and maybe relation/reference expansion.

This could become a major feature for users who want Fibery to be not only a workspace, but also a true AI-accessible second brain, especially for users with hundreds or thousands of notes, documents, decisions and linked entities.

Is exposing Semantic Search through MCP something already planned or being considered?

Thanks!

2

Hi! I’m adding my vote for this. It would be very useful to have the option to use semantic search via MCP. We are usually doing research with Claude MCP with qualitative data stored in Fibery and in one conversation we are combining multiple data to create an Artifact. I’m also curious if this is planned.