Cognitor indexes folders so AI agents can search them by meaning
Cognitor is an open-source semantic search engine that watches a target folder and its subfolders. It automatically chunks, embeds, and indexes the files so they can be searched with natural language. The post says it offers a REST API and can be used as a standalone search engine, a vector database, or an app backend.
Key points
- It indexes a chosen folder and its subfolders.
- It chunks files, creates embeddings, and stores searchable data.
- It supports natural language queries through a REST API.
- It is aimed at both AI agents and human users.
- A worker process watches the folder and updates the index when files change.
Quick term guide
- open-source
- Software whose code is shared publicly so others can inspect, use, or change it.
- semantic search
- Search that looks for similar meaning, not just exact matching words.
- search engine
- A website like Google or Bing that helps you find information on the internet.
- vector database
- A special type of storage that saves text as numbers so similar meanings can be found quickly, commonly used for AI memory
- database
- A large collection of organized data used for search and analysis.
- AI agents
- AI agents are AI tools that can carry out steps toward a goal, not just answer once.
- embeddings
- A way of converting text into numbers so that similar meanings can be found and compared mathematically.
- embedding
- A way to turn text meaning into numbers so similar text can be found.