Vera, your knowledge base

Knowledge base and edge server · Layer 1

"Reads everything. Remembers everything. Explains it clearly."

What Vera does

Give Vera your PDFs, notes, code and markdown. She indexes them on your machine and answers your questions with citations back to the source.

Vectors and a graph

Vector search finds similar passages. A knowledge graph captures how things relate. Together they answer better than either alone.

Honest answers

When Vera doesn't have enough in your documents, she says so instead of guessing.

Local-first

Your index lives on your machine in a ChromaDB vector store. Use Groq for a free cloud model, or Ollama to stay fully offline.

A collection per topic

Keep study notes, recipes, project plans, legal research or a codebase's docs in separate collections.

The edge server role

In the platform, Vera is also the name of the public edge server: a Linode VPS behind Caddy that hosts the central bus hub.

Same engine as the platform

Vera Home is a thin client over the Kevin daemon's knowledge-base API, so it improves when the platform does.

How it works

Four steps from a folder of files to a cited answer.

  1. Point Vera at a folder.
  2. She splits your files into chunks and indexes them in a vector store on your machine.
  3. When you ask a question, she retrieves the best passages and the related items from the graph.
  4. A model, Groq or local Ollama, writes the answer with citations.

Where it fits in the four layers

1 Infrastructure & roles Vera, Grant, the bus, the device registry
2 Data pipeline & state Flo, DynaRAG, the carousel
3 Reasoning & inference Rick, Annie, Max, Hank
4 UI & collaboration Marco, Noted, Norm, the graph

The Vera server role is layer 1: hosting, DNS and the central bus. Vera Home, the personal app, asks the daemon's knowledge-base API, which is where Annie's collections in layer 3 live.