Architecture
Why Origin Uses Markdown plus a Local Index
Origin stores human-readable Markdown records and uses a local database as the retrieval index, so AI memory stays inspectable and useful.
Article packet
Concepts
People who want AI memory they can inspect and trust
5 min read
01
Markdown is the durable record people can read, edit, export, and review.
02
The database stores retrieval indexes, graph structure, and search metadata.
03
This split keeps memory useful to agents without making it opaque to humans.
01
The problem with black-box memory
A memory layer can become risky when the only record lives inside a database or model profile that users cannot inspect. If the assistant retrieves stale or wrong context, it is hard to know where the mistake came from.
For long-running work, memory needs to be readable by people and searchable by agents.
02
Markdown is the record
Origin keeps memory artifacts in Markdown so people can open them, read them, correct them, export them, and reason about them without a special UI.
That does not mean Markdown has to be the only system. It means the human-readable artifact remains the source people can trust.
03
The local database is the index
Agents still need fast retrieval. Origin uses a local database for vector search, full-text search, graph context, provenance, and other metadata that make memories useful during an AI session.
The database helps find the right context. It should not be the only place where meaning is locked away.
04
Why the split matters
The Markdown-plus-index design gives both sides what they need: humans get control and readability, while agents get retrieval speed and context packaging.
That is the basis for Origin's trust story. Memory can be powerful without becoming invisible.
Keep memory readable and searchable
Origin pairs Markdown records with a local retrieval index so memory stays useful to agents and visible to people.
FAQ