Overview

What Chordian is, who it is for, and how to find your way around these docs.

Chordian is memory infrastructure for organisations. It takes the knowledge scattered across your CRM, your drive, your wiki, your inbox and your uploads, and turns it into a single typed, temporal memory that answers questions with citations.

The part that usually matters most in an evaluation: it can run entirely on your own infrastructure, with none of your content on ours.

What you get

Who these docs are for

You are…Start here
Evaluating ChordianQuickstart, then Core concepts
Integrating an applicationPlatform API and Memory API
Wiring an agent or IDEMCP integration
Assessing securitySecurity overview
An AI agent/llms.txt — every page is also available as .md

How the product is put together

At the centre is a memory: a complete, self-contained stack — application, vector engine, graph engine, cache and object store — that can be deployed in four postures without changing a line of its code.

Content passes the perimeter before it is stored; every answer is grounded in what was retrieved.

That single design decision — a memory as a deployable unit rather than a row in a database — is what makes the sovereignty story possible without forking the product. A feature that ships to the hosted service is simultaneously available to a customer running a box in their own data centre.

A note on honesty

These docs distinguish carefully between what is live, what is partial, and what is roadmap. Where a capability depends on how you deploy — several do — the condition is stated rather than glossed. If something here reads as hedged, that is deliberate.

Was this page helpful?
Report an issue

On this page