memoria
PricingDocsStatusSign in

Documentation

Everything you need to integrate Memoria. Every page is also available as raw markdown — append .md to any /docs URL, or fetch /llms.txt for the full site map in one request.

Quickstart

Get an API key, send your first memory, and recall it — under five minutes.

Read →

MCP server

Connect Memoria to Claude Code and other MCP clients. Twelve tools, OAuth 2.1 authentication, one install line.

Read →

CLI

Install the memoria CLI to wire up Claude Code in one line — including Claude Code on the web. Handles install, auth, and MCP config, and teaches the agent the mental model.

Read →

REST API

All Memoria REST endpoints, request/response shapes, and conventions. OpenAPI 3 spec available.

Read →

Brains

Brains are isolated memory workspaces. One per project, one per environment, or one per agent — your call.

Read →

Recall pipeline

How Memoria retrieves memories — dense vectors, BM25 keyword search, and graph traversal, fused and re-ranked.

Read →

Write pipeline

Seven LLM stages turn raw text into a bi-temporal knowledge graph — entities resolved, facts deduped, contradictions detected.

Read →

For AI agents

Memoria's docs are designed to be readable by both humans and agents. Every page renders as HTML at /docs/<slug> and as raw markdown at /docs/<slug>.md (or /docs/<slug>/raw). An AI-readable index of the whole site lives at /llms.txt.

memoria
DocsChangelogStatusPrivacyTermsllms.txt