Documentation

Orbit Developer Docs

Orbit is memory infrastructure for developer-facing AI products. Send signals, retrieve focused context, close the loop with feedback, and let memory quality improve over time.

Quick install

pip install orbit-memory

Orbit Setup Lab

Floating AI-assisted onboarding

Marvelous tone

Tap the Setup with AI helper that trails this page. Tell it your runtime (Cloud, Self-hosted, hybrid), your adapter, and we will build the exact prompt to copy into ChatGPT, Claude, or Cursor.

Prompt blueprint

prompt.txt

Prompt blueprint:
"Generate an Orbit MemoryEngine init snippet in Python using the Ollama adapter. Show ingest + retrieve call for entity_id 'alice' with conflict-aware metadata."

This keeps the docs humane and playful; the AI takes care of the boilerplate while you focus on the behavior you want.

Getting Started

Pick Cloud vs Self-Hosted setup, install Orbit, and run your first end-to-end memory loop in minutes.

>Quickstart >Installation & Setup Routes

Core Concepts

Understand the contract: event ingestion, retrieval quality, feedback signals, and inferred memory.

>Integration Guide >Event Taxonomy >Personalization

API Reference

Method-by-method SDK docs and endpoint-by-endpoint REST reference.

>API Reference >SDK Methods >REST Endpoints

Guides

Production-style examples for FastAPI, OpenClaw integration, and live chatbot testing.

>FastAPI Integration >OpenClaw Plugin >Examples

Operations

Deploy, configure, monitor, and troubleshoot Orbit with a Postgres-first runtime path.

>Cloud Dashboard >Deployment >Configuration >Monitoring >Troubleshooting

Metadata & Keywords

See the stack, keywords, and docs signals we highlight for SEO/clarity.

>Metadata Blueprint