Skip to content
Documentation Hub

AgentOS Documentation

Complete API reference, 0+ guides, and examples for building with AgentOS

API Reference

Split TypeDoc documentation for @framers/agentos: a stable public API surface for application developers and a deeper module-level reference for internal subsystems.

Public API stays concise; Module API exposes the deeper class and subsystem surface

Source Code

Explore the full source code, examples, and contribute to AgentOS on GitHub. Apache 2.0 licensed.

Getting Started

New to AgentOS? Start here with installation instructions and your first agent.

npm install @framers/agentos

// Set a provider key via env or pass it inline
// export OPENAI_API_KEY=sk-...
import { generateText } from '@framers/agentos'

const result = await generateText({
  provider: 'openai',
  apiKey: process.env.OPENAI_API_KEY, // optional — auto-detected from env
  prompt: 'Explain quantum computing in one sentence.',
})
console.log(result.text)
Read the Architecture Guide →

Development Note

Documentation is auto-generated from source code. To regenerate locally:

pnpm --filter @framers/agentos run docs   # Generate public/module API docs + search index
pnpm --dir apps/agentos.sh dev:full         # Watch docs and run the site locally
WHITEPAPER · COMING SOON

The AgentOS Technical Whitepaper

The full architecture and methodology, written for engineers and researchers who want a citable PDF instead of scrolling docs. Cognitive memory pipeline, classifier-driven dispatch, HEXACO personality modulation, runtime tool forging, full LongMemEval-S/M and LOCOMO benchmark methodology with bootstrap CI math, judge-FPR probes, per-stage retention metrics, and reproducibility recipes.

Architecture
Generalized Mind Instances, IngestRouter / MemoryRouter / ReadRouter, 8 cognitive mechanisms with primary-source citations
Benchmarks
LongMemEval-S 85.6%, LongMemEval-M 70.2%, vendor landscape, bootstrap CI methodology, judge FPR probes, full transparency stack
Reproducibility
Per-case run JSONs at seed=42, single-CLI reproduction, MIT-licensed bench at github.com/framersai/agentos-bench