Skip to main content
npmCIStarsForksTypeScriptLicense

AgentOS

Open-source TypeScript runtime for autonomous AI agents — unified graph orchestration, cognitive memory, streaming guardrails, voice pipeline, 21 LLM providers.

npm install @framers/agentos

Quick Start

import { generateText, streamText, generateImage, agent } from '@framers/agentos';

// Text — just set the provider, AgentOS picks the model
const { text } = await generateText({ provider: 'openai', prompt: 'Explain QUIC.' });

// Images — cloud or local (Ollama, SD WebUI)
const poster = await generateImage({
provider: 'stability',
model: 'stable-image-core',
prompt: 'Art deco travel poster for a moon colony',
providerOptions: { stability: { stylePreset: 'illustration' } },
});

// Streaming
for await (const delta of streamText({ provider: 'anthropic', prompt: 'Compare TCP vs UDP.' }).textStream) {
process.stdout.write(delta);
}

// Stateful agent with sessions and personality
const tutor = agent({
provider: 'openai',
instructions: 'You are a networking tutor.',
personality: { openness: 0.9, conscientiousness: 0.8 }, // HEXACO
});
const session = tutor.session('demo');
await session.send('What is QUIC?');
await session.send('How does it compare to HTTP/2?');

System Architecture

Full architecture guide →

Core Features

Multimodal Provider API

Text, images, embeddings, and speech from one API. 21 cloud providers (OpenAI, Anthropic, Gemini, Stability, Replicate) + local (Ollama, SD WebUI). Auto-fallback between providers.

Deep Research Agents

mission() API with goal-first planning, multi-source search, grounding verification, and human-in-the-loop review. Agents plan their own steps, discover tools, and self-evaluate.

Emergent Capabilities

Agents forge new tools at runtime — compose (chain existing tools) or sandbox (isolated V8 with allowlists). LLM-as-judge safety review, tiered promotion, portable YAML export.

Voice & IVR Pipeline

Full-duplex voice with 3 endpoint detection modes, barge-in handling, 28 STT/TTS providers, diarization. Twilio/Telnyx/Plivo telephony bridging for production IVR.

Graph Orchestration

Three authoring APIs — AgentGraph, workflow() DSL, mission() — compile to one IR. judgeNode for evaluation, checkpointing for time-travel, streaming events.

Cognitive Memory

Ebbinghaus decay, spreading activation, Baddeley working memory, personality-driven encoding (HEXACO). GraphRAG retrieval with episodic-to-semantic consolidation.

Streaming Guardrails

5-tier pipeline: PII redaction (regex + NLP + NER + LLM), ML classifiers (ONNX BERT), topicality drift, code safety (OWASP), grounding guard (NLI). Sentence-boundary buffered.

Evaluation Framework

Dataset-driven evals with candidates, graders, and experiments. LLM prompt runner and HTTP endpoint runner. Compare baseline vs challenger. Drizzle ORM with SQLite/Postgres.

Capability Discovery

3-tier semantic discovery: category summaries (150 tokens) → top-5 matches (200 tokens) → full schemas on demand. 89% token reduction. Agents self-discover tools mid-conversation.

Provenance & Audit

Signed event ledger (Ed25519 + SHA-256 hash chain), soft-delete tombstones, revision history, autonomy guard. Merkle anchoring for tamper-evident external verification.

37 Channel Adapters

Telegram, Discord, Slack, WhatsApp, Twitter/X, LinkedIn, email, SMS, and 29 more. Multi-channel routing, content adaptation engine, and custom adapter API.

Immutable Agents

Sealed storage policy, toolset pinning, secret rotation, soft-forget memory. Full provenance audit trail. Deploy agents that cannot be tampered with after initialization.