AgentOS

Open-source TypeScript AI agent framework

AgentOS is an open-source TypeScript AI agent framework for building production-ready autonomous agents with multi-agent orchestration, cognitive memory, multimodal RAG, AI guardrails, voice pipelines, and 11 LLM providers. A type-safe alternative to LangGraph, CrewAI, Mastra, and AutoGen.

Open-source TypeScript runtime for AI agents that remember, adapt, and spawn new tools. Agents write their own functions mid-task and the judge approves them; managers create specialists when the static roster does not cover the task. Cognitive memory grounded in neuroscience. 85.6% on LongMemEval-S. Apache 2.0.

See code examples
—GitHub Stars
—Forks
Agents that write their own tools
Mid-task TS function generation, judge-approved, sandboxed
Specialist spawning
spawn_specialist mints new agents at runtime
Cognitive memory
Ebbinghaus decay, reconsolidation, neuroscience-grounded
85.6% LongMemEval-S
0.4 below Emergence.ai (closed SaaS), +1.4 above Mastra
GDPR-ready guardrailsSelf-hosted on your infra

Runtime tool forging

Watch a tool get forged at runtime

Three HEXACO-distinct agents collaborate on a code review. When their static toolkit can't cover the task, the manager calls spawn_specialist, an LLM judge approves the spec, and all three invoke the forged tool on the next turn.

Three AgentOS agents with distinct HEXACO personalities collaborate on a code review, forge a new tool at runtime, the LLM judge approves the spec, and all three invoke it on the next turn.
Captured from node examples/emergent-hierarchical-spawning.mjs.View source on GitHub →
Wilds AI
Join the Wilds AI Discord
Official community for AgentOS and Paracosm support, developer onboarding, and questions.

PARACOSM

AI Agent Swarm Simulation Engine

Define a scenario as JSON. Run it with AI commanders that have different personalities. Watch their decisions compound into divergent civilizations from identical starting conditions. Built on AgentOS.

Launch SimulationLearn more about Paracosmnpm install paracosm

See AgentOS in action

Watch a live AgentOS agent handle a real workload end-to-end: tool use, memory recall, multi-step planning. Every clip is captured from a running runtime, no edits.

MIT LicensedOpen Source

AgentOS Workbench

The official desktop application for building, managing, and interacting with AgentOS experiences. Create powerful AI agents with a visual interface, test in real-time, and deploy anywhere.

  • Visual agent builder with drag-and-drop workflows
  • Real-time testing and debugging tools
  • One-click deployment to cloud or local
View on GitHub

Ready to Build Adaptive AI Agents?

Join developers building the next generation of intelligent systems with AgentOS.

DiscordBrowse CodeDocsWilds.aiTry Voice Chat

Built by

WE ARE THE FRAMERS

frame.dev@framersai