Aura Workshop documentation
Aura Workshop is a model-agnostic AI agent orchestration platform — a desktop app for macOS, Windows, and Linux and a headless daemon image for servers. One surface gives you 25+ LLM providers, multi-agent teams, self-paced autonomous agents, a native memory and knowledge stack, automation (schedules, chat listeners, webhooks, workflows), and 100+ tool integrations. This documentation covers Aura Workshop 1.36.2 and the Aura Enterprise Portal.
Explore the docs
Getting Started
Install on desktop or as a daemon, first launch, and the setup wizard.
Workspace & Tasks
The agent workspace, the composer, and the four task modes — Execute, Plan, Goal, Auto.
Projects
Group tasks, files, and context into projects with their own goals.
Multi-Agent Teams
Run cloud models, ACP peers, and CLI tools together — Discussion, Plan, Goal, Research, and parallel fan-out.
Autonomous Agents
Self-paced loops that run, carry state across wakes, fan out, and stop themselves — goal and monitor kinds, with guardrails.
Skills & Agent Tools
The 100+ agent tools and the bundled skills that extend what an agent can do.
Automation & Workflows
Schedules, chat listeners, webhooks, slash commands, and visual workflows.
Memory & Knowledge
Hybrid recall, layered distillation, CodeGraph, the Wiki, and the interactive knowledge graph.
Models & Providers
Connect providers, set per-model parameters, and use Aura Routing 2.0.
Settings
Every settings tab — security, memory, routing, credentials, and more.
Deployment & Security
The daemon image, remote deployment, the web viewer, isolation layers, and guardrails.
Integrations
Connect Slack, Gmail, GitHub, Notion, and 100+ services, plus MCP servers.
API Reference
The full REST API, WebSockets, and the OpenAI-compatible gateway.
CLI & Reference
The aura-daemon CLI, network ports, data locations, and license tiers.
Enterprise Portal
Managed governance: enrollment, policies, cost, audit, and the central LLM gateway.