Skip to main content
Version: 0.9.20

What is AlphaAgent?

AlphaAgent is an operating system for AI Agents that lets your team build, run, and collaborate with AI agents that work directly against your own data — all inside your own AWS account.

The product in one sentence

AlphaAgent gives analysts and builders a workspace where they can create AI agents, connect those agents to their databases, documents, and APIs, and then chat with them or run them on a schedule — while every piece of data, every conversation, and every result stays within the customer's own cloud boundary.

Two parts: Console and Studio

AlphaAgent is delivered in two parts. Understanding the split makes the rest of the documentation easier to follow.

Console (the portal)

The Console is a web portal hosted by PrometheusRL. You use it to:

  • Create your account and organization.
  • Manage billing and subscriptions.
  • Create and manage license keys.
  • Download the AlphaAgent Studio install bundle.
  • Monitor the health and usage of your deployed Studio.

The Console is a management portal only. It never connects to your AWS account and never sees your data.

Studio (the product you deploy)

Studio is the application your team actually works in. You deploy it once into your own AWS account using the bundle you download from the Console. Once it is running, your users sign in through your own identity provider and start building.

Everything Studio does — storing agent definitions, running code, holding conversations, building knowledge graphs from your documents — happens inside your AWS account. Nothing about your data is hosted by PrometheusRL.

[Screenshot: The Console portal showing the license and download pages]

What you can do in Studio

  • Build agents. Define AI agents with their own instructions and connect them to your data sources.
  • Compose swarms. Coordinate a team of specialist agents under a supervisor for more complex tasks.
  • Connect your data. Link databases, REST APIs, Snowflake, AWS data sources, and external tools.
  • Build knowledge graphs. Turn your documents into a queryable knowledge graph (AMPG) that agents can draw on.
  • Run code. Agents can run Python and shell code in an isolated sandbox to analyze data, produce tables, and generate charts.
  • Chat. Have rich conversations with your agents, with citations, visual reasoning maps, and file attachments.
  • Automate with workflows. Chain agents together into multi-step workflows and schedule them to run automatically.

Why it runs in your account

AlphaAgent is designed for organizations that cannot send their data to a third-party SaaS. Because Studio runs inside your own AWS account:

  • Your agent definitions, conversations, documents, datasets, and credentials never leave your cloud boundary.
  • Model inference runs on AWS Bedrock, pinned to a single residency zone (United States or European Union) that you choose.
  • The only information that ever flows out to PrometheusRL is license and usage information needed to keep your subscription healthy. See Licensing and telemetry for the exact details.

Where to go next