Skip to content

Liferay AI Commerce Accelerator documentation

Everything here documents the Liferay AI Commerce Accelerator (AICA): how to install it, how it works internally, and the platform behaviour it has to work around.

Community project

Not an official Liferay product and not covered by Liferay Support. About this project

AICA generates data for Liferay Commerce; it is not a commerce platform in its own right. If the platform itself is new to you, start with the Liferay Commerce documentation.

New to the project? Start with the Quick Start Guide.

I want to…

I want to… Go to
Install AICA and generate my first catalog Quick Start Guide
Run the seeder against an existing DXP, without LDM Local Execution & Operations Guide
Target a remote Liferay Experience Cloud (SaaS) instance SaaS Targeting Guide
See what AICA can actually generate Features & Capabilities
Understand how a generation run flows through the system Architectural Overview
Find which client extension owns a given responsibility System Map
Change what a generator asks the model for Prompt Templates
Drive AICA from an AI agent MCP Server
Configure Liferay's API authentication JSON Web Services Guide
Work on the shared SDK rather than AICA itself LDK / Accelerator SDK Blueprint
Understand a Liferay bug AICA works around Upstream Bugs & Limitations
Follow the repository's automation and release rules Automation Playbook

By subject

The table above routes on your question. This section lists the same documents by what they cover, for when you would rather browse than search.

Getting started

  • Quick Start Guide — deployment, configuration and execution, including the Sales Engineering Demo (LDM) path and manual developer setup.
  • Features & Capabilities — AI generation, real-time monitoring and visual assets, from a user's point of view.

Running it

  • Local Execution & Operations Guide — the runbook for executing AICA against an existing or restored DXP environment without LDM or .ldmp packages. The longest document here, and the one to read when something is misbehaving locally.
  • SaaS Targeting Guide — running the seeder locally while targeting a remote Liferay Experience Cloud environment, including why this bypasses browser CORS entirely.
  • MCP Server — connecting an AI agent over SSE to inspect, monitor, generate and clean up datasets programmatically.
  • Prompt Templates — the nunjucks syntax available to whoever writes a prompt: variables, conditionals, loops and filters. Templates are editable in the Configuration UI, so a prompt changes without a build.

How it works

  • Architectural Overview — the stateful workflow engine, entity dependencies, batch statuses, the WebSocket event contract and the security flows. Read this before the deep dives.
  • System Map — each client extension as a subsystem, with its responsibilities and data flow.
  • Architecture deep dives — the detailed specifications behind the overview: orchestration, the microservice split, API constraints, workflow state and UI standards.
  • LDK / Accelerator SDK Blueprint — the decoupled @liferay/accelerator-sdk repository that AICA and other accelerators build on.

Platform reference

  • JSON Web Services Guide — Liferay's /api/jsonws endpoints and when to prefer the Headless REST or GraphQL APIs instead.
  • Upstream Bugs & Limitations — upstream DXP bugs and undocumented behaviour that AICA has to accommodate, with current status for each.

Project

  • About these docs — what this project is, what it is not, and how the dates on each page are produced and enforced.
  • CONTRIBUTING — how to propose and land a change.
  • Code of Conduct — the standards expected of everyone taking part.
  • Automation Playbook — branch protection, release pipelines and automated backlog prioritisation, for maintainers and AI agents.
  • Repository README — project overview and the aica CLI command suite.
  • AGENTS.md — the canonical context for AI agents working in this repository.

Three community tools from the same maintainer. None is an official Liferay product or covered by Liferay Support.

Tool What it does
Liferay Docker Manager Run, manage and develop Liferay DXP and Portal environments in Docker.
Liferay AI Commerce Accelerator You are here. Generate and manage AI-driven commerce data in Liferay DXP.
Liferay Tunnel Securely expose a local Liferay environment on a public HTTPS subdomain.

Last Reviewed: 2026-10-15