View on GitHub

Lightspeed Core Stack

Lightspeed Core Stack

Lightspeed core stack

About The Project

Lightspeed Core Stack (LCS) is an AI-powered assistant that provides answers to product questions using backend LLM services, agents, and RAG databases.

Logo

Basic info

Overview

Getting started

User docs

Deployment Guide

Configuration options

Agent skills

A2A [Agent-to-Agent] Protocol

RAG configuration guide

BYOK guide

OKP guide

Authentication and Authorization

User data collection

Sentry error tracking integration

Splunk integration

OpenTelemetry tracing

Developer doc

Architecture

Architecture diagram

Installation on Linux

Installation on MacOS

Database structure

Providers

Container Orchestration Guide

Contributing guide

Conversation history

Persistent storage

Quota scheduler

API

OpenAPI specification

Conversations API

Query endpoint

Streaming query endpoint

OpenResponseAPI specification

Models

Common

Database

Requests

Successful responses

Error responses

Compaction

Class diagrams

Common

Database

Requests

Responses

Testing

Testing

End-to-End Tests Guide

List of e2e scenarios

Maintenance

Status of Lightspeed Core Stack versions

Supported versions

Unsupported versions

Migration guides

Branching

Releasing

LTS process overview

Design documents

** Agent skills **

Spike

Design

** Automatic conversation expiration **

Design

** BYOK PDF support **

Spike

Design

** Conversation compaction **

Spike

Design

** Supporting backport changes for releases **

Design

** Human-in-the-loop (HIL) **

Spike

Design

** Llama Stack config merge (unified lightspeed-stack.yaml) **

Spike

Design

Profiles (Deployment Guide)

** Low overhead deployment for server mode **

Design

** OpenTelemetry tracing in Lightspeed Core **

Design

Demos

LCORE introduction

CodeRabbitAI

Lunch and Learn 2026

LCORE weak points for AI-driven agentic flow