lexigram-ai

AI Layer for Lexigram Framework - Native LLM, Vector, RAG integration

Overview

lexigram-ai is the AI orchestration layer for the Lexigram Framework. It is a thin coordinator package that wires together independent AI sub-packages (lexigram-ai-llm, lexigram-ai-rag, lexigram-ai-agents, etc.) via the framework's DI container and entry-point discovery, and exposes a single AIModule entry point for application composition. Zero-config usage starts with sensible defaults.

Full documentation: docs.lexigram.dev

Install

uv add lexigram-ai
# Optional extras
uv add "lexigram-ai[llm,vector,rag,governance,observability]"

Quick Start

from lexigram import Application
from lexigram.di.module import Module, module

from lexigram.ai import AIModule

@module(imports=[AIModule.configure()])
class AppModule(Module):
    pass

app = Application(modules=[AppModule])
if __name__ == "__main__":
    app.run()

Configuration

Zero-config usage: Call AIModule.configure() with no arguments to use defaults.

Option 1 — YAML file

# application.yaml
ai:
  llm:
    default_provider: "openai"
    openai:
      api_key: "${OPENAI_API_KEY}"
  rag:
    enabled: true
  governance:
    enabled: true
    monthly_budget: 1000.0

Option 2 — Profiles + Environment Variables (recommended)

export LEX_PROFILE=production
# Environment variables for each field

Option 3 — Python

from lexigram.ai.config import AIConfig
from lexigram.ai import AIModule

config = AIConfig(...)
AIModule.configure(config)

Config reference

Module Factory Methods

Key Features

• AI orchestration: Thin coordinator wiring all AI sub-packages via DI
• Entry-point discovery: Discovers sub-packages via lexigram.ai.subsystems entry points
• Multi-provider support: Integrates LLM, RAG, agents, memory, governance, and observability
• Production security: Validates API keys in production environments

Testing

async with Application.boot(modules=[AIModule.stub()]) as app:
    # your test code
    ...

Key Source Files