Overview

Arch is an AI-native proxy server and the universal data plane for AI - one that is natively designed to handle and process AI prompts, not just network traffic.

Built by contributors to the widely adopted Envoy Proxy, Arch helps you move faster by handling the pesky low-level work in AI agent development—fast input clarification, intelligent agent routing, seamless prompt-to-tool integration, and unified LLM access and observability—all without locking you into a framework.

In this documentation, you will learn how to quickly set up Arch to trigger API calls via prompts, apply prompt guardrails without writing any application-level logic, simplify the interaction with upstream LLMs, and improve observability all while simplifying your application development process.

High-level network flow of where Arch Gateway sits in your agentic stack. Designed for both ingress and egress prompt traffic.

Get Started

This section introduces you to Arch and helps you get set up quickly:

Overview

Overview of Arch and Doc navigation

overview.html

Intro to Arch

Explore Arch’s features and developer workflow

intro_to_arch.html

Quickstart

Learn how to quickly set up and integrate

quickstart.html

Concepts

Deep dive into essential ideas and mechanisms behind Arch:

Tech Overview

Learn about the technology stack

../concepts/tech_overview/tech_overview.html

LLM Provider

Explore Arch’s LLM integration options

../concepts/llm_provider.html

Prompt Target

Understand how Arch handles prompts

../concepts/prompt_target.html

Guides

Step-by-step tutorials for practical Arch use cases and scenarios:

Prompt Guard

Instructions on securing and validating prompts

../guides/prompt_guard.html

Function Calling

A guide to effective function calling

../guides/function_calling.html

Observability

Learn to monitor and troubleshoot Arch

../guides/observability/observability.html

Build with Arch

For developers extending and customizing Arch for specialized needs:

Agentic Workflow

Discover how to create and manage custom agents within Arch

../build_with_arch/agent.html

RAG Application

Integrate RAG for knowledge-driven responses

../build_with_arch/rag.html