Introduction

xlai is a Rust-first AI integration workspace for building reusable AI calling flows across native applications and the browser. The design is inspired by moeru-ai/xsai.

The project is built around a unified API with pluggable backends. The long-term goal is to support API-based models, tool integration, skill management, knowledge retrieval, vector search, and local or device inference behind the same overall API model.

What works today

• Cargo workspace with native and wasm32 build support
• OpenAI-compatible chat backend
• Chat sessions with per-session tool registration
• Streaming chat output
• Configurable tool-call execution (concurrent by default, sequential optional)
• Unit/mock tests and ignored end-to-end tests
• GitHub Actions for build, test, formatting, clippy, and e2e

Where to go next

1. Getting started — clone, build, test, and optional JS package build
2. Configuration — .env and provider variables
3. Architecture — crates and request flow
4. Rust SDK — RuntimeBuilder, chat, agents, tools
5. JavaScript / WASM — @yai-xlai/xlai

For the full narrative and roadmap in the repository, see the README on GitHub.