AI-DLC Introduces Adaptive Workflow Steering Rules for AI Coding Agents

Overview of AI-DLC

AI-DLC (AI-Driven Development Life Cycle) is a system designed to convert AI coding assistants into structured and verifiable software-delivery workflows. This initiative aims to bring more order and reliability to the use of AI in software development, moving beyond simple code generation to integrated workflow management.

Broad Compatibility

The core of AI-DLC is harness-neutral, allowing it to run natively across a range of popular AI coding tools. Supported platforms include Claude Code, Kiro CLI, Kiro IDE, Codex CLI, Cursor, opencode, and GitHub Copilot. This broad compatibility allows developers using different AI assistants to integrate AI-DLC into their existing setups.

Installation and Setup

Installation is available for macOS, Linux, WSL, and Windows, with specific commands provided for each environment. The installer sets up the native `aidlc` command and all necessary harness runtimes without requiring Bun or Node.js. For Windows, it installs for the current user account and modifies the User PATH, while macOS, Linux, and WSL users may need to follow PATH instructions.

A manual installation path is also available for users who prefer to manage project files directly or cannot install a native executable. This method requires Bun and involves copying runtime directories from a downloaded archive into the project.

Configuring and Using AI-DLC

After installation, users configure AI-DLC by selecting their preferred harness (e.g., `claude`, `kiro`, `copilot`) using the `aidlc config --harness` command. An interactive setup is available if `--harness` is omitted. Once configured, users can describe work directly within their AI assistant, such as `/aidlc Build a REST API for inventory management`, to initiate structured development tasks.