Engineering With AI
Engineering With AI (EWAI) helps you plan, build and review software with an AI assistant. It gives the assistant a shared record of the project, a delivery workflow and checks against your engineering standards. You keep control of the decisions and approve implementation before it starts.
Website, guides and books: conversationalcoding.dev · Online documentation · Persona library
What's new in this version
Version 0.4.0 lets your team work on separate developer machines with one shared record of intents, delivery progress and approvals. Run a central coordination service, connect each developer's EWAI installation, and keep using your usual AI assistant and Git workflow.
Pick up where the team left off. Developers can see the same work and hand it over between machines. EWAI tracks who owns an active piece of work and checks for newer changes before accepting an update. Permission to develop remains separate from permission to approve.
Keep development on your machine. Your code, project documents, Docker services and tests stay local. Git shares the files; the central service coordinates the work. If your local documents differ from the shared record, EWAI asks you to reconcile them before continuing.
Shared coordination is optional. Existing projects continue to work locally by default. The team setup and recovery guide covers private HTTP/Tailscale connections, access permissions and moving an existing project into shared coordination.
Quick start
EWAI is free and works with Claude Code, Codex, Google Antigravity and Grok Build. You'll need Node.js 22.5 or newer and one of those assistants, installed and signed in.
npm install --global @thebackstoryis/engineering-with-ai
cd /path/to/your/project
ewai
ewai opens your assistant and walks you through setting up the project. See Install and start for details.
To use EWAI in a project without replacing a global installation, run:
npm install --save-dev @thebackstoryis/engineering-with-ai
npx ewai
To install or update EWAI globally, run npm install --global @thebackstoryis/engineering-with-ai.
Install and start
You'll need Node.js 22.5 or newer, npm, and a supported AI command-line tool such as Codex or Claude Code, installed and signed in.
Install EWAI, go to your project folder, then run ewai:
npm install --global @thebackstoryis/engineering-with-ai
cd /path/to/your/project
ewai
Replace the project path with your own. Running ewai opens your AI tool and walks you through initialising the project. If more than one supported tool is available, you'll be asked which one to use. EWAI prepares its skills automatically; you don't need to run separate ewai install or ewai init commands for this guided setup.
It first asks where to keep your project's SPECS records, then helps you describe what you're building. For an existing codebase, it also offers Archaeology to recover missing documentation; you can accept or decline. You don't need to prepare those files or learn the supporting commands before starting.
To return to the project, run ewai in the same folder again. It picks up the saved project state rather than starting setup from scratch. Keep your normal Git workflow for your application's code; EWAI itself is installed and updated through npm.
If npm returns E404, follow installation troubleshooting. That's a package-acquisition failure, not a premium-persona licence error. Don't paste licence keys into npm commands or chat.
Start with a conversation
Once EWAI opens your AI tool, describe what you need. You don't need to memorise EWAI's commands:
“Help us understand this codebase before we change it.”
“We need an export of the filtered support-ticket list. Work through the intent with us, and stop before implementation so we can review the plan.”
“Pick up the export feature and show us where we left it.”
The skills inspect the project's state, choose the relevant workflow and run the supporting commands. They'll ask for missing information and your decisions. ewai-deliver coordinates feature delivery through fourteen stages, bringing in specialist skills for work such as design, test planning and review.
The supporting CLI commands are available for direct control, troubleshooting and automation, but you don't need to work through them manually to use the guided workflow.
What stays in your project?
SPECS means Scope, Purpose, Evidence, Constraints and Strategy. These readable project records hold the purpose, requirements, decisions and evidence the team has agreed. They remain useful outside an AI session.
The dashboard runs locally and lets you inspect work and make supported choices. Its loopback address isn't a shared team website. The AI host does the guided work; the runtime records progress and checks the conditions for moving on. How these parts fit together.
Human judgement remains essential. A persona isn't a real stakeholder, green tests aren't human acceptance, and a delivery handoff isn't permission to deploy.
Licence
The official EWAI npm package is free to use for personal, professional and
commercial work. You can use it to build commercial products, work for clients
and create independent integrations or extensions.
The source is published for transparency and review, but EWAI is not open
source. You may not modify, repackage, redistribute, rebrand or commercially
exploit the EWAI core without separate written permission from Backstory Group.
The licence does not restrict the project content or output you create by using
EWAI. Read the Backstory Group Source-Available Licence for the full
terms.
Choose what you need
Start with the parts your project needs. Portfolio, Team Hub and other advanced dashboard views are optional; hiding a view doesn't disable mandatory project checks.
Personas
The included personas support the normal workflow. You can also create project-specific personas and use your own personal library.
Premium personas are optional specialist perspectives. If you have a subscription, enter your key privately and install the pack before the analysis you want it to support. Installing a persona doesn't give it authority to approve a requirement, bypass a check or speak for a real user.
Go deeper
Contributing to EWAI
If you're changing the harness itself, use the contributor guide and verification walkthroughs. Those source-checkout and regression-test instructions aren't part of setting up your own application.
