AfterAgile™
Lens
Get the PDF

The IDE Handbook

The stack, and how the team works in it

Coding Day happens in one shared IDE – an AI-native development environment the whole team works in together, on one screen, at the same time.

The Stack

AI-Native IDE

With a coding agent built in – for example Cursor, or VS Code with Claude Code or GitHub Copilot.

The Repository

Code, context files and decision records, in GitHub or GitLab.

The Pipeline

Continuous integration running every gate on every push.

Preview Links

A live preview of every change, for Definers to test on real devices.

Flag Service

Feature flags and one-click rollback.

Session Recording

Captures prompts, chat and decisions for the team's memory.

How It Works

Real time

One Shared Session

Four in the IDE, the rest watching. CX, UX, Product and Engineering work in the IDE, each in their own lane, and the keyboard rotates between them. The IDE is shared live with remote team members, and anyone else in the session watches every prompt and every change. Stakeholders see the platform only during Intent Review and Final Review.

Every prompt

Invoking AI

Context first, then one task at a time. The agent loads the Intent, the context files and the design system AI project before the first prompt. Each prompt points back to the Intent.

Engineering approves

Branches and Merges

One branch per Intent, small commits. AI writes the pull request summary. Engineering reviews the diff and approves. It merges to main behind the Intent's flag.

On save · on push

Tests

Running all day, not at the end. Tests run in the IDE on every save and in the Pipeline on every push. Persona and device walkthroughs run on the preview link.

Tool names are examples. Use the stack your company has approved for enterprise AI.