AIDEN Docs
AIDEN is an AI-native software engineering workspace — everything you need to build software with AI, in one place.
It is built for people who want the speed of AI without giving up the structure of real software development. Instead of reducing software work to a long stream of prompts, AIDEN gives you a system:
- projects
- stories and specs
- a board
- agents
- a focused story workspace
- Git-aware execution
- and now voice — talk to AIDEN and delegate work out loud
What AIDEN is for
AIDEN is for builders who want to:
- turn vague work into clear stories
- plan before implementing
- execute with AI inside a real workspace
- run multiple agents without losing control
- keep specs, context, and Git aligned
- ship software in a way that still feels disciplined
Start here
Installation
Download AIDEN, connect an AI provider, and get set up in a few minutes.
Getting started
Create your first project, plan your first story, and complete your first loop.
Core concepts
The mental model behind projects, stories, specs, agents, and the board.
Features
The main parts of the product and when to use each one — including Talk to AIDEN.
Recommended workflow
The best way to use AIDEN from idea to shipped work.
The shortest possible explanation
AIDEN works best when you follow this loop:
- Capture work inside a project.
- Shape it into stories.
- Plan the story into a spec.
- Execute in the story workspace.
- Review against the spec.
- Ship and update the board.
That is where the product becomes more than a chat UI.
What makes AIDEN feel different
It starts with structure
AIDEN wants work to live in projects, stories, specs, and statuses.
It keeps execution close to context
The spec, the chat, the terminal, the browser, the notes, and the Git-aware workflow all live in one place.
It treats AI like part of a real engineering system
Agents are useful because they work inside a workflow, not outside of one.
It runs on your machine
AIDEN is local-first: your code, projects, and keys stay on your machine, and agents run through the CLIs and providers you already use.
It stays close to shipping
The point is not to generate impressive output. The point is to finish real software work cleanly.