About SpecStep

A short read on what we do and why.

What we do

SpecStep generates the documentation your AI coder won't drift from. You answer a short interview — vision, users, requirements, architecture inputs — and we produce a complete documentation package that lays out the decisions, the non-functional requirements, the testing strategy, and the rubric your AI coder will check itself against. Delivered to a GitHub repository, ready to point an agent at on day one.

Why this exists

AI coders are excellent at the next ten lines of code and unreliable at the next ten thousand. They drift. They forget. They invent things that already exist. The fix isn't a smarter prompt — it's giving the agent a written contract about what the system is, how it's organized, and what "done" looks like. Hand-writing that contract is slow and error-prone. SpecStep automates it.

How it works

1. Interview — A short conversation with a SpecStep agent that captures the project's vision, users, scope, requirements, architecture inputs, and constraints. You can pause and resume anytime.
2. Generation — Multiple agents collaborate to draft, self-review, and (depending on the review profile you pick) get a fresh-eyes pass. The output is a complete documentation package, not a single README.
3. Delivery — The package is committed to a GitHub repository you specify. Your AI coder reads it, follows it, and stops drifting.

Three review profiles

• Fast — one self-review round. Best for prototypes and proofs of concept.
• Thorough — three self-review rounds plus a fresh-eyes pass. The default for production work.
• Exhaustive — five self-review rounds plus two fresh-eyes passes. For systems where the cost of getting it wrong is high.

Three surfaces

The same generation engine ships behind three surfaces, all on the same account: the Web app for interactive use, a REST API for CI integrations and scripts, and an MCP server so your AI coder can request a regeneration directly. Same review profiles. Same outputs. Same audit trail.

Where we're going

We're in preview while we tune the rubric, broaden the integrations, and expand the surfaces. Pricing is live; the FAQ covers the most common questions; and we love hearing from people early — contact us anytime.