title: Find the instructions you need
description: Route straight to the namespace that matches what you're trying to do, and permission to skip everything else.
url: https://legalquants.github.io/lq-ai/start/choose-your-path/
checked-against: 671eab0d (2026-10-25)
status: draft
kind: explanation
sources: docs/site/start/choose-your-path.md, README.md

You do not need to read this site in order, and you do not need to read all of it. Find your goal below and go straight there — nothing elsewhere on the site is a prerequisite for it, except a running deployment for anything that says "in a running deployment."

| Your goal | Start here |
|---|---|
| Decide whether LQ.AI is worth trying | [Is LQ.AI right for your team?](/lq-ai/start/is-it-for-you/) |
| Get a deployment running today | [Set up LQ.AI and try your first document](/lq-ai/start/quickstart/) |
| Install on a specific topology (macOS, Docker Compose, Kubernetes, air-gapped, behind a reverse proxy) | [Operate](/lq-ai/operate/) |
| Run it day to day: back up, upgrade, rotate a leaked key, recover from a failure | [Operate](/lq-ai/operate/) |
| Convince a security or procurement reviewer | [Trust](/lq-ai/trust/) |
| Understand or write a skill, or judge whether to trust one | [Skills](/lq-ai/skills/) |
| Package a client-facing, branded deployment | [Deliver](/lq-ai/deliver/) |
| Contribute code, a skill, or a review | [Contribute](/lq-ai/contribute/) |
| Work with a coding assistant on the repository | [Coding agents](/lq-ai/contribute/coding-agents/) |
| Look up a generated spec: configuration keys, skill frontmatter, the ADR index | [Reference](/lq-ai/reference/) |
| Find out what changed in a release | [Changelog](/lq-ai/changelog/) |

This table exists because the site is organized by namespace (`Operate`, `Trust`, `Skills`, and so on), not by reader role, and the theme grouping is deliberate: it's the same flat shape a coding agent reading the machine surface (`/llms.txt`) needs, not a guided tour that assumes you'll read every page in a role-based sequence. A goal that spans two namespaces — installing behind a reverse proxy while also satisfying a security review, say — legitimately sends you to two rows, in either order.

If your goal isn't on this list, the eight top-level namespaces (Start, Operate, Trust, Skills, Deliver, Contribute, Reference, Changelog) are each a hub page with their own onward links — [the entry page](https://github.com/LegalQuants/lq-ai/blob/671eab0d/docs/site/index.mdx) lists all eight.

## Next

- [What gets installed and where data goes](/lq-ai/start/what-it-touches/) — if you haven't installed yet
- [Where to find help](/lq-ai/start/where-help-lives/) — for day-to-day product questions once it's running