How to use this handbook
This is a handbook, not a novel and not a ticket dump. Use it in two modes.
Train
Follow the 30-day path if you are new to SaaS support, new to technical support, or rebuilding fundamentals. The path is ordered on purpose: evidence and tools first, then HTTP, then the browser, then data, then independent triage.
Do not skip the scenario labs. Reading error codes is not the same as choosing the next action under time pressure.
Look up
When you are on a live ticket, search (Ctrl+K / Cmd+K) or jump from the sidebar:
- HTTP status codes — what the number means, who owns it, what to check
- API error shapes — when the body disagrees with the status
- HAR files — capture, sanitize, import, find the failing call
- SQL for support and MongoDB for support — read-only query patterns
Every reference page follows the same skeleton:
- What the customer sees
- What it actually means
- What to check in 2 minutes
- What to ask
- When to escalate
If a page does not tell you the next action, it is not finished. Open a change against it.
Public vs private
The public site is product-agnostic craft. It must stay useful for any SaaS.
Product-specific runbooks (architecture, where this app stores request IDs, known failure modes) live under docs/product/ and are excluded from the public build. Include them locally with:
bash
npm run docs:dev:privateFill real tenant names, table names, and escalation aliases only in docs/product/local/, which is gitignored. The split is recorded in decisions/0002-public-vs-private.md.
If you attach a real product later, copy the templates, do not paste secrets into the public chapters.
How to add a page
- Create a Markdown file under
docs/. - Add it to the sidebar in
docs/.vitepress/config.ts. - Link it from the training path or a lookup page so people can find it.
- Run
npm run qabefore you call it done.
Sources this book trusts
Primary references, not blogs: