Explain My BuildHow Explain My Build works
Last updated: August 30, 2026
Explain My Build reads your app's actual source code — via a public GitHub URL (read then deleted), or a private repo or local folder that runs in browser memory or via the CLI. It produces a free runbook and fix-list, plus an optional end-user guide at proposed commercial features (not currently for sale). supported claims include evidence references and uncertain conclusions are marked.
How does Explain My Build read my code?
You give it your code in one of two ways, depending on how private your project is:
- Public GitHub URL. Paste the link. We read the repo on our server to generate your artifacts, then delete it — the code is subject to terminal cleanup under the designed flow.
- Private repo or local folder. Run it in browser memory, or via the CLI. A configured remote model may receive derived evidence.
Once it has the code, it maps your app: the routes and screens, the data models, the auth flow, the environment variables, and the third-party services it connects to. This works across apps built by Lovable, Bolt, v0, Replit, Cursor and more, because they all produce readable code.
What do I get, and what does it cost?
From one code read you get three artifacts, on an available Free workspace plus disabled commercial proposals:
| Artifact | What it's for | Price |
|---|---|---|
| Founder runbook | Operate and extend your app | Free |
| Code-verified fix-list | Ranked issues to hand to your AI | Free |
| End-user guide + sharing | Help for the people who use your app | proposed commercial features (not currently for sale) |
The end-user guide add-on includes a share link and access code, an in-app Help widget, and a portable export. It's proposed and not currently for sale. See pricing for the exact breakdown.
What does evidence-linked mean?
Supported claims retain a reference to the static code, build history, authorised runtime observation or audit result that informed them. The workspace also records confidence and coverage, and marks unsupported conclusions as unknown.
An evidence reference makes review faster; it does not prove that the evidence is complete or that a model interpretation is correct. Owners must review a guide before publication and use qualified human review for high-stakes decisions.
Where does my code go?
The answer depends on the selected path. Public repositories use isolated temporary server jobs with terminal cleanup. Private GitHub uses the documented transient relay. Local folders and ZIP files are read in browser memory, while CLI analysis runs locally.
A configured remote model may receive derived evidence on any path unless a local model is selected. Review the complete path-by-path privacy disclosure before analysis.
Frequently asked questions
How does Explain My Build turn my code into a user guide?
It combines detected source structure with available build history, authorised runtime observations and audit results. Supported claims retain evidence references; gaps and unsupported conclusions are marked for owner review.
Is my code stored or uploaded?
It depends on the selected path. Public repositories use isolated temporary server jobs with terminal cleanup; private GitHub uses the documented transient relay; local folders and ZIP files are read in browser memory; CLI analysis runs locally. A configured remote model may receive derived evidence.
What is free and what costs money?
The founder runbook and the code-verified fix-list are free. The end-user guide plus sharing (share link and access code, in-app Help widget, portable export) is proposed and not currently for sale.
Does it run my app or just read the code?
Static analysis reads source. An optional authorised Walker run can observe selected journeys, while Auger reports only the checks it completed. This is not an unrestricted penetration test and does not replace qualified security review.
Turn your app into a guide your users can follow.
Start free with no credit card. Runtime varies with repository size, evidence sources and queue state.