Overview | Portreeve Docs

Portreeve is an abuse firewall for SaaS signups, trials, checkouts, and logins: one API call per screened action, one verdict back — allow, review, or block — with the reasons behind it. This page is the map; each section below is a short, self-contained guide.

Start here

Guide	What you’ll do
Get set up in five minutes	Account → test key → first verdict via `curl` → see it in the dashboard.
Install the SDK	`npm install portreeve`. Zero runtime dependencies.
First verdict — Next.js	A complete signup route handler.
First verdict — Express	The same screening as reusable middleware.
Device fingerprinting	The browser snippet. Not optional if you take payments.
Test mode	Verify the integration end to end before going live.

Integrate

Topic	Covers
Keys	Secret vs publishable, live vs test, where they live.
What to send per event type	Canonical payloads, the required-fields matrix, signals that go dark without a field.
Failure behavior	Fail-open by default, `degraded` results, safe retries.
Feedback	Reporting confirmed abuse and false positives back to the graph.

Enforce

Topic	Covers
Handling verdicts	The one page to read fully: `review` never blocks, and what that means for your code.
The `review.resolved` webhook	Signature verification, revocation, retries.
Checkout patterns	Capture + refund vs Stripe manual capture, and when to pick which.
Fail-open vs fail-closed	When blocking on our outage is the right trade.

Reference

Topic	Covers
REST API	Endpoints, an example call, and the stored-event payload.
Rate limits	Per-key budgets and what a `429` means.
Idempotency	Safe retries and the one way to get a `409`.
Errors	Every error code the API returns.
Reason codes	Every code a verdict can carry, grouped by signal family.

Support

Email hello@portreeve.com — you’re talking to the founder, and anything wrong or unclear in these docs gets fixed fast.