The biggest barrier to adopting any developer tool is the signup flow. Create an account, verify your email, set up a project, generate an API key, copy it into your code, and then you can see if the product actually works. By the time you reach that first API call, you have already invested 10 minutes and a non-trivial amount of patience.

Today we are shipping Stream Mode — a fundamentally different approach to developer onboarding. Install the SDK, call one function, and start sending audit log events immediately. No account. No API key. No dashboard. Just data flowing in under 60 seconds.

How it works

Stream Mode is built on a simple idea: let developers prove value before asking them to commit. Here is the entire integration:

import { LogStitch } from '@logstitch/sdk';

const stream = LogStitch.stream();
// Console: "LogStitch Stream Mode — Claim at: https://logstitch.io/claim?token=<uuid>"

stream.log({
  action: 'user.signed_up',
  category: 'auth',
  actor: { id: 'user_123', type: 'user', name: 'Alice' },
  tenant_id: 'acme_corp',
});

That is it. The SDK generates a random claim token (a UUIDv4), and events are sent to an unauthenticated endpoint. No API key needed. The claim token is printed to the console so you can find it later.

What happens to the data

Stream events are stored in a separate, temporary table — isolated from authenticated project data. They receive the same built-in PII redaction that all LogStitch events get: credit card numbers, SSNs, email addresses, phone numbers, JWTs, bearer tokens, and AWS keys are automatically masked before storage.

No event data is readable through the stream endpoints. Only metadata — event count, status, and expiry — is exposed. Your data stays private until you claim it.

Claiming your stream

When you are ready to sign up, claiming is a one-click operation. Visit the claim URL printed to your console (or paste the token into the dashboard), select a project, and hit claim. All events are atomically migrated into your project — no data loss, no duplicates.

You can also claim programmatically via the API with a master key:

curl -X POST https://logstitch.io/api/v1/streams/claim \
  -H "Authorization: Bearer mk_..." \
  -H "Content-Type: application/json" \
  -d '{ "claim_token": "a1b2c3d4-e5f6-4a7b-8c9d-0e1f2a3b4c5d" }'

Claiming is irreversible and atomic. The token is permanently locked after use, and events are moved (not copied) into the target project.

Built-in guardrails

Stream Mode is designed for evaluation, not production workloads. We enforce sensible limits to prevent abuse of the unauthenticated ingestion path:

These limits are generous enough for a real evaluation but tight enough to prevent the endpoint from being used as a free data store.

Why this matters

Developer tools live and die by their onboarding experience. The tools that win are the ones that let developers feel value before asking for commitment. Stripe did this with test mode. Algolia does it with a generous free tier. We are doing it by removing the signup barrier entirely.

With Stream Mode, the conversation with a potential customer changes. Instead of “sign up and try it,” it becomes “try it and sign up when you are ready.” The SDK is the demo. The claim flow is the conversion.

Get started

Stream Mode is available now in @logstitch/sdk@0.2.0. Install the package, call LogStitch.stream(), and start sending events. When you are ready to keep your data, sign up and claim your stream from the dashboard.

npm install @logstitch/sdk

Read the full documentation in the SDK reference and the Streams API reference.