Skip to main content

Privacy policy

This notice explains what Chronik.io does with personal data, who receives it, and what you can ask us to do about it. It describes the service as it actually runs; where something is not yet decided, it says so rather than describing an intention.

Who is responsible

PUSHERY, ℅ HYPERLINK Solutions GmbH, Ringbahnstr. 42, 12099 Berlin, Germany. Contact: [email protected], +49 (0) 30 / 398 216 398. Full details are in the imprint.

We have not appointed a data protection officer; we are below the threshold in § 38 BDSG. Requests under Articles 15 to 22 GDPR reach a person at the address above.

What Chronik.io is, and why that shapes everything below

Chronik.io watches releases in the GitHub repositories you connect and turns them into messages in the places you choose. Two consequences run through this whole notice:

You decide who receives the announcements. Connecting a Slack workspace, naming a webhook endpoint or making a changelog page public are your choices, and each one sends release content — including the name of whoever published the release — somewhere we do not control. We deliver where you point us; we cannot judge whether that destination was the right one.

Release data is other people's data too. A release carries the GitHub identity of the person who published it. That person is usually not you, and usually has no account here.

Data we hold, and why

Your account. Name, email address, and — if you use one — a password hash, two-factor secret and recovery codes. Language, time zone and session length are settings you chose. We need these to give you an account and to keep it yours. Legal basis: Article 6(1)(b) GDPR (performing the contract) and Article 6(1)(f) (securing the account).

Signing in. You sign in to Chronik.io with your GitHub account, or with your email address and a password. Two-factor codes sent by email are stored only as a hash and expire. Devices you have signed in from are remembered as a one-way fingerprint, never as a stored IP address or browser string — enough to notice a new device, not enough to reconstruct where you were.

What you connected. The GitHub installation, the account it belongs to, its avatar, and the repositories it covers, each by its GitHub identifier, name and owner. Legal basis: Article 6(1)(b) GDPR.

Releases and the events behind them. The webhook payload GitHub sends us, kept as received, and the release we derive from it. The payload names the release author. Legal basis: Article 6(1)(f) GDPR — our and your legitimate interest in a reliable, replayable record of what was announced; a delivery that cannot be replayed cannot be corrected.

Changes to your account. Name and email changes are recorded with a timestamp so an unexpected change can be traced. Legal basis: Article 6(1)(f) GDPR.

Consent records. When you accept the terms or acknowledge this notice, we store the fact, the document version and the time, in a tamper-evident ledger. That record exists so we can demonstrate what you were shown, which Article 7(1) GDPR requires of us.

Cookies

Chronik.io sets no tracking cookies and shows no cookie banner, and those two facts belong together. The only cookies are the ones that keep you signed in and protect forms against cross-site request forgery. They are strictly necessary under § 25 (2) TDDDG, which is why no consent is asked for them.

Analytics

We measure page views with Matomo, on our own server, from the request that already happened. No script runs in your browser, nothing is stored on your device, and no third party is involved. This is the reason the previous paragraph can be as short as it is.

Who receives data

The table below is generated from a single machine-readable register in this application, so that it and our record of processing activities cannot drift apart.

Recipient What for Where Basis for transfer
GitHub, Inc. The platform Chronik.io reads releases from USA see below
Sentry (Functional Software, Inc.) Error reports Germany — Sentry's EU region none needed
Flare (Spatie BV) Error reports Belgium none needed
Digital­Ocean, LLC Database backups Germany — Frankfurt region none needed
Nightwatch (Laravel Holdings Inc.) Performance monitoring USA see below
Cloudflare, Inc. Delivery and denial-of-service protection USA see below
Matomo Usage analytics Germany — our own server none needed; no third party

Destinations you chose are not in this table, because we did not choose them. A Slack workspace, a Telegram chat or a webhook endpoint receives what you told Chronik.io to send there, and the operator of that destination decides what happens to it.

Transfers to the United States rest on the European Commission's adequacy decision for the EU–US Data Privacy Framework where the recipient is certified under it, and otherwise on standard contractual clauses. We are confirming the exact basis for each US recipient and will name it here rather than leave a general sentence standing in for it.

Email delivery has no provider yet. Chronik.io currently writes mail to a local log instead of sending it. When a provider is chosen it becomes a recipient of your address and will be named here before the first message goes out.

When you let Claude summarize a release note

Chronik.io can hand a release note to a language model and send the summary instead of the original. This section stands on its own because the feature is the exception in every respect: it is off by default, it runs on your account with Anthropic, and it is the only place where Chronik.io hands content to a recipient we do not operate ourselves.

None of it happens until you switch it on. Chronik.io holds no Anthropic key of its own; the key sits on your account, encrypted, and without it nothing is asked. That is also why Anthropic is not in the recipients table above — that table names who receives data because of how we run the service. Anthropic receives data because you switch a feature on.

What is sent when you do: the release notes of the release being announced, and the instruction your template carries for it. Nothing else — no account details, no addresses, nothing about other repositories. Release notes can contain the name of the person who published the release; where they do, that name travels with them.

Whose contract this is. The key is yours, so the contractual relationship with Anthropic is yours, and the calls are on your bill. ⚠️ Whether that takes us out of the role of processor is something we do not assert here — we open the connection and we write the instruction, and that is more than a pass-through. The question is not settled, and rather than claim a legal consequence we describe what actually happens.

Third-country transfer. Anthropic is in the United States. Which basis that transfer rests on is a lookup we are completing, and it will be named here once it is settled — as with every other US recipient in the table above. Our own legal basis for the processing is Article 6(1)(b) GDPR: you asked for the feature.

The output is marked. A summary a model wrote carries a line saying so, in the language of the summary — and we record per delivery which model produced it, under which preset, and when. That is our obligation as the provider of the output, and it cannot be switched off. What you owe as the publisher beyond that is in the help pages under When Claude writes your release notes.

What Anthropic does with the text is governed by Anthropic's own terms rather than ours. We do not pick the model for you and we train nothing.

How long we keep things

Your account and what you connected: for as long as you have an account. Release events and releases: for as long as the repository is connected, so that a changelog stays complete. Two-factor codes: until they are used or expire. Consent records: for as long as we may need to demonstrate them, which outlives the account itself — that is their purpose.

Backups: every night a copy of the database is stored with DigitalOcean in Frankfurt. Every copy is kept for 7 days, then one a day for 16 days, one a week for 8 weeks, one a month for 4 months and one a year for 2 years. Older copies are deleted.

Delete your account and the account data goes with it. Copies of it that are already in a backup are not edited; they are deleted on the schedule above. What was already published to a destination you chose is beyond our reach; you would have to remove it there.

Your rights

You can ask us for a copy of your data (Article 15), to correct it (Article 16), to erase it (Article 17), to restrict its use (Article 18), to receive it in a portable form (Article 20), and to object to processing we base on legitimate interests (Article 21). Chronik.io can produce your data as a file from within the application.

You can also complain to a supervisory authority. Ours is the Berliner Beauftragte für Datenschutz und Informationsfreiheit, Alt-Moabit 59-61, 10555 Berlin.

Changes to this notice

We publish a version number with every change. A change that affects you materially is announced rather than quietly deployed.