Privacy Policy

Who is responsible

The controller of the personal data described here is TO BE COMPLETED — registered company name, TO BE COMPLETED — registered address, including country, registered as TO BE COMPLETED — company number and register.

Questions about this policy, or any request about your own data, go to TO BE COMPLETED — privacy contact email. We answer within one month, as the GDPR requires.

We have not appointed a Data Protection Officer. We do not believe the conditions in Article 37 GDPR apply to us; if that changes we will appoint one and say so here.

What we hold

Four kinds of thing, and nothing else:

• Account data — your name, work email, password hash, workspace name, and the role and per-agent permissions you have been given.
• Workspace content — agent instructions, the questions you ask, the answers produced, documents you upload, the sources cited, and the record of each run including its cost, timing and hashes.
• Billing data — the contact details on a top-up request and the amount. Card details never reach us; see 'Credit and payment' in the Terms of Service.
• Technical logs — request metadata and error traces, used to keep the service up and to investigate abuse.

Why we are allowed to hold it

Every category above rests on one of these legal bases:

• Performance of a contract (Art. 6(1)(b)) — account data and workspace content. Without them the product cannot do the thing you asked it to do.
• Legitimate interests (Art. 6(1)(f)) — technical logs, for security, fraud prevention and keeping the service running. We have weighed this against your interests and consider it proportionate; you may object at any time.
• Legal obligation (Art. 6(1)(c)) — billing records we are required to keep for tax purposes.

We do not train on your content

We do not train models. We do not have models to train. Your prompts, documents and answers are not used to improve this or any other service.

Your content is sent to the model provider named on the agent version you chose, for the sole purpose of producing your answer. Which provider that is, is your decision, recorded on every single run.

Who else sees it

Only the sub-processors listed in our Data Processing Agreement, each for the narrow purpose stated there. We do not sell personal data, and we do not share it for advertising. There is no advertising on this product.

We will disclose data if a valid legal order compels us to. Where we are lawfully permitted to tell you first, we will.

Transfers outside the EEA

The model providers we rely on — Anthropic, OpenAI and Google — process data in the United States. When your prompt runs against one of their models, it is transferred there.

These transfers are made under the European Commission's Standard Contractual Clauses together with supplementary measures. You can ask us for a copy of the clauses in force.

If you need model calls to stay within a particular jurisdiction, say so before you sign. It is a deployment question with a real answer, not something to discover afterwards.

How long we keep it

Run content is retained for the period stated on your plan — 30 days on the free tier, one year on Starter, three years on Team, and a negotiated period on Enterprise.

When retention expires, the content of a run is redacted while its hashes, configuration and timing are kept. That is deliberate: it lets a version diff still prove whether behaviour changed, long after the text itself is gone. The retained remainder is not personal data in any ordinary sense.

Account data is kept while your workspace exists. If you close it, we delete account data and redact workspace content within 90 days, except where we must keep billing records for tax.

Your rights

Under the GDPR you may ask us to do any of the following, free of charge:

• Access — get a copy of the personal data we hold about you.
• Rectification — correct anything inaccurate.
• Erasure — delete it, where no legal obligation requires us to keep it.
• Restriction — pause our processing while a dispute is resolved.
• Portability — receive your data in a structured, machine-readable format, or have it sent to another provider.
• Objection — object to processing we base on legitimate interests.
• Withdraw consent — where processing rests on consent, withdraw it at any time, without affecting what was lawful beforehand.

Complaining

If you think we have handled your data badly, please tell us first at TO BE COMPLETED — privacy contact email — it is usually faster. You are not obliged to.

You have the right to lodge a complaint with a supervisory authority, in the member state where you live, where you work, or where the problem happened. Ours is TO BE COMPLETED — lead supervisory authority and its website.

Security

Passwords are stored as bcrypt hashes and never in plain text. Sessions use a signed token in an HttpOnly cookie. Provider API keys, where an Enterprise workspace supplies its own, are stored in that workspace's database and shown only as a masked fingerprint after saving.

Access to an agent is granted per person and per agent. Someone with no grant on an agent cannot see it, its threads, or its run records — a request for one returns the same response as a request for something that does not exist.

If you find a vulnerability, please report it to TO BE COMPLETED — security contact email before disclosing it publicly. We will not pursue you for good-faith research.

Changes

If we change this policy in a way that materially affects you, we will tell you before it takes effect, not after. The date at the top always reflects the current version.