> For the complete documentation index, see [llms.txt](https://docs.getkita.org/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.getkita.org/legal/data-processing-agreement.md).

# Data Processing Agreement

29 September 2026, Version 0.9

Between Kita Education Ltd ("Kita") and \[School name] (the "School").

### Background

A. The School uses the Kita coding education platform (the "Service") under Kita's Terms of Use and any order form or pilot agreement between the parties (together, the "Main Agreement").

B. In providing the Service, Kita processes personal data on the School's behalf. This Agreement sets out\
the terms required by Article 28 of the UK GDPR.

C. If this Agreement and the Main Agreement conflict on data protection, this Agreement takes precedence.

### 1. Definitions

1.1 "UK GDPR", "controller", "processor", "personal data", "personal data breach", "data subject" and "processing" have their meanings in the UK GDPR and the Data Protection Act 2018.

1.2 "School Data" means personal data that the School, its staff or its pupils put into the Service, or that Kita creates from their use of it.

1.3 "Sub-processor" means a third party Kita engages to process School Data.

### 2. Roles

2.1 The School is the controller of School Data. Kita is the processor.

2.2 Kita is a separate controller only for:

* (a) staff account and login data needed to run and secure the Service;
* (b) billing and contract records;
* (c) anonymised, aggregated statistics that cannot identify anyone.

Kita's Privacy Policy covers this processing.

### 3. Details of the processing

Schedule 1 sets out the subject matter, duration, nature, purpose, types of personal data and categories of data subject.

### 4. Kita's Obligation

Kita will:

4.1 process School Data only on the School's documented instructions. The Main Agreement, this Agreement, the School's settings in the Service, and written requests from an authorised School contact are all instructions. If Kita believes an instruction breaks data protection law, it will tell the School.

4.2 make sure everyone authorised to process School Data is bound by confidentiality.

4.3 keep the technical and organisational security measures in Schedule 2.

4.4 not use School Data to train artificial intelligence models, and not allow any sub-processor to do so.

4.5 not sell School Data or use it for advertising.

4.6 help the School, taking into account the nature of the processing, to:

* (a) respond to requests from pupils, parents or staff exercising their data protection rights;
* (b) carry out data protection impact assessments and any consultation with the ICO;
* (c) meet its security and breach obligations.

4.7 forward to the School, without undue delay, any rights request Kita receives directly about School Data. Kita will not answer it itself unless the School asks it to.

4.8 make available the information needed to show compliance with this Agreement, and allow audits under clause 9.

### 5. Sub-processors

5.1 The School gives general authorisation for Kita to use the sub-processors listed in Schedule 3, as updated under clause 5.2.

5.2 Kita will give at least 30 days' notice before adding or replacing a sub-processor, by email to the School's contact.

5.3 The School may object on reasonable data protection grounds within that notice period. The parties will discuss the objection in good faith. If it cannot be resolved, the School may end the Main Agreement for the affected service without penalty.

5.4 Kita will put data protection terms in place with each sub-processor that are at least as protective as this Agreement. Kita remains liable to the School for each sub-processor.

### 6. International Transfers

6.1 Kita stores School Data in the United Kingdom and the European Economic Area.

6.2 Kita will only transfer School Data outside the UK and EEA, or allow it to be accessed from outside, where:

* (a) Schedule 3, as updated under clause 5.2, lists the transfer; and
* (b) it is protected by a UK adequacy regulation (including the UK Extension to the EU-US Data Privacy Framework), the UK International Data Transfer Agreement, the UK Addendum to the EU Standard Contractual Clauses, or binding corporate rules approved for UK transfers.

### 7. Personal data breaches

7.1 Kita will notify the School's named contact without undue delay, and in any event within 48 hours of becoming aware of a personal data breach affecting School Data.

7.2 The notice will include, as far as known at the time:

* (a) what happened and when;
* (b) the categories and approximate number of people and records affected;
* (c) the likely consequences;
* (d) what Kita has done or proposes to do;
* (e) a contact for more information.

Kita will send further information as it becomes available.

7.3 Kita will not notify the ICO or affected individuals about a breach of School Data on the School's behalf without the School's agreement, unless the law requires it.

7.4 Kita will keep a record of all personal data breaches affecting School Data.

### 8. Retention, deletion and return

8.1 During the Main Agreement, School Data is kept while the School's account is open, and deleted in line with the School's own actions in the Service and this clause 8. Kita's data guide describes how retention and deletion work in the Service. It is provided for information and does not form part of this Agreement.

8.2 From 1 October 2026, Kita will permanently delete:

* (a) pupils, courses and groups that the School deletes in the Service, 30 days after deletion; and
* (b) conversations with the AI helper, 90 days after they take place.

8.3 At any time, the School may ask Kita in writing to permanently erase the School Data of a named individual. Kita will do so, and confirm in writing, within one month of the request.

8.4 When the Main Agreement ends, or the School asks Kita in writing to close its account:

* (a) the School may download its data for 30 days using the export tools in the Service, or ask Kita for an export;
* (b) Kita will then permanently delete all School Data within 90 days of the end or the request;
* (c) Kita will confirm deletion in writing on request.

8.5 Deleted data will be removed from Kita's backups within 7 days of deletion.

8.6 If a School's account has had no activity for 24 months, Kita may treat the Main Agreement as ended. Kita will first give at least 30 days' written notice to the School's contact, so the School can download its data or keep the account open.

8.7 Kita may keep School Data longer only where the law requires it. It will tell the School if so.

### 9. Audit

9.1 Kita will answer reasonable written questions about its compliance, normally within 30 days.

9.2 If the School reasonably needs more, it may carry out an audit, or have an independent auditor do so. This is limited to once a year unless there has been a breach, with at least 30 days' notice, during business hours, and subject to confidentiality. Each party pays its own costs.

### 10. Liability and Term

10.1 Liability under this Agreement is as set out in the Main Agreement.

10.2 This Agreement lasts as long as Kita processes School Data.

### 11. Governing law

The law of England and Wales governs this Agreement. The courts of England and Wales have exclusive jurisdiction.

### Schedule 1. Details of the processing

**Subject matter.** Providing the Kita coding education platform to the School.

**Duration.** The term of the Main Agreement, plus the deletion period in clause 8.\
**Nature of processing.** Collecting, storing, organising, displaying, sharing between users in the same School, collaborative real-time editing, automated analysis to generate learning feedback, backup, export and deletion.

**Purpose.** To let the School teach coding, set and mark work, give feedback (including AI-generated feedback where the School has it switched on), and track pupil progress.

**Categories of data subject.** Pupils, teachers and other School staff who use the Service.

**Types of personal data: pupils**

* First and last name, or a pseudonym chosen by the School. Username. School email address, if the School provides one. Over or under 13 indicator. Class, group and course membership.
* Code, files, projects, written answers, submissions and their saved versions. Marks, teacher\
  feedback and AI-generated feedback.
* Questions asked to the AI helper and its answers.
* Learning progress and activity records.
* Technical data: IP address, browser and device type, login times.

**Types of personal data: staff.** Name, work email, role, login and security data, content they create, support messages.

**Special category data.** None is requested. Pupils are told not to include personal information in code, answers or AI chat. If any is included, it is handled as School Data under this Agreemen&#x74;**.**

**Automated decision-making.** The Service does not make decisions about pupils that have legal or similarly significant effects. AI feedback is advisory. Teachers make all assessment decisions.

### Schedule 2. Security Messages

**Encryption.** All data is encrypted in transit (TLS 1.2 or higher) and at rest.

**Separation between schools.** Every record is tied to one School. The database enforces this with row-level security, so one School cannot see another's data.

**Access control**

* Multi-factor authentication on all administrative and cloud accounts.
* Kita staff access to production data is limited to named people with a specific staff role. Roles cannot be self-granted.
* Staff administrative actions that change or delete data are logged.

**Application security**

* Code changes are reviewed before release.
* Automated protection against bots on sign-up and login.
* Uploaded images are screened for explicit content on Kita's own servers.
* Application logs hold technical data only, such as IP addresses and internal IDs. Names and pupil work are kept out of logs.

**AI safeguards**

* AI is provided by Microsoft Azure OpenAI Service in the UK (UK South). No other AI provider is used.
* Microsoft does not use prompts or outputs to train models.
* The AI receives only the learning context it needs: the pupil's code, error messages, test results, the lesson and the pupil's question. It does not receive pupils' names or identifiers.
* The AI cannot access the internet or other systems.
* Instructions tell the AI not to ask for personal information, and to direct safeguarding concerns to a trusted adult.
* Daily usage caps limit misuse.
* The AI assistant is clearly labelled as automated.
* Kita follows the DfE's Generative AI Product Safety Standards.

**Resilience.** Automated daily database backups, kept for 7 days. Recovery objective: 24 hours for critical systems.

**Organisation**

* Information Security Policy, reviewed annually, aligned with Cyber Essentials.
* Staff and contractors are bound by confidentiality.
* Personal data breach procedure with a breach log.

### Schedule 3. Sub-processors

Pupil data is stored in the UK, Ireland and Germany. All AI processing takes place in the UK. Kita gives at least 30 days' notice before adding or replacing a sub-processor (clause 5.2).

**Sub-processors that handle pupil data**

<table><thead><tr><th width="121.05889892578125">Sub-processor</th><th width="114.88525390625">Purpose</th><th width="120.4052734375">Data</th><th width="116.108154296875">Stored in</th><th>Access from outside
UK/EEA and
safeguard</th></tr></thead><tbody><tr><td>Heroku (Salesforce,<br>Inc.)</td><td>Runs the Kita<br>application and in-<br>memory cache</td><td>All platform data<br>passes through it</td><td>Ireland (Heroku EU<br>region)</td><td>Limited access for<br>platform operation<br>and support. EU-US<br>Data Privacy<br>Framework and UK<br>Extension; SCCs and<br>binding corporate<br>rules in the<br>Salesforce Data<br>Processing<br>Addendum</td></tr><tr><td>Neon, Inc.</td><td>Main database</td><td>Courses, code,<br>submission, marks,<br>feedback, AI<br>conversation<br>history. No names</td><td>Germany (Frankfurt)</td><td>None</td></tr><tr><td>Supabase, Inc.</td><td>Sign-in and account<br>security</td><td>Names, usernames,<br>email addresses</td><td>Ireland (Dublin)</td><td>Support staff in the<br>US and Singapore.<br>UK Addendum to<br>the EU SCCs</td></tr><tr><td>Amazon Web<br>Services</td><td>File and image<br>storage, course asset<br>delivery</td><td>Files and images<br>uploaded by<br>teachers and pupils</td><td>United Kingdom<br>(London)</td><td>None</td></tr><tr><td>Microsoft Azure<br>OpenAI Service</td><td>AI hints,<br>explanations,<br>feedback and<br>marking</td><td>Code, error<br>messages, test<br>results, questions to<br>the AI helper. No<br>names or identifiers</td><td>United Kingdom (UK<br>South)</td><td>None. Microsoft<br>does not use the<br>data for training or<br>share it with OpenAI</td></tr><tr><td>Resend</td><td>System emails:<br>password resets,<br>invitations, parental<br>consent requests</td><td>Email address and<br>the message</td><td>Ireland (AWS eu-<br>west-1)</td><td>None</td></tr><tr><td>Better Stack</td><td>Application logs</td><td>IP addresses and<br>internal IDs. No<br>names or pupil work</td><td>EU</td><td>None</td></tr><tr><td>Cloudflare, Inc.</td><td>Bot protection on<br>sign-up and sign-in<br>(Turnstile)</td><td>IP address and a<br>browser check token</td><td>Global network</td><td>EU-US Data Privacy Framework and UK Extension / UK<br>Addendum</td></tr></tbody></table>

**Sub-processors that handle staff data only (no pupil data)**

<table><thead><tr><th width="121.05889892578125">Sub-processor</th><th width="114.88525390625">Purpose</th><th width="178.9774169921875">Data</th><th width="161.94775390625">Location and Safeguard</th></tr></thead><tbody><tr><td>Stripe</td><td><p>Card payments for paid</p><p>plans</p></td><td>Billing contact and<br>payment details. Kita never sees full card numbers</td><td>United States. [Data Privacy Framework and<br>UK Extension / UK Addendum]</td></tr><tr><td>Slack</td><td>Kita's internal alerts and<br>support inbox</td><td>Support messages from teachers</td><td>United States. [Data Privacy Framework and<br>UK Extension / UK Addendum]</td></tr><tr><td>Mailchimp (Intuit)</td><td>Product news for adult<br>users who opt in</td><td>Name, email address,<br>whether each email was<br>opened and which links<br>were clicked, IP address<br>and approximate location,<br>device and email client.</td><td>United States.</td></tr><tr><td>Microsoft and Google</td><td>Single sign-on, only if the<br>School enables it</td><td>Name, school email and<br>account ID, from the<br>School's identity provider</td><td>Under the School's own<br>agreement with its identity provider</td></tr></tbody></table>

**Content loaded from other services.** These are not sub-processors, but a pupil's browser contacts them directly, so they receive an IP address: jsDelivr and cdnjs (the in-browser Python runtime and the p5.js graphics library); YouTube and H5P, only when a teacher embeds them in a lesson.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.getkita.org/legal/data-processing-agreement.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
