vivibe makes a desktop app that maps how your code runs, and an API that powers its optional AI features. This Privacy Policy explains how we collect, use, disclose and protect personal data when you use the vivibe app, the vivibe API and vivibe.net (together, the “Services”).
vivibe is run by an individual, Hà Gia Kính, in Vietnam, who decides how and why the personal data described here is processed (the controller, “we”). Contact: privacy@vivibe.net.
One policy applies wherever you live. Where a local law gives you more rights, you have those rights too; see Regional disclosures. A Vietnamese version is here; both say the same.
1. Personal data we collect
Data you give us
- Account data. If you sign in, your name, email address, profile picture link and sign-in provider, received from Auth0 when you sign in. During the private beta you need an approved account to use the app.
- Beta requests. If you ask to join the beta: your email address and, if you add them, your name and what you would use vivibe for, plus the version of this policy you agreed to. We also receive your IP address, browser user agent and language, used to prevent spam. We compare the email with the one your sign-in provider verified to let you in.
- Support requests. If you use the contact form or email us: your email address, the type of request, what you write and, if you add it, your app version. With a form request we also receive the version of this policy you agreed to, your IP address and browser user agent, used to prevent spam.
- Error reports. Only when you click Report in the app and then Send: the error message, stack trace and component stack, the screen you were on, your comment if you add one, your email address if you are not signed in (your account if you are), the last part of the app log, and a screenshot of the app window if you tick the box. The report also carries the random install ID, app version, operating system, CPU type and language, and we receive your IP address and user agent. The app log and screenshot may show file names, paths or code that was on screen, so check before you send.
Data we receive when you use the Services
- Sign-in records. Time, IP address, app version, operating system and version, CPU type, language and a random install ID.
- AI requests. Only while an AI action you started is running: the code excerpts, diagram structure, screen text or coding-agent steps, and the question that action needs, plus the language for the answer. To answer, our AI assistant may also ask the app to search the repository you have open (by text or file name) and read whole files from it. The app refuses files that look like secrets (such as
.envfiles, SSH keys and.pemor.keyfiles) and masks strings that look like passwords, tokens or keys before anything is sent. - Coding-agent history. The app reads Claude Code and Codex transcripts on your computer. When you use an AI feature on that history (for example asking about a session, or a summary of it), the transcript steps it needs are sent as part of that AI request.
- GitHub connection. Only if you connect GitHub: your GitHub username and ID, the installations you grant, and an access token that we store encrypted.
- Update checks. App version, platform and the random install ID.
- Usage statistics (app). Only if you allow it when the app asks (change it any time in Settings → Share usage statistics): which screens and features you use, how often and for how long, whether an analysis worked, plus app version, operating system, language, the random install ID and, when signed in, your account. Never your code, file or folder names, paths, repository names, prompts or anything you type.
- Usage statistics (website). Our own counter records which page was viewed, the site you came from, campaign tags and country; your IP address is not stored, only a hash that changes every day, and downloads of the app are counted the same way. The website also uses Cloudflare Web Analytics, which counts page views and page load times without cookies and without identifying you. Neither sets cookies.
- Anti-spam check. The beta and support forms use Cloudflare Turnstile, which receives your IP address and browser signals to tell people from bots.
- Server logs. Requests to the API, used to run and secure it. They do not contain your code.
Data that stays on your computer
vivibe analyses your repositories on your computer. Apart from what an AI action or error report you start sends (above), we do not receive your source code, the map, analysis results, terminal sessions, the coding-agent transcripts the app reads, run environments or test data. On macOS your sign-in session is kept in the Keychain.
What we do not collect
We do not use advertising or cross-site tracking tools, in the app or on this website, and we do not set cookies for tracking or advertising. The app sends error reports and usage statistics only to the vivibe API, never to a third-party analytics or crash-reporting service. We never sell personal data or share it for advertising.
2. How we use personal data
- To provide the Services you ask for: sign-in, AI actions, GitHub access and updates.
- To run the private beta: decide who gets in and check that a signed-in account is approved.
- To find and fix the errors you report, and to reply to you about them.
- To keep the Services secure, prevent spam and abuse and fix problems, using sign-in records, server logs and the anti-spam check.
- To roll out updates in stages, using the install ID.
- To understand which features and pages are used and improve them, and to report aggregate numbers (for example active users or downloads) to partners and investors. Only totals leave vivibe, never data about one person.
- To answer your support and privacy requests, and to email you about your beta request or account.
- To keep a record of the consent you give and of the actions taken on your data, meet legal obligations and enforce our Terms.
We do not sell personal data, share it for cross-context behavioural advertising, or use it to profile you.
3. AI features and model training
AI actions run only when you ask. The request goes to the vivibe API and then to our model provider, OpenAI, to produce the answer. While it works, the assistant may ask the app for more of the open repository: search results and the content of files it needs, with secret-looking files refused and credential-like strings masked as described above. We do not keep the content of AI requests, the files read or the answers after the answer is sent. Answers may be cached on your computer.
We do not use your code, transcripts, questions or answers to train models. OpenAI does not train on data sent through its API and may keep it for up to 30 days to detect abuse.
AI output can be wrong. It cites the code so you can check it before you rely on it.
4. How we disclose personal data
We disclose personal data only to the service providers below, who process it on our instructions under data processing terms:
| Provider | Purpose | Location |
|---|---|---|
| Amazon Web Services | API hosting, database, logs, website and downloads | Singapore; global edge |
| Cloudflare | DNS, network and secure tunnel for all website and API traffic, rate limiting, anti-spam check (Turnstile), website analytics | United States; global network |
| OpenAI | AI model for AI actions | United States |
| Auth0 (Okta) | Sign-in | United States |
| GitHub | Repository access you grant | United States |
| Google Workspace | Email (hello@ and privacy@vivibe.net) | United States; global |
We may also disclose data when the law requires it, to protect people or the Services, or as part of a merger or acquisition, in which case this policy continues to apply.
5. Retention
| Data | How long |
|---|---|
| AI request content and files read during it | Not stored after the answer |
| Server logs | 14 days |
| Database backups | 7 days |
| Error reports | 12 months, then deleted automatically |
| Sign-in history | 12 months, then deleted automatically |
| Support requests | 24 months, then deleted automatically |
| Beta requests | 24 months, then deleted automatically |
| Records of emails we send you | 24 months, then deleted automatically |
| Usage statistics (app and website) | 24 months, then deleted automatically |
| Administrator audit log (who did what to an account or request) | 5 years, then deleted automatically |
| Account, credits and GitHub link | Until you delete your account |
| Data on your computer | Until you delete it or uninstall the app |
When you delete your account (Settings → Account → Delete account), we delete at once your account, sign-in history, credits ledger, GitHub connection (and revoke its token), your error reports, and the beta and support requests filed under your account’s email. Your Auth0 sign-in account is deleted automatically at the same time, so no sign-in profile is left with our sign-in provider. Copies in database backups are gone within 7 days. Entries about your account in the administrator audit log are kept until their 5-year limit, as the record of what was done with your data, and are then deleted automatically. You can also ask by email: write to privacy@vivibe.net.
6. Your rights and choices
Depending on where you live, you can ask to be informed, to access, correct, delete or export your personal data, to restrict or object to its processing, and to withdraw consent. Withdrawing consent does not affect processing done before. You can export and delete your data yourself in the app (below); for anything else, or if you cannot use the app, email privacy@vivibe.net or use the contact form with the type Privacy. We answer within the time the applicable law sets. You can also complain to your data protection authority.
Controls in the app
- Everything except AI runs on your computer; nothing from your repositories is sent unless you start an AI action or send an error report.
- Error reports are sent only when you click Send; the screenshot is included only if you tick the box.
- Export my data in Settings → Account downloads a JSON file with the personal data we hold for your account. If you ask by email instead, we send you the same file.
- Sign out, disconnect GitHub, and Delete account in Settings → Account. Deleting your account also deletes your Auth0 sign-in account.
- Reopen this notice in Settings → Account → Privacy.
- Uninstall the GitHub App from your GitHub settings at any time.
7. International transfers
Your data is processed outside Vietnam and possibly outside your country, including in Singapore and the United States and on Cloudflare’s global network, by the providers listed above. We rely on safeguards such as the providers’ data processing terms and the European Commission’s Standard Contractual Clauses, and we prepare and file the cross-border transfer impact assessment that Vietnamese law requires.
8. Security
Traffic is encrypted with TLS, tokens are encrypted at rest, and access to production systems is limited. See Security for details. If a breach affects you, we will tell you and the authorities as the law requires.
9. Children
vivibe is a developer tool and is not meant for children under 16. We do not knowingly collect their data.
10. Regional disclosures
Vietnam
We process personal data under the Law on Personal Data Protection (Law No. 91/2025/QH15) and Decree No. 356/2025/NĐ-CP detailing it. We tell you, through this policy and before you use a form or a feature, what we process, why and for how long; we ask your consent where it is needed (for example the tick box on the beta and support forms, usage statistics and error reports) and keep a record of it; we prepare and file the personal data processing impact assessment and cross-border transfer impact assessment with the Ministry of Public Security; and we handle your requests and any breach within the times the law sets. Personal data in error reports, sign-in records and support requests is basic personal data; we do not ask for sensitive personal data.
EU, EEA, UK and Switzerland
The controller of the personal data described here is named in the introduction. Our legal bases:
| Purpose | Legal basis |
|---|---|
| Sign-in, AI actions, GitHub access, updates | Performance of a contract |
| Security, abuse and spam prevention, staged updates | Legitimate interests |
| Beta and support requests | Consent (the tick box), and steps you ask for before a contract |
| Error reports | Consent: you choose to send each one |
| Usage statistics in the app | Consent, which you can withdraw in Settings |
| Website visit and download counts (no cookies) | Legitimate interests |
| Legal obligations, audit log | Legal obligation |
United States
In the last 12 months we collected identifiers (name, email, IP address, install ID), internet activity (sign-in records, server logs, usage statistics, error reports) and, if you connected GitHub, account credentials (an encrypted token). We used them for the purposes above and disclosed them only to the service providers listed. We do not sell or share personal information, and we do not use sensitive personal information to infer characteristics. You will not be treated differently for using your rights.
Other countries
We also honour the rights given by laws such as Brazil’s LGPD, Singapore’s PDPA, Japan’s APPI, Korea’s PIPA, Canada’s PIPEDA, Australia’s Privacy Act and India’s DPDP Act.
11. Changes to this policy
We date every version; this one takes effect on November 16, 2026. If a change matters, the app shows the notice again before you continue, the forms ask you to agree to the new version, and we update the date above.
12. Contact us
Questions or requests about privacy: privacy@vivibe.net.