Privacy Policy
Last updated: October 13, 2026
This policy explains what data doc0 collects, why, who processes it, how long we keep it, and the rights you have over it. It covers the doc0 app, the documentation sites we host for you, and the integrations you connect. Questions? Write to alessandro@doc0.dev.
1. Information We Collect
- Account data. When you sign in with GitHub we receive your GitHub username, email address, and avatar. We use them to authenticate you and show your profile in the app.
- Repository data. When you connect a repository we clone it temporarily to generate your documentation and knowledge graph. What we persist is described in §2.
- AI chat. Questions you ask doc0 (in the app, or through a connected Slack or Linear workspace on the Team plan) and the answers it gives are stored as conversation history.
- Billing data. When you subscribe, our payment provider processes your payment details. We never see or store your card number.
- Product usage. Page views, feature interactions, and error reports inside the doc0 app, collected from your first visit unless you reject analytics (§6). While you use doc0 signed in, our servers also record what your account does with it (for example that a wiki was generated or a subscription started), linked to your account. For AI features we record usage figures (model, tokens, cost, response time), never your prompts or the answers.
- Documentation-site visits. When someone reads a wiki we host, we count the visit pseudonymously so project owners can see traffic (§7). We do not use cookies for this and we do not store raw IP addresses with these counts.
- Waitlist. If you join the waitlist we store the email address you enter, only to tell you when you can sign in. Our team sees each new address in its own Slack workspace.
- Support and email. If you contact us or receive transactional email (account, billing, invitations), we process the messages involved.
We do not sell your data, and we do not use your data to train AI models.
2. Source Code & Repository Data
Your repository is cloned temporarily and analyzed in memory. We do not keep a copy of your repository. We persist: the generated documentation, the code knowledge graph (symbol names and their relationships), file paths and content hashes (so we can detect what changed on your next push), and - to write and update your documentation - indexed chunks of your source code together with their vector embeddings. Search and chat run on the documentation and the code knowledge graph, not on these chunks. Chunks are encrypted at rest, never used for training, and deleted when you delete the project. The cloned copy is deleted as soon as processing finishes.
During generation, documentation content is sent to the AI processors listed in §5 for inference only - never for training.
3. How and Why We Use Data
- Authenticate you and manage your account and team
- Generate, host, and update documentation for your repositories
- Answer questions through AI chat, MCP, Slack, and Linear
- Process billing and manage your subscription
- Show project owners how their documentation is used
- Improve the product through usage analytics
- Send transactional email about your account
- Tell people on the waitlist when they can sign in
- Detect and investigate abuse or security incidents
Under the GDPR, our legal bases are: contract for providing the service you signed up for (accounts, generation, chat, billing); legitimate interests for product analytics (you can object at any time by rejecting analytics), service security, abuse prevention, and documentation-site visit counts; consent for session replay and for the waitlist, which you can withdraw at any time; and legal obligation for records we must keep, such as billing.
4. Data Storage & Security
Your data is stored with encryption at rest and in transit. Secrets such as GitHub installation tokens, integration credentials, and MCP keys are stored with an additional layer of envelope encryption. Webhooks are verified with signed signatures, access is controlled per account with row-level security, and security-sensitive actions are recorded in an audit log. All traffic to doc0 is served over HTTPS. No system is perfectly secure, but we design so that a single failure does not expose your data.
5. Sub-processors
We rely on the following processors to operate doc0. Each processes only the data necessary for its role.
- Supabase (US/EU) - authentication, PostgreSQL database, secrets storage
- Vercel (US) - hosting, edge network, deployment and traffic analytics, and AI Gateway, which routes AI requests to Google and Voyage AI (listed below); generation, chat and embeddings run with zero data retention, and neither provider uses the data to train models
- Trigger.dev (US) - background job processing (wiki generation, incremental updates, scheduled retention jobs)
- Google (Vertex AI, through Vercel AI Gateway; processing may take place outside the EU) - AI documentation generation, chat, and text embeddings
- Voyage AI (US, through Vercel AI Gateway) - search-result reranking (short text snippets only)
- PostHog (EU region) - product analytics and session replay (replay starts only after you accept analytics, and masks all text and inputs)
- Sentry (US) - error tracking and performance monitoring
- Resend (US) - transactional email delivery (account, billing, invitations)
- Polar (US) - subscription billing and payment processing
- Calendar0 - the booking calendar behind every Book a call button, shown inside our PR reviews page; it receives the details you enter when you book a call
- Upstash (Global) - Redis-backed rate limiting and ephemeral caches (no PII stored)
- GitHub (US) - OAuth authentication and the GitHub App used to clone repositories you connect
- Slack / Linear (US) - only if you connect them on the Team plan; they deliver the questions you ask doc0 there and receive its answers
- Slack (US) - our team's own workspace, where product feedback you send from the app and the email address of each new waitlist sign-up reach us
6. Cookies & Tracking
We use strictly necessary cookies for authentication (your session) and to remember your documentation-mode preference. These are required for the service to function and do not require consent.
We use analytics cookies (PostHog, Vercel Analytics) from your first visit to understand how doc0 is used. If you reject analytics in the cookie banner, analytics stop in your browser and stay off on your next visits. Session replay starts only after you explicitly accept. You can change your choice at any time - .
Rejecting analytics stops tracking in your browser. The account events our servers record while you use doc0 signed in (§1) are part of running the service and continue. Documentation sites served on a customer's own domain do not load our product analytics.
7. Documentation-Site Visit Counts
Hosted wikis show their owners how many people read each page. This counting is cookieless and pseudonymous: each visit is recorded under a one-way hash of connection attributes, computed with a secret salt that rotates every day and is itself deleted after 48 hours. That means visit counts cannot be linked back to a person, and cannot be linked across days. Raw IP addresses are not stored with these counts.
8. Data Retention
- Account data: retained while your account is active
- AI chat conversations and messages: 180 days
- Repository history used for changelogs and expertise (commit and pull-request metadata): 180 days
- Page-view analytics and analytics sessions: 365 days
- Wiki page feedback: 90 days
- Wiki regeneration events: 180 days
- Audit log entries: 90 days for routine actions, 730 days (2 years) for security-sensitive actions (auth, account or project deletion, key rotation, subscription, membership changes)
- Archived wiki versions: capped at the 20 most recent per wiki
- Indexed code chunks and their embeddings: kept for the current version of your documentation, and for a new version while it is being built. A version loses its chunks as soon as another replaces it (any left behind are swept after 7 days), a failed build loses them with its record after 24 hours, and all are deleted with the project
- File paths and content hashes of archived versions: 90 days
- Visitor-count salts: 48 hours (§7)
- Waitlist email addresses: until the waitlist closes, or sooner if you ask us to remove yours
You can delete your account at any time from account settings; this removes your profile and all associated projects, wikis, and conversations.
9. Your Rights (GDPR)
If you are in the EU/UK, you have the right to access, correct, export, restrict processing of, or delete your personal data, and to object to certain processing. You can:
- Export a copy of your data via the "Export my data" button in account settings (Article 20 - data portability)
- Delete your account and all associated data at any time, including your product-analytics profile
- Object to analytics by rejecting them in the cookie banner or via the footer link
- Lodge a complaint with your supervisory authority if you believe we've mishandled your data
For any other request, contact alessandro@doc0.dev. We respond within 30 days.
10. International Transfers
Several of our processors are based in the United States. Where personal data is transferred outside the EU/UK, we rely on the applicable processor's Standard Contractual Clauses or equivalent safeguards.
11. Children
doc0 is a developer tool and is not directed at children under 16. We do not knowingly collect data from children; if you believe a child has created an account, contact us and we will delete it.
12. Changes
We may update this policy as the product evolves. We will notify you of significant changes via email or an in-app notification before they take effect, and the date at the top always reflects the current version.