Privacy Policy

Last updated: 2026-11-13.

This policy explains what WikiFix does with your data. WikiFix is operated by TOTEMDEV SRL, trading as Totem Dev (“Totem Dev”, “we”), company number (IDNO) 1022609000314, registered at Miron Costin 19/4, ap. 37, Chișinău MD-2068, Republic of Moldova. For anything not covered here, email security@wikifix.ai. Region-specific rights are at the end, under For EU residents and For non-EU residents.

Who this applies to

WikiFix is a B2B product that comes two ways: WikiFix at app.wikifix.ai, where a team signs up, connects Confluence Cloud or Notion and pays through Paddle, and WikiFix for Confluence, installed from the Atlassian Marketplace, which can also read the Notion workspaces and other Confluence sites you connect to it. The “customer” is the team that signs up, or the Atlassian organization that installs the app. The data we handle is the content of the sources the customer connects, which we process as the customer’s processor under our Data Processing Agreement, and the account details of the people who use WikiFix, which we hold as controller. The headings below say which is which.

Cookies on this website

This section is about wikifix.ai itself – the site you’re reading right now – separately from the product described below. We use one analytics tool, PostHog, on its EU instance, to see which pages help and where visitors drop off.

Nothing loads until you click Accept on the cookie banner shown on your first visit. If you reject it or leave it unanswered, no analytics script loads at all – the only cookie set is the one remembering your choice, which lasts about 6 months before asking again. Change your mind anytime via Cookie preferences in the footer.

If you do accept, two things go further than page counting, and you should know about them:

  • Session replay is on, and form inputs aren’t masked. That includes what you type into the contact form.
  • If a contact-form submission fails to reach us, the details you typed are sent to PostHog – name, work email, company, role, and your notes – so we can follow up by hand instead of losing your message.

The same toggle covers script-error reporting, since it runs through the same PostHog integration.

WikiFix at app.wikifix.ai sets no analytics cookie: its PostHog analytics keep nothing in your browser between visits. Session replay, in WikiFix at app.wikifix.ai and in WikiFix for Confluence, masks every input and the visible text on the screen, leaves out elements that hold your text in an attribute, and records layout and clicks. It doesn’t run on the Strict Data handling level.

What your browser keeps. Your browser keeps your sign-in, so you stay signed in, and a cache of what you’ve already loaded, so pages open faster. At app.wikifix.ai that cache holds finding text, page titles, people’s names and the email addresses of your team’s members, and a reply-to address if you typed one into feedback. Signing out clears it. In WikiFix for Confluence the cache lives in the Confluence frame, which has no sign-out: it stays until your browser’s data is cleared, and its feedback entry is keyed with your Atlassian account id. Once data is on your device it is the device’s copy, not ours. Deleting a team does not reach into browsers that already loaded it.

What we collect

Source content (processed, mostly not stored). When you run a scan, we read the pages in the spaces or Notion pages you select, directly from Confluence or Notion. We do not keep a copy of your wiki. What persists in our database is:

  • Text derived from your pages, which can repeat their wording and is kept whether or not it leads to a finding, and a numeric index of them.
  • Questions and answers on a finding, and comments you ask us to post.
  • When a fix is previewed, the passage it would replace and its proposed text, so it can be applied and reverted. For a duplicate-pages merge, that is the whole text of each page merged away.
  • A search index, so you can search your findings.
  • The ids of each page’s creator, owner and last editor in your source, so a finding can say whose page it is.

Pages are read live from your source on each scan. The text above is encrypted column by column.

From here to Server logs, we hold the data as controller, apart from the address a scan notification goes to, which is part of what we process for the customer. AI requests and responses, at the end, is customer content.

Your web account (app.wikifix.ai). To sign you in, we use Amazon Cognito, part of AWS. It holds your email address and your password, which it stores as a verifier we never see. In our own database we keep your email address (encrypted), your role in your team (admin or member), and the name you give your team. We don’t ask for your name. When an admin invites someone, we keep the invitee’s email address (encrypted), who invited them, and when the invitation expires.

Source connections (app.wikifix.ai). When an admin connects Confluence or Notion, and when a member links their own account to apply fixes, we hold the access tokens Atlassian or Notion issues for that person, encrypted. We also keep which person connected each source, their Atlassian account id or Notion user id, and the Notion workspace’s name and id. When a finding needs to name a page’s owner, we look up their display name in Notion at that moment, without their email address, and don’t store it.

Account identifiers (WikiFix for Confluence). To run scheduled scans and to apply or revert fixes on your behalf, we hold the access tokens the Atlassian Forge platform issues for each participating user, encrypted, and rotate them. We use an internal user id (our own identifier, mapped from your Atlassian account at sign-in) as the primary key across our systems. Amazon Cognito also holds a record for each person who uses WikiFix for Confluence, named with their Atlassian account id, with no email address or password. While you sign in, a short-lived record in DynamoDB, a database table from AWS, holds your Atlassian account id and site id for about five minutes, and, while an admin picks a site to connect, the sign-in state with encrypted tokens for ten minutes. Those records are deleted automatically when they expire, which AWS can take up to about 48 hours to do; the table behind the first has a point-in-time backup that keeps removed items for up to 35 days. When someone signs in to a Notion workspace, or to a Confluence site connected beside the one WikiFix is installed on, we hold the access and refresh tokens that service issues to them, encrypted, with their account id there – and, for Notion, the workspace’s id and name. We read no email address from Notion.

Email. In WikiFix at app.wikifix.ai, we email the codes that confirm your sign-up and reset your password, invitations (with the team’s name, the role, and the inviter’s email address), and scan-completion summaries and monitoring digests, all to the address on your account. In WikiFix for Confluence, to email the scan-completion summary to the report’s owner, we read that user’s email address from Atlassian at send time (via Confluence’s read:email-address scope, using the app identity). We never store it, and we use it only for sending that completion email – apart from the feedback case described next, which you can turn off before you send.

Feedback you send us. When you send feedback – from the Feedback button, or the optional note when you ignore a finding – on the Standard Data handling level the form shows your email address and a tick-box saying we may reply to you. It starts ticked, so if you leave it as-is, your address goes to PostHog with your message, and PostHog passes both on to our Slack, where we read it and can answer you. Untick it and no address is sent; you can also type a different one. On Restricted and Strict, your message goes without an address. In WikiFix for Confluence, if your Atlassian profile keeps your email private, we never see it, and the form simply offers you an empty box instead.

Payments (app.wikifix.ai). Paddle sells WikiFix on the web as our merchant of record, so it collects your billing name, email address, address and payment details itself, under its own privacy notice. We never see your card. When you check out, we send Paddle your team’s WikiFix id and what you’re buying – a plan, or credits and their price. From Paddle we keep the ids of your Paddle customer, your subscription and each payment; the plan you’re on, its status, when its billing period ends and when a cancellation takes effect; for each payment, what it bought, the amount and currency, when it was made, and any refund of it; and which of Paddle’s billing events we have processed. We also keep your auto top-up settings, and when and by which admin they were agreed to. Invoices stay with Paddle: when you open one, we fetch a link to it. No page content goes to Paddle.

Product analytics and error data. We use PostHog for product analytics and Sentry for error tracking. Neither receives page content. PostHog events pass an allow-list per event on our servers, and a block-list of content and external-id fields in the browser. Comments you choose to write in feedback, and the reply-to address described above, go there too. Events also carry page addresses, browser and approximate country. Sentry receives stack traces, error codes and page addresses, with no names or emails, and performance traces: the timings of requests and page loads, with addresses but no query strings. What each receives depends on your team’s Data handling level; on Strict, only feedback you send goes to PostHog, and nothing we can tie to your team goes to Sentry. The web sign-in and sign-up screens come before we know your team, so on every level they send error reports and a page-load event as on Standard. PostHog is set to discard IP addresses, and these screens aren’t recorded.

Server logs. Our servers log each API request with its IP address, for security, our internal ids and our vendors’ reference numbers. They hold no names, email addresses or page content.

AI requests and responses (content-bearing). On the Standard level, we record each AI request and response to Langfuse, our monitoring subprocessor, so we can look into a problem when you ask us to. Unlike product analytics and error tracking, this does include content: the full prompt we send to the model and the model’s reply embed the page text under review. Restricted keeps only cost and timing, and Strict records nothing.

Where we process it

Your data is stored and processed in AWS Frankfurt (eu-central-1), with one exception described below. Sign-in at app.wikifix.ai runs on Amazon Cognito in the same region. Default AI inference runs on AWS Bedrock through an EU inference profile that stays within EU regions; cross-continent inference is blocked in code. Analytics (PostHog) and error tracking (Sentry) store their data in the EU, and can process some of it in the US as described below, and Langfuse is pinned to an EU-resident instance.

Two analysis checks: in the EU with EU residency, otherwise in the US. These two checks help decide whether a finding stands. With EU residency required by your team or install, they run on different AI models, which reside in the EU: eu/jev, operated by Bevelites GmbH on its own server in Germany, while every other AI step stays on AWS Bedrock in the EU; a scan that cannot meet the requirement stops rather than running elsewhere. Without it, on every Data handling level, they run on TypeSafe, whose endpoint processes in the United States. Either way, a request receives excerpts of the pages it checks, which can come from several pages: each passage with its page’s title and headings and the section around it, up to 2,500 characters of section. On a short page that can be the whole page; it never receives the whole space. eu/jev doesn’t store the excerpts it receives, in its database or its logs, and doesn’t use them to train AI models. We use TypeSafe with zero data retention: it doesn’t keep the excerpts it receives or use them for anything else, including training AI models. Today we run one region, and EU is the only residency we can serve. Residency is a setting on the Residency tab – under Admin at app.wikifix.ai, and in WikiFix Admin in WikiFix for Confluence; see Data residency.

Exception – bring-your-own-key. If you configure your own Anthropic API key, the AI calls made on it – the analysis checks above included – go to Anthropic’s API directly, whatever residency you’ve set, which is outside our EU-region pin and under your own agreement with Anthropic. Indexing stays on AWS Bedrock, and those calls are still recorded to Langfuse as your Data handling level sets. Use the default path, with EU residency required, if EU residency for every inference call matters to you.

Transfers outside the EU

Where personal data leaves the EU/EEA, this is where it goes and what protects it:

  • WikiFix itself, in Moldova. TOTEMDEV SRL is established in the Republic of Moldova, which has no EU adequacy decision. Your data is stored in Frankfurt, but we administer WikiFix remotely from Moldova, and reaching data from a third country is a transfer under the GDPR even though nothing is stored there, apart from working files from a read you permitted in writing, which we delete afterwards. Storing so little in Moldova lowers the risk without removing the transfer. For content we process on a customer’s behalf, the safeguard is the EU Standard Contractual Clauses (Module 2, controller to processor), which are part of our Data Processing Agreement, together with the UK Addendum for UK data. We keep a written assessment of this transfer, as the Clauses require. Moldova is a party to Council of Europe Convention 108 and has ratified the protocol that modernises it (Convention 108+), which is not yet in force. Its Law No. 195/2024, in force since 23 August 2026, transposes the GDPR. The account, analytics and log data we hold as controller sit in the same places and are reached the same way.
  • TypeSafe, in the United States. With EU residency required (choose European Union under Admin → Residency, or WikiFix Admin → Residency in WikiFix for Confluence), nothing goes to TypeSafe: its checks run on eu/jev in Germany. Without it, on every Data handling level, TypeSafe receives excerpts of the pages it checks. We use TypeSafe with zero data retention, so it doesn’t keep the excerpts it receives or use them for anything else, including training AI models. The transfer is covered by the EU Standard Contractual Clauses in TypeSafe’s Data Processing Addendum; our Sub-processors page has the detail.
  • Paddle, in the United Kingdom, receives what you give it at checkout and, from us, your team’s WikiFix id and what you’re buying. The United Kingdom is covered by the European Commission’s adequacy decision.
  • Sentry and PostHog store their data in the EU, but both are US companies and can process some of it in the US, under the EU-US Data Privacy Framework and the EU Standard Contractual Clauses. We send neither of them page content. Feedback you write reaches PostHog as you wrote it.
  • Slack, a US company, receives the feedback you send us, which reaches our Slack through PostHog.
  • Anthropic, only if you bring your own Anthropic API key: your prompts go to Anthropic’s API under your own agreement with them, outside our EU-region pin.
  • Atlassian receives what you type into the contact form on this website, through Jira Service Management. That data isn’t pinned to an Atlassian data-residency region, so Atlassian may hold it outside the EU; the transfer is covered by the EU Standard Contractual Clauses in Atlassian’s Data Processing Addendum.

Some of our providers, AWS among them, are incorporated in the United States even where they store and process your data in the EU; Government access below explains what that means.

How we use it

To run the service: sign you in, scan your selected spaces and pages, find outdated and contradictory pages, propose fixes, and – when you approve – apply or revert them. Your content is never used to train AI models, or to tune or improve our AI prompts; default inference on AWS Bedrock does not use prompts or completions for training per AWS’s terms. A person at WikiFix reads your content only when your organization asks us to in writing or gives explicit written permission, for the problem it raised.

The GDPR asks us to name the legal basis for each use of personal data. For the data described above:

  • Source content, including what reaches Langfuse: we process it on your organization’s behalf, on its instructions and under our Data Processing Agreement, so the basis is your organization’s.
  • Your web account, source connections, account identifiers, and the email we send you: the contract with you – without them we can’t sign you in, scan your sources, apply your fixes or tell you a scan is done.
  • Payments: the contract for what you buy. Paddle, as merchant of record, keeps the accounting records under its own legal obligation and handles your payment details as its own controller.
  • Product analytics, error data and server logs: our legitimate interests in keeping WikiFix secure and working and in seeing which parts of it are used – limited to what’s described under What we collect.
  • Feedback you send us, which reaches us through PostHog and on to our Slack: our legitimate interests in improving WikiFix. Your email address goes with it only if you leave the reply tick-box ticked. This also rests on our legitimate interest in answering you, and you can untick it.
  • Cookies and analytics on this website: your consent, given on the cookie banner. You can withdraw it at any time via Cookie preferences in the footer.
  • The contact form on this website: our legitimate interests in answering you.

How long we keep it

WikiFix at app.wikifix.ai. When an admin deletes your team, every member loses access at once, and we delete the team’s data 30 days later, automatically – a daily job starts the deletion of teams once they pass the 30-day mark. When someone is removed from a team, their account row stays with the team until then, so their past fixes stay attributed to them. We stop using their source sign-ins and ask Notion to revoke theirs; their encrypted tokens stay until the team’s data is deleted. Your sign-in itself – your email address and password in Amazon Cognito – stays during those 30 days, so you can create or join another team. When the team’s data is deleted, your sign-in goes with it unless you still belong to another team; email security@wikifix.ai to delete it sooner.

WikiFix for Confluence. We delete your data about 30 days after uninstall, automatically – a daily job starts the deletion on its first run past the 30-day mark, whether or not you ask, once every install on the team is uninstalled. The 30 days exist so that a re-install picks up where you left off. Uninstalling stops our scans of your Confluence site at once and deletes the app token we cached for it. Notion workspaces and other Confluence sites that someone added by hand stay connected for those 30 days, with their scheduled scans waiting, and their sign-in tokens are deleted when the 30 days are up.

Disconnecting a source deletes the sign-in tokens we hold for it, and nothing already stored from it: its findings and reports stay until the team is deleted. Deleting a team at app.wikifix.ai deletes its tokens at once. We also ask Notion to revoke the Notion ones, whenever we delete them. Atlassian gives apps no way to revoke a token, so a person’s WikiFix grant stays listed under Connected apps in their Atlassian account until they remove it. If you want anything gone sooner, email security@wikifix.ai. Database backups roll off after 7 days.

What runs on its own clock. Deleting a team doesn’t reach what’s listed here: each expires on its own clock below. That job covers our database and sign-in records, apart from your team’s change history (who changed its configuration, permissions and consent, and when, and who created or opened its share links) and the ids that tie it together (ours, Atlassian site, account and group ids, and Notion workspace ids), which we keep, each entry for three years from its date, even after your team is deleted. Email security@wikifix.ai and we’ll delete it sooner. Text in Langfuse is deleted about 30 days after it’s recorded, or usually within about a day of moving to Restricted or Strict, and longer for a team with many records; server logs expire after 90 days; analytics events go after 3 months, at PostHog’s next monthly cleanup; error data after 30 days, and session replays after 30; database performance records after 15 months; the terms of Brevo, which delivers our emails, set no fixed retention period. TypeSafe keeps none of the excerpts it receives and uses none of them for anything else, including training AI models. eu/jev, which runs those checks with EU residency, doesn’t store the excerpts it receives. A summary kept in Atlassian’s Forge storage for the Rovo agent has no expiry from us while WikiFix is installed, and our deletion job doesn’t clear it; Atlassian deletes app storage after an uninstall, on its own retention schedule. What your browser holds, described under Cookies on this website, is on your device and stays there until you sign out or clear it.

Who we share it with

We use third-party sub-processors to run the service. Rather than half-list them here and let the two versions drift apart, the current and authoritative list – every vendor, what it’s for, the data it processes, and where it sits – is on our Sub-processors page, and every change to it is published to a feed you can follow. Two are worth calling out in this policy because they receive content: Langfuse, on the Standard level (see “AI requests and responses” above), and TypeSafe (see “Where we process it”). Payments for WikiFix at app.wikifix.ai run through Paddle, and for WikiFix for Confluence through the Atlassian Marketplace. In WikiFix for Confluence there is one more recipient, and you choose it: when a person uses the WikiFix agent in Atlassian’s Rovo, Atlassian’s Rovo AI receives finding text, page titles and the passages behind them, the text of any page the agent reads for them and, when they settle a finding, the before and after text. Your page bodies are already at Atlassian. It happens on every Data handling level, Strict included, because the level doesn’t apply to it. To settle a finding, the agent also keeps one summary per person in Atlassian’s Forge storage until they settle it or prepare another, with no expiry from us while WikiFix is installed. Beyond these, we share personal data only with our professional advisers (such as lawyers and accountants) where they need it to advise us, and with authorities where the law requires it (see Government access). We don’t sell your data to anyone, ever.

Government access

WikiFix is operated by TOTEMDEV SRL (trading as Totem Dev), incorporated in Moldova. Where authorities can compel data, they reach it through the providers that hold it and through legal-assistance channels; our place of incorporation does not put it beyond reach. Your content is held by AWS, a US-incorporated provider that can be compelled under the US CLOUD Act regardless of where it stores data, including the EU. Authorities can also seek assistance from Moldova under mutual-legal-assistance treaties (the US and Moldova are both parties to the Budapest Convention on Cybercrime, UNTOC, and UNCAC). We operate no government data-access program: we require valid legal process, challenge overbroad or unlawful requests, and notify you where legally permitted. What limits exposure in practice is EU-region processing, data minimization (we keep what What we collect lists, not a copy of your wiki), and encryption.

For EU residents

If you are in the EU or EEA, the GDPR applies. Your organization is the data controller for the content of the sources it connects; WikiFix (Totem Dev) acts as your processor. For your sign-in at app.wikifix.ai, we decide what is collected and why, so for that data we are the controller. You have the rights of access, rectification, erasure, restriction, portability, and objection, and, where we rely on your consent, the right to withdraw it at any time. To exercise them, email security@wikifix.ai – we answer within one month, as the GDPR requires; you can also delete your team or uninstall WikiFix at any time, which ends our reading of your sources (deleting a team ends it at once; an uninstall stops scans of that site at once, and the rest follows as How long we keep it describes) and leads to deletion of its data about 30 days later, apart from the records How long we keep it lists on their own clocks. Your data is stored in the EU (AWS Frankfurt). With EU residency required, all AI processing stays in the EU too, unless you’ve also added your own Anthropic key; without it, two checks run in the US, as described under Where we process it. Our GDPR-aligned Data Processing Agreement applies automatically when you accept our Terms, with nothing to sign. If your procurement process needs a countersigned copy, email security@wikifix.ai. You may lodge a complaint with your national supervisory authority. Totem Dev’s home jurisdiction, Moldova, is a party to the Council of Europe’s Convention 108 on data protection and has ratified the protocol that modernises it (Convention 108+), which is not yet in force; see Transfers outside the EU for what that means for your data.

For non-EU residents

If you are outside the EU/EEA, your data is still stored in the EU (AWS Frankfurt) and processed there, with the one US exception described under Where we process it – EU is the only residency we can serve today. We extend the same core handling to everyone, regardless of location: data minimization, encryption, no use of your content to train AI models, and access, correction, or deletion on request via your organization at security@wikifix.ai. The Government access section above applies to you as well.

US residents (California / CPRA). We do not sell or share your personal information, and we do not use it for cross-context behavioral advertising. You may request to know, access, or delete the personal information we hold by emailing security@wikifix.ai.

Changes

We’ll update this page when our practices change and update the date at the top. The data-handling detail on the security page is kept in sync with this policy.

Contact

security@wikifix.ai.