Skip to content

Privacy Policy

Last updated: April 2026

Overview

VibeWorks (the "service") is a fluency platform that scores how you work with AI. To do that, we ask you to upload a sample of your AI conversations and, optionally, to connect code or project sources. This page describes what we collect, how we use it, how long we keep it, and how you can delete it.

We keep this page short and product-accurate. If something here disagrees with what the product actually does, treat that as a bug and email us at hello@fluencity.ai.

What we collect

We collect only what we need to run the service. That falls into a few buckets:

  • Account information. Your email address and, if you set a password, its hash (bcrypt, 12 rounds). We never store the password itself.
  • Your AI conversation upload. The ChatGPT or Claude export (or pasted transcript) you submit for analysis. We immediately filter and sample it down to a small corpus (see "How analyses are stored" below) and discard the rest.
  • Your scores and progression history. The dimension-level fluency scores and snapshots produced from your uploads.
  • Connected integration tokens. If you link an external service, we store the OAuth or API tokens you grant us, encrypted at rest. We request only the minimum scopes needed to read project metadata.
  • Portfolio content. Anything you write, edit, or import into your VibeWorks portfolio.
  • Basic usage analytics. Page views, portfolio visits, and feature usage events. Portfolio view records store SHA-hashed IP addresses, not raw IPs, so we can count unique viewers without being able to identify them individually.

We do not collect or store payment card information. If we charge for a feature in the future, payments are processed by Stripe and subject to their privacy policy.

How analyses are stored

Your raw AI conversation export is the most sensitive thing you give us, so it gets the most aggressive treatment:

  • Uploads are filtered server-side to strip API keys, secrets, and other sensitive content before anything is stored.
  • We sample down to a small corpus, at most around 90 KB of conversation snippets per user, which is what the rubric needs to compute your scores.
  • The full, unfiltered export is discarded after that pipeline runs. We do not keep your complete chat history.
  • The sampled corpus is retained for 90 days so we can re-score it when the rubric improves, then auto-deleted. Your scores and progression history persist after that. Only the underlying text is removed.
  • ZIP uploads enforce entry-count, uncompressed-size, and compression-ratio limits to block zip-bomb attacks, and file types are validated server-side.

How we use your data

We use the data described above to:

  • Compute your fluency scores and produce your dashboard and snapshots.
  • Render your public portfolio and Growth Snapshot pages, when you choose to publish them.
  • Show you product analytics about your own portfolio (such as view counts).
  • Improve our scoring rubric and product. When we re-score the sampled corpus to evaluate rubric changes, that work happens on your existing stored sample. We do not pull additional data to do it.
  • Detect abuse, debug failures, and operate the service.

We do not sell your data. We do not use your conversations, code, or portfolio content to train external AI models, and we do not share them with third parties to do so.

Third parties we share data with

VibeWorks runs on managed infrastructure and uses a small number of vendors to do its job. Where data is shared, it is the minimum necessary for the integration to work:

  • Hosting and database providers. Your account data, scores, and sampled corpus are stored on managed cloud infrastructure with SOC 2 Type II compliance.
  • AI scoring providers. To compute scores from your sampled corpus, we send sampled snippets to a large-language-model provider acting as our subprocessor. We do not authorize them to retain your content for model training.
  • Connected integrations. If you link an external service, we exchange data with that service on your behalf using the tokens you grant us. You can revoke any integration at any time by emailing us, and you can revoke our access from the third party as well.
  • Email. Transactional email (magic links, password resets, notifications) is sent through an email-delivery vendor.
  • Stripe. If you ever pay for the service, payment details go directly to Stripe and are subject to their privacy policy. We do not store card numbers.

Privacy of your portfolio

Your scores, snapshots, and portfolio content are private by default and live on a private dashboard. They are only visible publicly if you explicitly publish your portfolio or share a Growth Snapshot link.

Public portfolio pages and shared snapshot pages are intentionally unauthenticated. Anyone with the URL can view them. Treat them as public.

Cookies and tracking

We use a small number of cookies and similar identifiers:

  • A signed, HTTP-only session cookie so you stay logged in. It is invalidated on logout.
  • An optional "remember me" cookie, set only if you leave "Keep me signed in" ticked at sign-in, so you stay logged in across browser restarts.

We do not use third-party advertising trackers and we do not run cross-site behavioural ads.

Retention and deletion

You can delete your data at any time:

  • Delete your conversation corpus. Email us and we'll purge the stored sampled corpus. Your scores stay; the underlying text is gone.
  • Disconnect integrations. Disconnecting an integration in your dashboard removes the stored OAuth or API token.
  • Delete your account. Email us to delete your VibeWorks account. Personal data, portfolio content, scores, and connected integration tokens are permanently removed from our production systems within 30 days. Anonymized aggregate analytics may be retained for longer.
  • Automatic expiry. Sampled conversation corpora auto-expire 90 days after upload even if you take no action.

Send export or deletion requests to hello@fluencity.ai from your account address.

Children

VibeWorks is not directed at children under 13, and we do not knowingly collect personal information from them. If you believe a child has provided us data, email us and we will delete it.

Changes to this policy

When we make material changes to this policy we'll update the "Last updated" date at the top of the page and, where appropriate, notify users in-product or by email. Continued use of VibeWorks after a change constitutes acceptance of the updated policy.

Contact

Questions about this policy or your data? Email hello@fluencity.ai. For security-specific concerns, see our Security Policy.