Buy me a coffee
Plain-language privacy

Privacy Policy

Claude2PDF is designed to be simple and auditable. This policy states exactly what passes through the server, what stays in your browser and what the current application does not collect.

Effective: June 19, 2026Temporary server extractionNo conversation database

1. Scope

This Privacy Policy applies to the Claude2PDF website and hosted conversion service. Claude2PDF converts supported public conversation share pages from Claude, ChatGPT, Gemini and Grok into a browser-rendered document that you can print or save as a PDF.

2. The exact processing path

Claude2PDF is not a fully local converter. The extraction step is performed by the Claude2PDF server, while parsing, preview and PDF generation happen in your browser.

  1. Your browser sends the public share URL to Claude2PDF's /api/extract endpoint.
  2. The server opens that public page with Puppeteer, a headless browser.
  3. The server reads the rendered page HTML, which necessarily includes the visible conversation content.
  4. The server returns that HTML to your browser.
  5. The browser-side parser removes interface noise, identifies the conversation turns and builds the document preview.
  6. Your browser's print engine creates the PDF.

In practical terms, the conversation content passes through Claude2PDF's server during extraction. It is not accurate to describe the complete conversion as local-only.

3. What the server can access during extraction

Because the server must load and render the public page, it can temporarily access the public share URL and the full rendered conversation content needed for conversion. This processing occurs in server memory for the purpose of completing the request.

  • Claude2PDF does not sign into your AI account.
  • Claude2PDF does not access private chats that are not available through the submitted public link.
  • The current application code does not intentionally write conversation HTML or parsed conversation content to a database or file.
  • The current application code does not intentionally print conversation HTML, conversation bodies or complete request payloads to application logs.

Public share links can be accessible to anyone who has the URL. Do not submit a link containing information you are not authorized to process or information you would not want handled temporarily by an online service.

4. What Claude2PDF does not currently collect

The current open-source application does not include:

  • user accounts, authentication profiles or mandatory email collection;
  • a conversation database or conversion-history database;
  • Google Analytics, Google Tag Manager, PostHog, Mixpanel, Hotjar, Meta Pixel, LogRocket or similar behavioral analytics tools;
  • Sentry or another third-party application-error reporting SDK;
  • an in-product feedback form, rating database, newsletter or mailing-list signup;
  • code that sends conversation content to an AI API or another content-processing API.

This list describes the current application code and may change only if the product changes and this policy is updated.

5. Browser session storage

Claude2PDF uses browser sessionStorage to pass the source URL, normalized conversation and conversion status between the home, loading and preview pages. This information is stored in the current browser session, not in a Claude2PDF user account.

Session data is cleared when you start another conversion, when the application removes it, when you clear site data or when the browser session ends. Claude2PDF does not use localStorage as a persistent conversation archive.

6. PDF generation

The preview and document template are rendered in your browser. The Download PDF action invokes the browser's print function. Claude2PDF does not receive, upload or intentionally store the finished PDF file.

7. Application logs and infrastructure metadata

The application records limited operational messages such as the source hostname, extraction success or failure and error messages. The current code does not intentionally log the full public URL, rendered HTML or conversation body.

The hosting provider and normal internet infrastructure may still process technical metadata such as your IP address, timestamps, request headers, response status, traffic volume and security events. Those systems are operated under their own infrastructure and retention practices.

8. Cookies and analytics

Claude2PDF does not currently include behavioral analytics, advertising pixels or advertising cookies. The active conversion flow relies on browser session storage. External resources and external pages may use their own cookies or similar technologies under their own policies.

9. Third-party services

Claude2PDF relies on or links to third parties:

  • Source AI platforms: Anthropic, OpenAI, Google and xAI host the public pages you ask Claude2PDF to retrieve.
  • Google Fonts: font files may be requested from Google when a page loads, which can expose ordinary request metadata to Google.
  • Tailwind CDN: the home interface currently loads Tailwind CSS from a content-delivery network.
  • GitHub: the open-source repository and “Star on GitHub” link are hosted by GitHub.
  • Buy Me a Coffee: the support button opens an external page only when you choose to visit it.
  • Hosting infrastructure: the website and Puppeteer extraction server run on a hosting provider that processes ordinary network and operational data.

Claude2PDF does not control how these third parties process data. Review their terms and privacy policies before using their services.

10. Retention

The current application is designed to process conversation content transiently and does not intentionally retain conversation HTML or parsed conversation content on the server after the extraction request is completed. The normalized conversation may remain in your browser's session storage for the active session as described above.

Normal hosting, security and network logs may be retained by infrastructure providers for their own operational periods. Claude2PDF does not promise that no transient copies can ever exist in memory, network buffers, crash data or provider-managed infrastructure.

11. Security and sensible use

Claude2PDF uses URL validation, supported-host restrictions, request rate limits and HTTPS when deployed on its production host. No online service can guarantee absolute security.

Use Claude2PDF only with public share links. Avoid processing highly sensitive, confidential, regulated or private information unless you have independently decided that temporary server-side handling is acceptable for your use case.

12. Your choices

You can stop a conversion before submitting a link, close the tab, clear browser site data or avoid optional external links. Claude2PDF does not maintain a user profile or conversion-history account for you to manage.

13. Open-source transparency

Claude2PDF's source code is publicly available at github.com/Dunegerb/claude2pdf. The repository allows anyone to review the application's extraction flow, parser logic, dependencies and storage behavior. Do not include private conversation content or sensitive personal information in a public GitHub issue.

14. Children

Claude2PDF is not directed to children under 13. Do not submit personal information about a child unless you have the legal authority to do so.

15. Changes

This policy may change if the architecture, supported providers, dependencies or legal requirements change. The effective date will be updated when material revisions are published.

16. Questions

Privacy questions and technical reports can be submitted through the public GitHub repository. Do not post a private share URL, conversation text, personal data or other sensitive material in a public issue.

Claude2PDF is an independent open-source tool and is not affiliated with, endorsed by or operated by Anthropic, OpenAI, Google or xAI.