Privacy Policy

1. Controller

Sandro Vogel
Klingsorstr. 66
12167 Berlin
Germany

Email: sandrovogel@pm.me

2. Local project processing

Ledgermachine reads project folders, ZIP archives, source files, and Ledgermachine archives locally in your browser. It analyzes imported source as text. Project source, scan inputs, findings, and exports are not uploaded to us or evaluated on our servers.

Full source files are kept only for the active browser session and are not persisted by Ledgermachine after that session. Results-only project ledgers may include scan summaries, detected values, findings, decisions, settings, file paths, and small evidence snippets needed to explain a finding.

Exports are generated locally. They leave your device only if you choose to copy, save, publish, or share them.

3. Browser storage and offline use

Ledgermachine uses IndexedDB to save results-only project ledgers in your browser. It uses localStorage to remember your selected appearance theme. This data remains on the same device and browser, and we cannot access it.

A service worker may store Ledgermachine's own application files in browser Cache Storage so the interface can remain available offline. Project source and project ledgers are not placed in that application cache.

You can delete a project ledger inside Ledgermachine. You can remove all locally stored data by clearing the website's storage in your browser.

4. Local command-line and agent integrations

If you install or run Ledgermachine's command-line tool, Claude Code skill, or local MCP server, those components read files on your machine only when invoked. They do not send project content to the Ledgermachine website or to us.

The local MCP process and a connected AI client are separate processing boundaries. When an MCP client requests a tool or resource, the returned result becomes available to that client and may be included in model context. Results can contain project and token names, authored values, selectors, relative file locations, findings, evidence, or user-written decision reasons. Their handling is governed by the connected provider, account settings, retention policy, and any data-processing agreement. Compact findings are the default; full source-derived evidence requires explicit confirmation.

Do not connect projects containing credentials, secrets, customer records, production data, or sensitive personal data. Organisations using an AI integration are responsible for assessing their legal basis, transparency, data minimisation, access controls, retention, processor terms, and international transfers. Ledgermachine's local-first design does not by itself establish GDPR compliance for that separate use.

The optional Live Snapshot runner makes an explicit network request from your device to the public HTTPS page you select and to separately approved stylesheet hosts. The selected target can observe your IP address, user agent, requested path, and timing. The runner writes a source-free receipt locally; fetched HTML and CSS are not uploaded to Ledgermachine or retained in the receipt.

The optional Local Verified Render runner opens the local or HTTPS origin you select in a fresh Chrome or Edge context. Target JavaScript executes locally and the selected origin can observe the request. The runner does not use your personal browser profile and does not send captured page content to Ledgermachine.

5. Hosted live website scanning

When you explicitly choose Scan a live website, your browser sends the public HTTPS URL you entered to Ledgermachine's dedicated hosted scanner. The scanner requests that one page, its robots policy, and a bounded set of linked or imported static stylesheets. It does not use your cookies, credentials, or browser profile, and it does not execute the target website's JavaScript.

Fetched HTML and CSS are held only in the scanner's working memory for the bounded request and are not written to the browser project, a database, object storage, or application logs. The browser receives normalized visual primitives, findings, target identity, capture time, and boundary counts. The target operator can observe the scanner's IP address, user agent, requested path, and timing.

Abuse-prevention and operational security data may include a short-lived request challenge, a pseudonymous network-address hash, request timing, response status, and bounded resource counts. The submitted target URL must not be written to application logs. Infrastructure providers may still process network and access metadata needed to deliver and protect the hosted scanner.

6. Technical access data

When you visit the website, technically necessary access data may be processed to deliver the site and maintain security and stability. This can include your IP address, approximate location derived from it, date and time of access, requested URL, referrer information, browser and device information, status codes, diagnostics, and error data.

7. Purposes and legal bases

The website is provided on the basis of Article 6(1)(f) GDPR. Our legitimate interest is to provide a secure, stable, and functional website.

Browser storage is used to provide the local persistence, appearance, and offline features requested by the user. Where Section 25 TDDDG applies, storage and access are based on Section 25(2) No. 2 TDDDG because they are necessary to provide those requested functions.

Project data that remains exclusively on your device is not received or processed by us.

8. Hosting and source-code platforms

The website is hosted on Vercel. Vercel may process access, usage, device, and service-generated information to deliver, operate, and protect the website. Details are available in the Vercel Privacy Notice.

Vercel may process data in the United States and other countries outside the European Economic Area. Where required, such transfers rely on an applicable legal mechanism, such as an adequacy decision or standard contractual clauses.

The Ledgermachine source code and deployment workflow may be hosted on GitHub. GitHub is not involved in browser-based project analysis. If you follow a link to GitHub or interact with a repository there, GitHub processes that visit under its General Privacy Statement.

9. Cookies, analytics, and tracking

Ledgermachine does not use advertising cookies, analytics cookies, tracking pixels, or behavioral profiling. The browser storage described above supports only local application functionality.

10. Recipients and retention

Recipients of technical website-access and live-function data may include Vercel as the website and scan-function hosting provider. Folder and ZIP project inputs and locally saved ledgers are not transmitted to us. A live-scan URL and transient public page responses are processed only for the requested capture and are not written to an application database.

Local Ledgermachine data remains in your browser until you delete it or clear the site's storage. Hosting access data is retained only as long as necessary for delivery, security, troubleshooting, and abuse prevention, unless a longer period is legally required.

11. Your rights

Subject to the statutory requirements, you have rights to access, rectification, erasure, restriction of processing, data portability, and objection. Where processing is based on consent, you may withdraw that consent with effect for the future.

Because project inputs and local ledgers are not transmitted to us, these rights mainly concern technical data processed when the website is accessed or hosted.

12. Right to lodge a complaint

You have the right to lodge a complaint with a data protection supervisory authority. In Berlin, the competent authority is the Berlin Commissioner for Data Protection and Freedom of Information.

13. Automated decision-making

We do not use automated decision-making, including profiling, within the meaning of Article 22 GDPR.

14. Changes to this policy

We may update this policy if Ledgermachine's functionality, hosting setup, or legal requirements change. The current version is published on this page.