Privacy policy

Privacy Policy

Last updated: April 24, 2026

dir-tree is designed as a local-first directory tree tool. This page explains the boundary between the core browser-only workflow and the optional AI annotation workflow, so you can decide what is safe to load, edit, and send.

TL;DR

Your files never leave your browser for core features. dir-tree reads names and hierarchy to build an editable tree, not source code or document contents. Optional AI annotation sends only compact visible path structure and existing annotations.

Core directory reading, editing, rendering, and export run in your browser.
dir-tree does not upload file contents for core tree generation.
AI annotation is optional and sends compact path structure, existing annotations, and language preference.

Important note about path names

Even without file contents, path names can reveal sensitive context. Review visible names before using AI annotation, especially when trees include client names, internal codenames, unreleased products, private identifiers, or contract details.

clients/acme-secret-project/contract-draft.md

SECTION 01

Core features process locally

For the core workflow, dir-tree processes directory structure in the browser. It needs names, hierarchy, visibility settings, manual annotations, export settings, and optional display metadata only when you enable those display options.

  • Opening or dragging a local folder.
  • Parsing a local ZIP archive in the browser.
  • Importing an ASCII, JSON, XML, HTML, or Markdown tree file.
  • Editing, hiding, annotating, rendering, copying, and downloading tree output.

SECTION 02

What dir-tree does not collect for core features

dir-tree does not upload your files for core directory tree generation. The product is built around structure, so the browser can render and export the tree without sending file bodies to dir-tree servers.

  • File contents are not uploaded for core tree generation.
  • Manual tree edits stay in the browser unless you copy or export them yourself.
  • Account creation is not required for the current core workflow.

Not collected for core tree generation

Source code content
Document body text
Image or binary file contents
Secrets, API keys, or credentials inside files
File permissions
Modification times unless metadata display is enabled

SECTION 03

Optional AI annotation

AI annotation is optional and only runs when you explicitly request it. The feature is designed to generate first-draft comments from paths and visible structure, not from file contents.

  • What may be sent: visible file and folder paths, file/directory markers, node count, existing annotations, UI locale, and browser language preference.
  • What is not sent: file contents, source code content, document body text, binary file contents, file permissions, file size, or modification time.
  • If names reveal sensitive customer names, internal codenames, unreleased products, or private identifiers, use manual annotation or hide/rename those paths before requesting AI comments.

AI annotation data boundary

AI comments are generated from structure, not from file contents. The request should contain only the visible tree context needed to draft useful labels.

May be sent when AI is used

  • Visible file and folder paths
  • Node type markers such as file or directory
  • Approximate node count and visible depth
  • Existing manual annotations
  • UI locale and browser language preference

Never sent by AI annotation

  • Source code or document contents
  • Binary file contents
  • Secrets inside files
  • File permissions
  • File sizes
  • Modification timestamps

SECTION 04

Remote ZIP URLs

If you enter a remote ZIP URL, your browser fetches that URL so dir-tree can parse the archive. The remote host may receive standard request information from your browser, such as IP address, user agent, and requested URL.

  • dir-tree should not proxy or store the remote ZIP unless a future server-side fetch feature is explicitly introduced and documented.
  • Only use remote ZIP URLs you are allowed to access and process.

SECTION 05

Browser storage

dir-tree may store UI preferences, display settings, reader control settings, and theme preference in browser storage so the interface can remember your choices.

  • This data stays on your device.
  • You can clear it through your browser settings.
  • The theme toggle stores either light or dark as a local preference.

SECTION 06

Hosting, logs, and third-party services

The website may be hosted on a third-party infrastructure provider. That provider may process standard infrastructure logs such as IP address, request path, timestamp, and user agent. Optional AI annotation uses a third-party AI service to process the compact tree payload described above.

  • Infrastructure logs are separate from local tree processing.
  • AI requests should only contain compact path structure and annotation context.
  • Provider retention and model-training settings must be verified against the actual production AI provider contract before launch.

SECTION 07

Data retention

Core local tree data is not stored on dir-tree servers. AI annotation requests should be processed transiently and not used for model training unless the selected AI provider and product policy explicitly say otherwise.

  • This policy should be updated if user accounts, cloud saving, collaboration, analytics, or server-side imports are introduced.
  • Future paid cloud features will need a separate retention and deletion policy.

SECTION 08

Children's privacy

dir-tree is a developer tool and is not directed at children under 13. We do not knowingly collect personal information from children.

SECTION 09

Changes to this policy

We may update this policy as the product evolves. Material changes should be reflected on this page with an updated effective date.

SECTION 10

Contact

For privacy-related questions, contact the project maintainer through the public project channel or the contact method published on dir-tree.com.