Team Bookmarks

A lightweight open source web app to manage, share and version your team's bookmarks as a YAML file.

• Accessible: keyboard navigation, ARIA, Lighthouse accessibility score = 100
• Lightweight: main bundle < 20 KB, heavy dependencies lazy-loaded
• Self-hosted: single Docker image, no database, no auth
• CLI: init and export bookmarks from the terminal, no app required

Features

• Create folders and sub-folders of links
• Import a .yml bookmark file
• Export as YAML (for Git versioning), NETSCAPE HTML (browser import) or JSON
• All edits saved in localStorage, persists across page reloads
• Example collection pre-loaded with Slashgear's favorite tools

CLI

The package ships a CLI so your team can manage bookmarks directly in the repository, no web app needed.

Init a new bookmarks file

npx @slashgear/team-bookmarks init "My Team"
# bunx @slashgear/team-bookmarks init "My Team"
# pnpx @slashgear/team-bookmarks init "My Team"

Creates a bookmarks.yml with a starter structure and inline usage comments. Commit it to your repository.

Export to HTML or JSON

# Generate a Netscape HTML file ready to import in any browser
npx @slashgear/team-bookmarks export bookmarks.yml

# Generate JSON
npx @slashgear/team-bookmarks export bookmarks.yml --json

# Custom output path
npx @slashgear/team-bookmarks export bookmarks.yml -o dist/bookmarks.html

Workflow with Git

init bookmarks.yml  →  edit YAML  →  commit  →  share with team
                             ↓
                    npx … export → bookmarks.html → import in browser

Edit with an AI agent

An edit-bookmarks skill is included under .agents/skills/edit-bookmarks/SKILL.md. It is compatible with any agent runner that supports the open skills format (opencode, Claude Code, Cursor…).

skill: edit-bookmarks — add a link to Figma in the Design folder
skill: edit-bookmarks — rename "General" to "Resources"
skill: edit-bookmarks — remove the link to example.com

The agent reads and edits the YAML directly, preserving comments and formatting.

Quick start

Docker

docker run -p 3000:3000 ghcr.io/slashgear/team-bookmarks:latest

Then open http://localhost:3000.

Docker Compose

services:
  team-bookmarks:
    image: ghcr.io/slashgear/team-bookmarks:latest
    ports:
      - "3000:3000"
    restart: unless-stopped

docker compose up -d

YAML format

name: My Team
folders:
  - name: Dev Tools
    links:
      - label: GitHub
        url: https://github.com
    folders:
      - name: Monitoring
        links:
          - label: Grafana
            url: https://grafana.com

Save this file in your repository and import it whenever you need to update your team's bookmarks.

Workflow

Browser app:

Edit in browser  →  Export .yml  →  Commit to Git  →  Share with team
                                         ↓
                               Anyone can re-import,
                               edit and export again

CLI / Git-native:

npx … init  →  Edit YAML  →  Commit  →  npx … export  →  Import in browser
                  ↑
        /bookmarks "add a link to…"   (Claude Code)

Tech stack

| Layer | Technology | | ----------- | ---------------------------------------------------------- | | Frontend | Preact + TypeScript | | Validation | Zod mini | | YAML | js-yaml (lazy-loaded) | | Server | HonoJS on Node 22 | | Compression | Brotli (q11) + Zopfli | | Build | Vite | | Tests | Vitest | | Linting | Oxlint | | Formatting | oxfmt | | Releases | Changesets |

Local development

Requirements

• Node.js 22+
• pnpm 10+

Setup

git clone https://github.com/slashgear/team-bookmarks.git
cd team-bookmarks
pnpm install
pnpm dev

Available scripts

| Script | Description | | -------------------- | ----------------------------------------- | | pnpm dev | Start dev server | | pnpm build | Build frontend + compress + bundle server | | pnpm build:analyze | Build with bundle visualizer | | pnpm preview | Serve production build locally | | pnpm test | Run tests in watch mode | | pnpm test:run | Run tests once | | pnpm lint | Lint with Oxlint | | pnpm format | Format with oxfmt |

Build your own Docker image

docker build -t team-bookmarks .
docker run -p 3000:3000 team-bookmarks

Deployment guides

• Scaleway Serverless Containers

Contributing

See CONTRIBUTING.md.

License

MIT © Antoine Caron