mesamirh/linkding-bot repository overview

⁠Linkding Bot

A lightweight, zero-dependency Telegram bot that saves links to your Linkding⁠ instance with AI-generated tags and notes.

The bot automatically fetches website content as clean Markdown and uses your preferred LLM to intelligently tag and summarize your bookmarks.

⁠Features

• AI Auto-Tagging & Notes: Extracts context from the webpage and generates concise notes and up to 5 relevant tags.
• Multi-URL Support: Send multiple links in a single message; the bot processes them concurrently.
• Reply to Tag: Forgot a tag? Simply reply to the bot's success message with new tags (e.g., golang, tutorials) to append them to the bookmark.
• Duplicate Prevention: Checks if a URL is already in your Linkding instance before saving.
• Ultra-Lightweight: Written in Go with 0 external dependencies. Minimal Docker footprint.

⁠Prerequisites

Before installing, make sure you have the following ready:

1. Telegram Bot Token: Create a new bot and get the token from @BotFather⁠ on Telegram.
2. Linkding Instance: Your Linkding URL and API Token (found in Settings > Integrations of your Linkding instance).
3. LLM API Key: Any OpenAI-compatible API endpoint (e.g., OpenAI, a proxy, or a local LLM).

⁠Installation & Setup

You can run the bot either using Docker (recommended) or locally from source.

⁠Option A: Using Docker Compose (Recommended)

The easiest and most reliable way to run the bot is using Docker Compose.

1. Create a docker-compose.yml file:

services:
  bot:
    image: mesamirh/linkding-bot:latest
    container_name: linkding-bot
    restart: unless-stopped
    env_file:
      - .env

1. Create a .env file next to it (see Configuration⁠ below for details):

TELEGRAM_TOKEN=your_telegram_bot_token
LINKDING_URL=https://linkding.yourdomain.com
LINKDING_TOKEN=your_linkding_api_token
LLM_API_KEY=your_llm_api_key
LLM_BASE_URL=https://api.openai.com/v1
LLM_MODEL=gpt-4o
TAG_LIMIT=5
ALLOWED_CHAT_IDS=123456789,987654321

1. Start the container in the background:

docker-compose up -d

⁠Option B: Running Locally (Without Docker)

If you prefer to run it manually or for development purposes:

1. Clone the repository:

git clone https://github.com/mesamirh/linkding-bot.git
cd linkding-bot

1. Set your environment variables (copy .env.example to .env and fill it out):

cp .env.example .env
# Edit .env with your favorite editor

1. Source the environment variables and run the bot:

set -a; source .env; set +a; 
go run ./cmd/bot

⁠Configuration

The bot uses the following environment variables. All are required unless specified otherwise.

⁠Usage

1. Open Telegram and start a chat with your newly created bot by sending /start.
2. Send the bot a message containing one or more URLs.
3. The bot will automatically:
   • Fetch the webpage content.
   • Generate tags and summary notes using AI.
   • Save the bookmark directly to your Linkding instance.
4. Custom Tags: If you want to add specific custom tags later, simply reply to the bot's success message with your new tags separated by commas or spaces.

⁠Credits

• Linkding⁠ - The fantastic self-hosted bookmark manager.
• Jina AI⁠ - Their Reader API powers the clean text extraction for LLM processing.