A multi-platform Telegram bot that automatically detects media links in private messages and group chats, extracts the content without watermarks, and sends it straight back to the conversation.
- Broad platform support: Works across dozens of popular social networks, video hostings, and image boards — designed to easily scale to hundreds of sources.
- All media formats: Fetches videos, separate audio tracks, single images, and multi-photo carousels (delivered as clean Telegram media albums).
- Built for group chats: Silently spots links in the flow of conversation and replies with the media without cluttering the chat.
- High-quality & large files: Handles heavy, high-resolution videos seamlessly without aggressive compression or tight size limits.
- Java 21 LTS or later.
- Installed CLI tools (for local/standalone runs):
ffmpeg,ffprobepython3,python3-pipyt-dlp(pip install yt-dlp)gallery-dl(pip install gallery-dl)curl_cffi(pip install curl_cffi)quickjs(qjs)
- Docker and Docker Compose (optional, for containerized deployment).
Copy the example configuration file:
cp .env.example .envConfigure variables in .env:
TELEGRAM_BOT_TOKEN: Bot token from @BotFather.TELEGRAM_API_ID&TELEGRAM_API_HASH: Obtained from my.telegram.org (required for the local Bot API server).ALLOWED_TELEGRAM_USERS: Comma-separated Telegram User IDs permitted to execute commands.APP_ENV: Environment profile (devorprod).
Launches both the local Telegram Bot API Server and the bot:
# Build images and start services in the background
docker compose up -d --build
# View application logs
docker compose logs -f bot
# Stop services
docker compose down- Build the standalone package with Maven Wrapper:
./mvnw clean package
- Run the executable Fat JAR:
(The application automatically reads configuration from
java -jar target/jsaver.jar
.envin the current working directory).
The project is decoupled from x86-specific containers and can run natively in Android's Termux environment:
- Install prerequisites in Termux:
pkg update && pkg upgrade pkg install openjdk-21 ffmpeg python quickjs pip install yt-dlp gallery-dl curl_cffi - Clone the repository and build the JAR (or transfer a pre-built
jsaver.jar):./mvnw clean package -DskipTests
- Run the bot:
java -jar target/jsaver.jar
# Run unit tests (JUnit 5 + Mockito + AssertJ)
./mvnw clean test
# Complete build, testing, and Fat JAR packaging
./mvnw clean verify
# Quick packaging skipping test execution
./mvnw package -DskipTests