Automated Telegram bot to scrape and publish job offers across different categories with separated scraping/sending architecture.

Use Jobspy <3

# Initialize database tables (safe, preserves existing data)
make db-init

# Migrate existing JSON data (if any)
make db-migrate

# Create backup before changes
make db-backup

TELEGRAM_BOT_TOKEN=your_token_from_botfather
TELEGRAM_GROUP_ID=your_group_id
DATABASE_URL=your_database_url
SKIP_INIT_JOB=1 # Skip initial job setup (set to 0 to run init job)
DOCKER_ENV=true # true if running in Docker
FORCE_DOCKER_MODE=0  
RESULTS_WANTED=10   # Number of job offers to scrape per category               
SCRAPE_DELAY_MIN=2.0 # Minimum delay between scrapes (in seconds)
SCRAPE_DELAY_MAX=5.0 # Maximum delay between scrapes (in seconds)

Note: All categories with their topic IDs and schedules are managed through categories.yml file.

The bot uses a separated architecture with independent scraping and sending operations:

• Scraping: Distributed across 24 hours (max 3 categories per hour)
• Sending: Runs 1 hour after scraping for each category
• Even distribution: No clustering, optimal resource usage
• Independent operations: Scraping failures don't block sending

00:00 - Scrape: aeronautique, immobilier, vente
01:00 - Send: aeronautique, immobilier, vente | Scrape: agent_assurance, industrie  
02:00 - Send: agent_assurance, industrie | Scrape: agriculture, ingénieur_electronique
...

View complete schedule: python -m france_chomage utils info

See: categories.yml

• Complete Documentation - All guides and references
• Development Guide - Setup for developers
• Database Setup - Database configuration
• Safe Deployment Guide - How to add new job categories safely

For development instructions and adding new job categories, see Development Guide.

Shield:

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.