Moodle Mate

What It Does

• Polls Moodle for new notifications and forwards them to providers (Discord, Pushbullet, NTFY)
• Optional AI summarization via OpenAI‑compatible endpoints (configurable model/endpoint)
• Lightweight (~50 MB RAM), suitable for 24/7 operation or low-end hardware

Quick Start

1. Clone and install:

git clone https://github.com/EvickaStudio/Moodle-Mate
cd Moodle-Mate
python -m venv venv && source venv/bin/activate
pip install -r requirements.txt

1. Generate config:

python main.py --gen-config

1. Run:

python main.py

Docker alternative:

docker compose build
docker compose up -d

Configuration (config.ini)

• [moodle]: url, username, password
• [ai]: enabled, api_key, model, temperature, max_tokens, system_prompt, endpoint
• [discord]: enabled, webhook_url, bot_name, thumbnail_url
• [pushbullet]: enabled, api_key
• [notification]: fetch_interval, max_retries

Use --gen-config to generate and validate this file interactively.

Architecture Overview

• CLI entrypoint (main.py): parses args, generates config, initializes services, runs main loop
• Config loader: singleton that reads config.ini into typed sections
• Moodle API wrapper: handles login, token/session, and REST calls
• Notification pipeline: fetch newest notification → convert HTML→Markdown → optional LLM TL;DR → send to providers
• Providers: Discord (webhook embed), Pushbullet, with room for custom providers

Processing Workflow

1. Fetch newest Moodle popup notification
2. Extract subject and HTML message; convert to Markdown
3. If AI enabled, call LLM to produce a concise summary
4. Send subject/message/optional summary to each enabled provider

Notes

• Supports custom OpenAI‑compatible endpoints (e.g., self‑hosted) for summaries
• Backoff and retries are applied in the main loop; keep fetch_interval reasonable

Links

• Repository: github.com/EvickaStudio/Moodle-Mate
• See README for provider setup and advanced configuration