Content Aggregator #16

	name: Content Aggregator
	on:
	workflow_dispatch:
	schedule:
	- cron: '20 8 * * *'

	jobs:
	summarize:
	runs-on: ubuntu-latest
	timeout-minutes: 120 # 2 hour timeout
	steps:
	- uses: actions/checkout@v4

	- name: Set up Python 3.10
	uses: actions/setup-python@v4
	with:
	python-version: '3.10'

	- name: Install system dependencies
	run: \|
	# TODO:check if all dependencies are indeed needed
	sudo apt-get update
	sudo apt-get install -y \
	python3-dev \
	libxml2-dev \
	libxslt-dev \
	chromium-chromedriver \
	jq

	- name: Install Python dependencies
	run: \|
	python -m pip install --upgrade pip
	pip install -r requirements.txt

	- name: Install Playwright browsers
	run: \|
	playwright install chromium
	playwright install-deps

	- name: Run summarizer
	env:
	GEMINI_API_KEY: ${{ secrets.GEMINI_API_KEY }}
	GEMINI_MODEL_SUMMARIZE: ${{ secrets.GEMINI_MODEL_SUMMARIZE }}
	GEMINI_MODEL_DATE_EXTRACT: ${{ secrets.GEMINI_MODEL_DATE_EXTRACT }}
	DEBUG: "false"
	ARTICLES_LIMIT: "500"
	run: python main.py

	- name: Upload output files
	uses: actions/upload-artifact@v4
	with:
	name: summaries
	path: \|
	outputs/*.json
	outputs/*.txt
	retention-days: 1

	- name: Create GitHub Issue
	if: success() # Only run if previous steps succeeded
	env:
	GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
	run: \|
	FILE_NAME="issue.md"
	./scripts/create_issue_body.sh $FILE_NAME
	cat $FILE_NAME

	# Create issue using GitHub CLI
	gh issue create \
	--title "Daily Content Summary $(date +'%Y-%m-%d')" \
	--body-file $FILE_NAME \
	--label automated

Provide feedback