# Milan News Briefing > Generate a mature daily Milan news brief in Chinese using Python with weather, 20-day strike check, Italian media headlines, world news, AI news, and Milan events; persist daily outputs with dedup so next-day reports avoid repeats; manage RSS/API source pipelines. Use when the user asks to create, run, debug, or extend the Milan briefing workflow, especially for OpenClaw skill-style automation. - Skill: `vulcanusalex/milan-news-briefing` (Agent Skill, multi-file: 10 files) - Install (CLI): `npx skillmds@latest add vulcanusalex/milan-news-briefing` - Raw SKILL.md: https://api.skillmd.com/api/skills/vulcanusalex/milan-news-briefing/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Integrations & APIs - Author: vulcanusalex (https://skillmd.com/u/vulcanusalex) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/vulcanusalex/milan-news-briefing --- Run the briefing pipeline and manage sources with deterministic scripts. ## Execute Daily Brief 1. Ensure dependencies are installed: `python3 -m venv .venv && source .venv/bin/activate && pip install -r requirements.txt` 2. Generate report: `python -m src.news_briefing.main` 3. Use dry run for debugging: `python -m src.news_briefing.main --dry-run` 4. Default report date follows config timezone (`Europe/Rome`) when `--date` is omitted. 5. Agent can choose layout/order at call time: `python skills/milan-news-briefing/scripts/run_briefing.py --layout editorial --section-order weather,strikes,ai_news,world_news,italian_news,milan_events` 6. Agent can request JSON output for post-processing: `python skills/milan-news-briefing/scripts/run_briefing.py --output-format json` ## Manage Source Pipelines Use `scripts/manage_sources.py` in this skill folder. 1. List configured sources: `python skills/milan-news-briefing/scripts/manage_sources.py list` `python skills/milan-news-briefing/scripts/manage_sources.py --json list` 2. Add RSS source: `python skills/milan-news-briefing/scripts/manage_sources.py add --section ai_news --name "Example AI" --type rss --url "https://example.com/rss.xml"` 3. Add JSON API source with parser: `python skills/milan-news-briefing/scripts/manage_sources.py add --section strikes --name "Strike API" --type json --url "https://example.com/strikes" --parser italy_transport_strikes_v1` 4. Add HTML source (for parser using webpage table): `python skills/milan-news-briefing/scripts/manage_sources.py add --section strikes --name "Strike HTML" --type html --url "https://example.com/strikes-page" --parser italy_mit_strikes_html_v1` 4. Remove source by exact name: `python skills/milan-news-briefing/scripts/manage_sources.py remove --section world_news --name "BBC World"` ## Daily Ops Use `scripts/daily_ops.py` for daily automation. 1. Run once with precheck + auto degrade + retry: `python skills/milan-news-briefing/scripts/daily_ops.py --auto-degrade --max-retries 2 --retry-delay 120` 2. Run once and send fail alert: `python skills/milan-news-briefing/scripts/daily_ops.py --auto-degrade --alert-webhook "https://example.com/webhook"` 3. Print cron line (7:00 daily): `python skills/milan-news-briefing/scripts/daily_ops.py --print-cron --cron-hour 7 --cron-minute 0` 4. Optional success alert: `python skills/milan-news-briefing/scripts/daily_ops.py --alert-webhook "https://example.com/webhook" --alert-success` 5. Skip precheck explicitly (not recommended): `python skills/milan-news-briefing/scripts/daily_ops.py --skip-precheck` ## Cron Scheduling Use `scripts/manage_cron.py` to manage one tagged cron entry. 1. Install/replace 07:00 daily run: `python skills/milan-news-briefing/scripts/manage_cron.py install --hour 7 --minute 0` `python skills/milan-news-briefing/scripts/manage_cron.py install --hour 7 --minute 0 --timezone Europe/Rome` `python skills/milan-news-briefing/scripts/manage_cron.py install --hour 7 --minute 0 --timezone Europe/Rome --layout editorial --output-format markdown` 2. Install with webhook: `python skills/milan-news-briefing/scripts/manage_cron.py install --hour 7 --minute 0 --webhook "https://example.com/webhook"` 3. Show current managed entry: `python skills/milan-news-briefing/scripts/manage_cron.py show` `python skills/milan-news-briefing/scripts/manage_cron.py --json show` 4. Remove managed entry: `python skills/milan-news-briefing/scripts/manage_cron.py remove` ## Health Check And Validation 1. Check all configured sources: `python skills/milan-news-briefing/scripts/check_feeds.py --timeout 12 --write-report output/logs/feed-health.json` 2. Validate full stack locally: `python skills/milan-news-briefing/scripts/validate_stack.py --skip-network` 3. Validate with network checks: `python skills/milan-news-briefing/scripts/validate_stack.py` ## Operate Safely 1. Keep configuration in `config/sources.yaml` as source-of-truth. 2. Keep strike source URL non-empty if real strike monitoring is required. 3. Keep per-section `count` at 5 for stable daily format unless user asks otherwise. 4. Preserve SQLite database `data/briefing.db`; do not clear it unless user asks to reset dedup history. 5. Check `output/logs/` after failures and use webhook payload `tail` for quick triage. 6. Use `--auto-degrade` in production cron to bypass temporary source outages. 7. Prefer `--json` flags on management scripts when an agent needs machine-readable state. ## Extend Parsers 1. Add parser logic in `src/news_briefing/parse.py`. 2. Register parser in `src/news_briefing/pipeline.py` parser maps. 3. Reference parser key in `config/sources.yaml` source item. Read `references/source_schema.md` before adding JSON-based sources.