Dedicated agent for the Daily Trades project — a trend-focused market intelligence newsletter targeting 20-30yr old males using FOMO/hindsight hooks.
1. Set up Beehiiv account + newsletter template design
2. Set up cron automation for weekday pipeline runs
3. Beehiiv API integration (auto-create draft, Josh reviews before send)
4. Add agriculture/mining RSS feeds (still missing)
5. Continue tuning prompts based on article review
prompts/filter.md, score.md, write.md (easy to iterate)scored after writing, only become used when manually marked (dashboard button or future automation)memory/MEMORY.md contains persistent context about the project state, architecture decisions, known issues, and next steps. Read this at the start of every session. Update it when you make significant changes so the next session (on any machine) has full context.
When finishing a dev session, you MUST update CLAUDE.md (this file) with:
This file is read by the EA agent on homelab to generate the daily briefing dashboard. If you don't update it, Josh sees stale project status. Do not wait until "later" — update before the session ends.
memory/MEMORY.md — agent memory (read first every session)product.md — product definition, audience, comms strategy (LOCKED IN) ← source of truthpipeline.md — pipeline architecture (v0.3)example-format.md — 5 complete example issues showing the target outputmfm-style-guide.md — Shaan Puri voice analysis for writing toneprompts/filter.md — filter prompt (supports {active_clusters})prompts/score.md — scoring prompt (supports {recent_context})prompts/write.md — writing prompt (supports {other_articles}, {recent_history})scripts/run.py — full pipeline + dashboard web serverscripts/init_db.py — database schemadata/ — SQLite database (gitignored)
python3 scripts/run.py # Full pipeline
python3 scripts/run.py --ingest # Fetch RSS feeds
python3 scripts/run.py --filter # Filter + rank new headlines
python3 scripts/run.py --decay # Apply time decay
python3 scripts/run.py --score # Score all filtered headlines
python3 scripts/run.py --write # Write top 3 articles
python3 scripts/run.py --top # Show scored headlines
python3 scripts/run.py --pool # Pool health check
python3 scripts/run.py --sources # List feeds
python3 scripts/run.py --serve # Dashboard on :8080
systemctl --user status daily-trades-dashboard # check
systemctl --user restart daily-trades-dashboard # restart after code changes