ai-cohort-scoreboard-marketing

name	ai-cohort-scoreboard-marketing
category	content
description	Generate daily Reddit and X/Twitter marketing content from the AI Cohort scoreboard, with automatic fallback to the most recent available data when today's file is missing.

AI Cohort Scoreboard Marketing Content Generation

Purpose: Generate daily Reddit and X/Twitter marketing content from the AI Cohort scoreboard, with automatic fallback to the most recent available data when today's file is missing.

Trigger Conditions:

• Daily cron job scheduled after 8:30am ET (Eastern Time)
• Manual invocation when marketing content needs update
• Error recovery: When generate_marketing_content.py fails with "Today's scoreboard not found"

Workflow Steps:

1. Check for Today's Scoreboard

   eastern = tz.gettz('America/New_York')
   today = datetime.now(eastern).strftime('%Y-%m-%d')
   # VAULT_ROOT should be: '/opt/data/home/obsidian-vault/FACorreia'
   scoreboard_path = f"{VAULT_ROOT}/{today} — AI Scoreboard.md"
   

   • The scoreboard file is saved directly in the vault root by the scoreboard generator script
   • Ensure VAULT_ROOT matches the scoreboard script's VAULT_ROOT
   • If today's scoreboard exists → use it
   • If missing → find latest available scoreboard (by date descending)
   • Never skip generation; always produce output

2. Generate Marketing Content

   • Parse the scoreboard markdown into sections
   • Format Reddit post with: Quick Take, Winners Cohort, Live Scoreboard, Insider Buys, Top Signals, Source attribution
   • Format X/Twitter thread with 5 tweets covering: performance gap, scoreboard table, top signals, insider highlights, cohort attribution
   • Write files to /opt/data/home/.hermes/output/marketing/
     • reddit_ai_scoreboard_{date}.md
     • x_thread_ai_scoreboard_{date}.md

3. Report Status

   • Success with today's date when using current data
   • Success with fallback notice when using older data, including which date was used
   • Include error context if generation failed completely

Pitfalls & Gotchas:

• ⚠️ Path Consistency: All scripts must use the same VAULT_ROOT path. The correct path is /opt/data/home/obsidian-vault/FACorreia (without the .hermes prefix). Inconsistent paths cause "scoreboard not found" errors even when the file exists.
• ⚠️ Missing historical data: If NO scoreboard files exist, the task cannot proceed. Verify the cohort directory contains at least one scoreboard file.
• ⚠️ Date format consistency: Scoreboard files use YYYY-MM-DD format with en dash separator. Ensure path construction matches exactly.
• ⚠️ Output date: Always stamp output files with today's date, even when using fallback data. This maintains cron job consistency.

Verification:

# Check output files exist
ls -lh /opt/data/home/.hermes/output/marketing/ | grep "reddit_ai_scoreboard_$(date +%Y-%m-%d)"

# Validate content structure
head -20 /opt/data/home/.hermes/output/marketing/reddit_ai_scoreboard_$(date +%Y-%m-%d).md

References:

• references/error-transcript-2026-05-06.md - Full error context and resolution steps
• references/scoreboard-directory-structure.md - Expected vault organization

Related Skills:

• content-digests - Periodic content digest patterns
• external-content-ingesting - Content ingestion workflows
• x-post-ingestor - Social media content ingestion

Last Updated: 2026-05-06