메뉴
Configuration, maintenance, and troubleshooting of AI Cohort scoreboard scripts including vault path setup and script updates.
| name | AI Cohort Scoreboard Maintenance |
| description | Configuration, maintenance, and troubleshooting of AI Cohort scoreboard scripts including vault path setup and script updates. |
| triggers | [] |
This skill covers the configuration, maintenance, and troubleshooting of the AI Cohort scoreboard scripts. It includes setting up the vault path, updating scripts, and verifying functionality.
The AI Cohort scoreboard consists of three main Python scripts that generate daily, weekly, and marketing content from the AI Cohort data:
ai_scoreboard.py - Generates the daily scoreboardgenerate_marketing_content.py - Creates Reddit and X/Twitter threadsweekly_review.py - Produces the enhanced weekly reviewThese scripts write their output to the Obsidian vault in the Wiki/Finance/AI Cohort/ directory.
The scripts use a VAULT_ROOT constant to define the vault location. This should be set to:
'/opt/data/home/.hermes/obsidian-vault/FACorreia'
Steps to configure:
Create symlink (if not already present):
mkdir -p ~/.hermes/obsidian-vault
ln -s /opt/data/obsidian-vault/FACorreia ~/.hermes/obsidian-vault/FACorreia
Create required directories:
mkdir -p ~/.hermes/obsidian-vault/FACorreia/Wiki/Finance/AI\ Cohort/Enriched
To update any of the AI Cohort scripts to use the new vault path:
Add VAULT_ROOT constant after imports (typically after the last import statement or after a Config header)
Replace hardcoded paths:
ai_scoreboard.py: Replace vault_dir = '/opt/data/obsidian-vault/FACorreia/Wiki/Finance/AI Cohort' with vault_dir = VAULT_ROOTgenerate_marketing_content.py: Replace path = f'/opt/data/obsidian-vault/FACorreia/Wiki/Finance/AI Cohort/{today} — AI Scoreboard.md' with path = f'{VAULT_ROOT}/{today} — AI Scoreboard.md'weekly_review.py: Replace p = f'/opt/data/obsidian-vault/FACorreia/Wiki/Finance/AI Cohort/{date_str} — AI Scoreboard.md' with p = f'{VAULT_ROOT}/{date_str} — AI Scoreboard.md'If file updates result in corruption (duplicate lines, syntax errors), restore from backup:
cp /opt/data/home/.hermes/scripts/ai-scoreboard/ai_scoreboard.py.backup /opt/data/home/.hermes/scripts/ai-scoreboard/ai_scoreboard.py
Then use a clean Python script to apply updates (see full skill for details).
To verify vault access, create a test file:
import os
import sys
sys.path.insert(0, '/opt/data/home/.local/lib/python3.13/site_packages')
test_path = os.path.join(os.path.expanduser('~'), '.hermes', 'obsidian-vault', 'FACorreia', 'Wiki', 'Finance', 'AI Cohort', 'test.txt')
with open(test_path, 'w') as f:
f.write("Test successful")
os.remove(test_path)
/opt/data/home/.hermes/scripts/ai-scoreboard/~/.hermes/obsidian-vault/FACorreia//opt/data/scripts/ai-scoreboard/Scan vault raw/ folder for hyperlinks embedded in markdown notes; POST each unseen URL to LuminaVault server /v1/capture/safari so Hermes ingests + memorizes. Usage: /kb-vacuum [folder=raw/]
Compile all uncompiled raw/ content into the wiki. Writes source summaries, creates/updates concept articles with Obsidian backlinks, and updates the index. Run after /kb-ingest to process new content.
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.
Build and deliver periodic content digests (news, stock, entertainment) to multiple platforms: save to vault Raw/ and print full markdown to stdout for cron-based delivery.
Capture external content (X/Twitter articles, web posts) and ingest into Obsidian vault Raw/ with automatic theme detection, summarization, and structured frontmatter. Handles X/fixupx links via multi-strategy extraction (direct fetch → r.jina.ai → nitter fallback).
Weekly founder content research and production engine that transforms trending news into ready-to-use content assets.