Credit: This skill is based on last30days by @mvanhorn. The original skill researches topics across Reddit, X, YouTube, and web. This version adds Gemini synthesis for briefings and prompts.

Original skill: github.com/mvanhorn/last30days-skill

last30days v2.1

Research any topic across X (Twitter), YouTube, and web. Find what's actually being discussed, recommended, and debated right now.

Setup

# Environment (should already be set)
export AUTH_TOKEN=your_x_auth_token
export CT0=your_x_ct0_token  
export BRAVE_API_KEY=your_brave_key

# Config
mkdir -p ~/.config/last30days
cat > ~/.config/last30days/.env << 'EOF'
BRAVE_API_KEY=your_key_here
EOF

Usage

# Quick research (faster, fewer sources)
python3 {baseDir}/scripts/last30days.py "AI agents" --quick

# Full research
python3 {baseDir}/scripts/last30days.py "AI agents" 

# Output formats
python3 {baseDir}/scripts/last30days.py "topic" --emit=json    # JSON for parsing
python3 {baseDir}/scripts/last30days.py "topic" --emit=compact  # Human readable
python3 {baseDir}/scripts/last30days.py "topic" --emit=md       # Full report

Output for AI Synthesis

The --emit=json flag outputs structured JSON that can be fed to Gemini for:

• Expert briefing generation
• Copy-paste ready prompts
• Trend analysis

Sources

Synthesis

This skill researches and returns raw data. For AI-generated briefings and prompts, pipe the JSON output to Gemini:

python3 {baseDir}/scripts/last30days.py "topic" --quick --emit=json | python3 -c "
import json, sys, os
import urllib.request, urllib.parse

data = json.load(sys.stdin)
prompt = f'Synthesize this research into an expert briefing and 3 copy-paste prompts:\\n{json.dumps(data)}'

body = json.dumps({
    'contents': [{'parts': [{'text': prompt}]}],
    'generationConfig': {'temperature': 0.7, 'maxOutputTokens': 2048}
})

req = urllib.request.Request(
    'https://generativelanguage.googleapis.com/v1beta/models/gemini-2.0-flash:generateContent?key=' + os.environ.get('GEMINI_API_KEY'),
    data=body.encode(),
    headers={'Content-Type': 'application/json'}
)
print(json.load(urllib.request.urlopen(req))['candidates'][0]['content']['parts'][0]['text'])
"

Attribution

• Original Author: Mike Van Horn (mvanhorn)
• Original Repository: github.com/mvanhorn/last30days-skill
• License: MIT (per original)
• Contributors: Thanks to @steipete for yt-dlp + summarize inspiration