| name | trend-radar |
| description | Find and score current AI/social content signals for Social Pulse and Clip Mine. |
| category | genx-marketing |
| version | 1.1.0 |
Trend Radar
Goal: identify current, source-backed content opportunities. No fabricated sources.
Inputs
- topic or scan mode
- audience: B2C, B2B, Both
- destination: social, blog, carousel, clip mining
- time window: default last 24-72h; Signal Scan last 6h
Source Order
Fast Scan: YouTube, Reddit, Tech News, Product Hunt, approved/public X.
Deep Social Scan: Instagram/TikTok public surfaces, then X/Reddit/Tech News.
Signal Scan: X/public, Hacker News, GitHub Trending, ArXiv, Reddit if public.
Manual Assisted Scan: user-provided URL/screenshot/caption/handle.
Respect access limits. If blocked, mark access_status: blocked and move on.
Normalize
Every signal must map to:
platform, creator, url, published_date, views, likes, comments,
caption, hook, format_type, topic, category,
why_it_is_trending, can_genx_adapt_this, confidence, access_status
Use goblin_recon.tools.social_intake before scoring when feasible.
Score
Score 0-100:
- velocity/recency: 30
- source proof: 20
- audience fit: 20
- format usefulness: 15
- differentiation: 15
Advance threshold: 60. If below, shelve.
Categories:
- Latest AI News
- Controversial/Polarizing
- Upgrade/Democratization
- Analytical/Deep-dive
Output
Start with ## Decision.
For each opportunity include:
- recommendation: approve / monitor / shelve
- category
- source URLs and publication dates
- hook and format
- creator/source driving it
- effort
- confidence
- vault check
- fallback angle
- AI search potential
- next step
Save Social Pulse reports to vault/reports/YYYY-MM-DD-social-pulse.md.