Research what people actually say about a topic in the last 30 days - pulling posts and engagement from Reddit, X, YouTube, TikTok, Hacker News, GitHub, Polymarket, and the web, with citations.
---
name: Last 30 Days
description: Research what people actually say about any topic in the last 30 days, pulling posts and engagement from Reddit, X, YouTube, TikTok, Hacker News, Polymarket, GitHub, and the web. Use for recency-focused sentiment, trend, and reaction research with citations.
---
Gather and synthesize what real people said about a topic recently across many social and web sources, with engagement signals and citations.
## How to use
1. Invoke with a focused query, e.g. "last30days nvidia earnings reaction" or "last30days what users want in react".
2. The skill runs Node/Python scripts to pull recent posts and engagement from Reddit, X, YouTube, TikTok, Hacker News, Polymarket, GitHub, Bluesky, and the web.
3. Configure optional API keys via env (e.g. SCRAPECREATORS_API_KEY as the primary, plus optional X/Bluesky/search provider keys) to widen and deepen coverage.
4. It filters to the recent window, weighs engagement, and synthesizes themes - what people praise, complain about, and want.
5. Output is a cited summary of recent sentiment and trends; check `--help` for flags like competitor comparison.
Requires `node` and `python3`; many sources work without keys but coverage improves with them.
Full skill & source: https://github.com/mvanhorn/last30days-skill/tree/main/skills/last30days