Last 30 Days

Researches a topic across Reddit, X, Hacker News, Polymarket, and the web over a 30-day window.

Updated Jun 11, 2026
One-click install
npx skills add https://github.com/aaronjmars/aeon-bd --skill last-30-days-aaronjmars
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: Last 30 Days
Source: https://github.com/aaronjmars/aeon-bd/tree/main/skills/last30
Command: npx skills add https://github.com/aaronjmars/aeon-bd --skill last-30-days-aaronjmars

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Platform-by-platform recaps of social discussion bury the actual story: the same narrative appears on Reddit, X, and Hacker News with different sentiment, and flat "top posts" lists miss that divergence. This Skill clusters cross-platform discussion into narratives and surfaces where communities disagree, which is usually the most actionable signal. ## Core Features & Use Cases - Cross-platform research: Queries Reddit (public JSON API), X (via the X.AI Grok x_search API), Hacker News (Algolia API), Polymarket/Kalshi prediction markets, and long-form web articles within a configurable lookback window. - Narrative clustering and sentiment splits: Deduplicates shared URLs, clusters items by anchor tokens, ranks narratives by cross-platform presence and divergence, and maps per-platform sentiment (bull/bear/mixed/neutral). - Snapshot deltas: Compares against a prior run stored in memory to report what is NEW, FADED, FLIPPED, or HEATING since the last snapshot. - Use Case: Ask what people are saying about a crypto project or AI model over the last month, and receive a report with a verdict, ranked narratives with direct quotes, a contrarian view, a sentiment map, prediction market odds, and open questions. ## Quick Start Research what people are saying about Solana Firedancer across Reddit, X, Hacker News, and the web over the last 30 days.

Frequently Asked Questions about Last 30 Days

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I research what people are saying about a topic on Reddit and X?▼

Provide a topic and the Skill queries Reddit's public JSON search, X via the X.AI Grok x_search API, Hacker News, and web articles over the last 30 days. It clusters results into narratives and writes a report with sentiment splits per platform.

What is narrative clustering in social media research?▼

Narrative clustering groups posts from different platforms that share anchor tokens like project names, numbers, or domains into a single story. Items with two or more overlapping anchors are merged, so the same event discussed on Reddit and HN counts as one narrative.

Does this research require an X.AI API key?▼

Yes, XAI_API_KEY is declared in requires and is the primary path for X/Twitter data via Grok's x_search tool. If the key is unset or the call fails, the X layer degrades to WebSearch while Reddit, HN, Polymarket, and web layers still run.

Can I change the 30-day lookback window or run a lighter search?▼

Yes, append --days=N to the topic to change the lookback window, or --quick for a lighter pass using at most 15 sources. Quick mode runs fewer subreddits, skips comment threads, and fetches fewer web articles.

Why does the Reddit search return empty results?▼

Reddit's anonymous JSON API caps at roughly 10 requests per minute and returns empty 200 responses without a descriptive User-Agent. The Skill backs off, retries once, and falls back to WebFetch on the same URL if the response stays empty.

What are the limitations of cross-platform sentiment analysis?▼

Sentiment is classified from the tone of the highest-engaged takes per platform, not comment averages, so it reflects dominant voices rather than full distributions. Topics visible on only one platform become standalone signals rather than ranked narratives.