mvanhorn/last30days-skill

Overview

This skill researches any topic from the last 30 days across Reddit, X, YouTube, TikTok, Instagram, Hacker News, Polymarket, and the web and returns a grounded synthesis with copy-paste-ready prompts. Before starting research it confirms parsed intent to the user by printing the following block: I'll research {TOPIC} across Reddit, X, TikTok, and the web to find what's been discussed in the last 30 days. Parsed intent:

• TOPIC = {TOPIC}
• TARGET_TOOL = {TARGET_TOOL or "unknown"}
• QUERY_TYPE = {QUERY_TYPE} Research typically takes 2-8 minutes (niche topics take longer). Starting now.

How this skill works

I parse your request for topic, target tool (if any), and query type (PROMPTING, RECOMMENDATIONS, NEWS, or GENERAL) and display that parsed intent before running searches. Then the skill runs a foreground research script that fetches posts, transcripts, captions, and market odds from the listed platforms, supplements findings with focused web searches, and synthesizes cross-platform signals into actionable insights and ready-to-use prompts.

When to use it

• You need a concise, evidence-backed update on what people are saying about a topic in the last 30 days.
• You want copy-paste prompts tailored to a specific tool (e.g., ChatGPT, Midjourney) based on current community practice.
• You need a list of concrete recommendations or product names trending now.
• You want to surface prediction-market signals (Polymarket) alongside social buzz.
• You want cross-platform verification of a claim or breaking story.

Best practices

• Provide a clear topic phrase; include the target tool if you want prompts formatted for that tool.
• Avoid asking about the tool before research—if no tool is provided, research runs first and the tool is asked about afterward.
• Use --days=N or --quick/--deep flags when you need narrower/faster or broader/deeper coverage.
• Expect 2–8 minutes for balanced runs; allow up to 5 minutes for the script to complete.
• Rely on cross-platform signals and Polymarket odds as higher-weight evidence in syntheses.

Example use cases

• ‘best open-source LLMs last30’ → get a ranked list of specific projects with community pros/cons.
• ‘image-generation prompts for Midjourney last30’ → receive copy-paste prompts tuned from recent posts.
• ‘what’s new with Nano Banana Pro last30’ → news-style summary and any official X handle findings.
• ‘crypto memecoin sentiment last7 --quick’ → rapid social sentiment, top tweets, and Polymarket snapshots.
• ‘Claude Code skills prompts last30’ → curated prompting techniques and ready prompts for Claude Code.