What is last30days-skill
last30days-skill is an open-source AI Agent research skill that lets Agents automatically scrape real discussions from major overseas social platforms over the past 30 days and synthesize a report with citations. The tool deeply crawls Reddit comments, highly upvoted X replies, full YouTube transcripts, TikTok engagement, Polymarket odds, etc., ranking by real human participation. last30days-skill supports intelligent topic parsing, cross-source story clustering, automatic competitor comparison, and shareable HTML briefings, helping users quickly grasp the real community consensus and latest developments on any topic.
Main features of last30days-skill
-
Cross-platform deep aggregation: Searches in parallel across more than ten platforms including Reddit, X, YouTube, TikTok, HN, Polymarket, GitHub, and Bluesky, scraping comments, replies, full transcripts, and odds data.
-
Real popularity ranking: Scores by human participation signals such as upvotes, likes, views, and Polymarket odds.
-
Intelligent pre-research: After a topic is entered, it automatically parses related entities, such as X accounts, GitHub repositories, subreddits, and TikTok hashtags, then searches precisely.
-
Cross-source clustering and merging: When the same story appears on different platforms, it is automatically merged into one entry to avoid duplication.
-
Shareable HTML briefing: Generates a self-contained dark-mode HTML file that can be placed directly into Slack, email, or Notion.
-
GitHub personal mode: For people-related topics, it automatically switches to author-dimension queries, showing PR merge rate, star count, and release activity.
-
ELI5 mode: One-click rewriting of complex reports into an easy-to-understand plain language version.
-
Automatic competitor comparison: Supports
--competitorsparameter to automatically discover and compare leading competitors in parallel.
Follow on WeChat and reply “open source” to join theAI Open Source Project Communication Group
How to use last30days-skill
-
Install the skill: Claude Code users run
/plugin marketplace add mvanhorn/last30days-skilland/plugin install last30days; other hosts runnpx skills add mvanhorn/last30days-skill -g。 -
First run: Enter
/last30days <话题>, and the four zero-configuration sources Reddit, HN, Polymarket, and GitHub work immediately. -
Unlock more sources: Follow the prompts to configure X browser login, install yt-dlp, or set the ScrapeCreators/Perplexity API Key to enable YouTube, TikTok, Instagram, etc.
-
Generate a briefing: Directly ask
/last30days <话题>, give me a shareable HTML briefto get an offline-usable HTML report. -
Deep follow-up: After the briefing is generated, continue asking questions in the same session, and the AI answers based on the full context already scraped.
Core advantages of last30days-skill
-
Depth over breadth: Supports scraping comments, replies, and full transcripts, where high-quality information is often buried in the comments.
-
Real signal driven: Rankings are decided by humans voting with likes and real money, not by search engine optimization.
-
Multi-platform bridging: Connects more than ten mutually independent platform walled gardens, providing a global perspective at once.
-
Intelligent entity parsing: The v3 engine understands the topic before searching and finds content that v2 cannot discover.
-
Privacy and Open Source: Open source under the MIT License, no tracking or analytics, and all research files are kept on the local machine.
Project address for last30days-skill
- GitHub Repository:https://github.com/mvanhorn/last30days-skill
Comparison of last30days-skill with similar competitors
| Comparison Dimension | last30days-skill | Perplexity |
|---|---|---|
| Product Positioning | An open-source AI Agent research skill designed specifically for aggregating real community discussions across multiple social platforms | An AI search engine focused on real-time web search and Q&A with cited sources |
| Coverage | More than a dozen social platforms, including Reddit, X, YouTube, TikTok, HN, Polymarket, GitHub, Bluesky, etc. | Web pages, news, academic papers, Reddit (partially partnered), YouTube |
| Content Depth | Captures deep data such as comments, highly upvoted replies, full video transcripts, prediction market odds, PR merge rates, etc. | Web page summaries, partial Reddit content, academic abstracts, video descriptions |
| Ranking Mechanism | Scored by real human engagement (upvotes / likes / views / odds / stars), anti-SEO | Model ranking based on relevance and credibility algorithms |
| Real-Time Capability | Focuses on community activity over the past 30 days and continuously updates based on real popularity | Real-time web indexing covering the latest news and events |
Use cases for last30days-skill
-
Pre-meeting person research: Enter the target person's name to automatically aggregate their posts, code commits, and community discussions on X, GitHub, Reddit, and other platforms over the past 30 days, generating a dynamic profile more authentic than LinkedIn.
-
Learning new technical concepts: Enter emerging concepts such as "loop engineering" to obtain first-hand community discussions, diverging opinions, and best practices, avoiding reliance solely on second-hand reposted information.
-
Product selection comparison: Enter queries such as "OpenClaw vs Hermes" to generate multi-dimensional comparison reports based on real user feedback, GitHub star trends, and community controversies.
-
Hot event tracking: Enter topics such as "Kanye West new album" or international conflicts to aggregate news, prediction market odds, in-depth YouTube analysis, and social platform reactions for a complete picture of developments.
-
Consumption and travel decisions: Enter items such as "Universal Epic Universe" or electronics to get the latest community feedback on queue times, repair issues, real experiences, and value-for-money discussions.