research / skill

last30days

Research a topic across Reddit, X, YouTube, Hacker News, Polymarket and the web, and get one summary of the last 30 days.

Repository owner mvanhorn

Not rated yet
0 comments

Like 0 Save

Published 11 Oct 2026Listing updated 12 Oct 2026

At a glance

Purpose: Research a topic across Reddit, X, YouTube, Hacker News, Polymarket and the web, and get one summary of the last 30 days.
Runs in: Chat app with skills, Local agent with shell, MCP server
Operating systems: macOS, Linux, Windows
Needs: Runtime: Python 3.12+; Needs internet: Yes; Needs accounts or API keys beyond your agent: Not stated
Files and data: Reads files: Yes; Writes files: Yes; Sends data to other services: Yes
Tool type: Skill
Delivery: From GitHub
Version: Latest on the main branch
Access: Code on GitHub
Compatibility: According to the README, it installs on Claude Code, Grok (xAI Build CLI), Claude's web app, Claude Desktop and OpenClaw, and on Codex, Cursor, Copilot, Gemini CLI and other Agent Skills hosts through the Skills CLI. It also names Windsurf, Cline, Continue, Roo, Aider-Desk, OpenCode and Goose among 50 or more supported hosts.
External costs: The repository is MIT-licensed. According to the README, Reddit, Hacker News, Polymarket and GitHub need no key, and the arXiv and Techmeme command-line tools are free. Other sources can cost money. The README lists a ScrapeCreators key (10,000 free calls, then pay as you go), Perplexity (pay as you go), X API credits from your own X developer project, and a Brave Search key (2,000 free queries a month). Your coding agent or model provider may also charge for usage.
Permissions: The skills act through your coding agent with the access you give it. According to the README, the skill runs Python scripts that query the sources you enable, writes research files to a folder on your machine, and reads keys from environment variables, .env files or the macOS Keychain. For X, you can log in to X in a browser instead of using a key, and the README describes reading login cookies from Chromium-based browsers such as Brave, Edge, Vivaldi, Opera and Arc. The README says first-run setup installs some free command-line tools, that its install command uses -g to install for your user across all projects, that publishing briefs to a hosted page is opt-in, and that a --preflight option shows what a run would read and write without starting research. It also describes a watchlist script for scheduled runs with optional Slack or webhook delivery of new findings.
Usage terms: MIT License, as stated in the repository. See its LICENSE file for the exact terms.

Code on GitHub

mvanhorn/last30days-skill

MIT

Python64,088 stars5,590 forksLast pushed 11 Oct 2026 UTC

AI agent skill that researches any topic across Reddit, X, YouTube, HN, Polymarket, and the web - then synthesizes a grounded summary

Topics: ai-prompts ai-skill bluesky claude claude-code clawhub deep-research hackernews

Public GitHub data fetched 13 Oct 2026 UTC · 0 days old

Skill folders in this repository (1)

Folders that contain a SKILL.md file, from GitHub's file list on 13 Oct 2026 UTC.

skills/last30days

View on GitHub ↗

What it does

last30days is a skill for AI coding agents that researches a topic across many online sources and writes one summary of the last 30 days. According to the README, it searches sources such as Reddit, X, YouTube, TikTok, Hacker News, Polymarket, GitHub, Bluesky and the web in parallel. It scores the results by engagement, and an AI agent synthesizes them into one brief.

The README lists uses such as researching a business before a sales call, reading a company's job pages for hiring signals, comparing tools, and a discovery mode that ranks trending topics. It also describes saving briefs to a local folder, JSON and HTML output, an optional SQLite store with watchlist runs, and a searchable library of past research.

You run it with /last30days and a topic. The README says Reddit, Hacker News, Polymarket and GitHub work with no configuration, and that a setup wizard unlocks more sources.

Who it is for

People who want a summary of recent online activity about a business or topic, using an AI coding agent.

Example tasks

Research a business or topic before a meeting or sales call.
Read a company's job pages for hiring signals.
Compare two or more tools side by side.
Find what is trending in a field such as AI agents.
Search your saved research for a topic you covered before.

Getting started

For Claude Code, the README gives /plugin marketplace add mvanhorn/last30days-skill and then /plugin install last30days. For Codex, Cursor, Copilot, Gemini CLI and other Agent Skills hosts, it gives npx skills add mvanhorn/last30days-skill -g. It also covers Grok, a .skill upload for Claude's web app, a .mcpb bundle for Claude Desktop, OpenClaw and a manual clone. It says to run /last30days once to start the setup wizard. The README has the exact commands.

Limitations

It depends on an AI coding agent that loads the skill and writes the brief. According to the README, several sources need something from you. X needs a token, a browser login or a provider key. YouTube needs yt-dlp. TikTok, Instagram, Threads, Pinterest and LinkedIn need a ScrapeCreators key. The README lists a Brave Search key for web search. It says SKILL.md is the source of truth for current commands and setup. The README says its Claude Desktop bundle needs Python 3.12 or later on your PATH and comes for macOS and Linux x86_64 only, and that Windows support is deferred. In Claude's web app, code execution and file creation must be turned on. The README says to use one install method per machine. The repository can change at any time; check its README before you install.

Data practices

According to the README, it saves research files on your machine, by default in a Last30Days folder inside your Documents folder, and you can choose another folder. It also describes an optional SQLite store that keeps findings across runs and a generated feed of local files. The README says searching your saved library is offline. It says publishing a feed is opt-in and public by default. For the Claude Desktop bundle, the README says keys are stored in your operating system keychain. The skill queries the sources you enable. What your coding agent and its model provider receive depends on your agent.