last30days

Researches recent public discussion about a topic across Reddit, X, YouTube, and other social sources.

1|Updated Jul 2, 2026
One-click install
npx skills add https://github.com/filippolmt/skills --skill last30days-filippolmt
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: last30days
Source: https://github.com/filippolmt/skills/tree/main/skills/last30days
Command: npx skills add https://github.com/filippolmt/skills --skill last30days-filippolmt

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? Finding out what people actually said about a topic in the last 30 days requires manually searching Reddit, X, YouTube, TikTok, Hacker News, Polymarket, GitHub, and the web, then reconciling scattered posts and engagement metrics into a coherent picture. ## Core Features & Use Cases - Multi-source research engine: Pulls posts and engagement from Reddit, X, YouTube, TikTok, Hacker News, Polymarket, GitHub, Bluesky, and web search, then ranks evidence clusters by relevance. - Query-type handling: Supports general, news, comparison (A vs B), discovery (trending topics in a domain), and deep-research modes with structured synthesis templates. - Doctor health check: Diagnoses broken, rate-limited, or unconfigured sources before and after runs so missing coverage is never mistaken for silence. - Use Case: Run a comparison query like "codex vs claude code" to get a community-sentiment verdict with quoted comments, per-source coverage stats, and a saved raw research file. ## Quick Start Ask the assistant to run last30days on a topic, for example: research what people are saying about nvidia earnings reaction in the last 30 days.

Frequently Asked Questions about last30days

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

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

Invoke the skill with a topic, and the engine queries Reddit, X, YouTube, TikTok, Hacker News, Polymarket, GitHub, and web search for the last 30 days. It returns ranked evidence clusters with engagement stats and quoted community comments synthesized into a narrative report.

How do I compare two products using community sentiment?

Phrase the query with "vs" or "versus", such as "codex vs claude code". The engine runs a comparison mode that produces a quick verdict, per-entity sections, and a head-to-head analysis drawn from community discussion across all configured sources.

What API keys does multi-source social research require?

All keys are optional; the engine degrades gracefully per source. Supported keys include SCRAPECREATORS_API_KEY, X_BEARER_TOKEN, OPENAI_API_KEY, PERPLEXITY_API_KEY, BRAVE_API_KEY, and others, with SCRAPECREATORS_API_KEY as the primary one.

Why does a source show no results during a research run?

A source may be rate-limited, unauthenticated, unreachable, or simply quiet. Run the built-in doctor health check to diagnose configuration before a run, or doctor --postmortem afterward to see each source's actual status from the last run's cache.

Can I discover trending topics instead of researching one topic?

Yes, the --discover flag sweeps river listings in a domain, ranks accelerating topics, and runs a full research pass on each survivor. A bare --discover runs global trending across every feed's hot list.