What is rising right now
The posts rising fastest on Hacker News, GitHub, Product Hunt and more, with the keywords that are taking off. Read every 2 hours. Pick a niche for today's content ideas.
Rising posts
Loading the latest read...
Deals and freebies
Free games and keys, free memberships and limited-time offers that are still open, newest first. An offer with an end date shows it. Each deal is deleted 30 days after it was posted.
Loading the open deals...
- GamerPower. Active free game, DLC and beta key giveaways (Epic, Steam, GOG and more) from the GamerPower API, with the end date when it has one.
- V2EX. Latest topics of the V2EX free-giveaway node (免费赠送) through the public V2EX API.
- NodeSeek. Posts from the public NodeSeek RSS feed whose title offers something free or discounted; buy-and-sell posts are skipped.
Game giveaways data by GamerPower.com. Forum offers are leads posted by other people: check each one on its page before you sign up.
How the ranking works
A post is rising now when it was published in the window you pick and was still on its source's list in the last 6 hours.
Gain per hour. Each post keeps its last two readings. When they are 30 minutes or more apart, the gain per hour is the change between them. A post read once is measured from its publish time.
Heat. Each post is ranked against the other posts of its own source, from 0 to 100, so points, stars and views never mix.
Rising keywords. Words from titles whose share grew most in the last 3 days against the 3 days before. A word needs 3 posts. New means the earlier days never used it.
The sources
- Hacker News. Front page and stories with 20+ points from the last 3 days, through the Algolia HN Search API. Counted in points.
- DEV Community. Top articles of the last 3 days from the public Forem API. Counted in reactions.
- GitHub. Repositories created in the last 7 days with 20+ stars, through the GitHub search API. Counted in stars.
- Product Hunt. The public Product Hunt Atom feed, ranked by its own order. Counted in list order, since it publishes no count.
- Lobsters. The public hottest.json feed. Counted in points.
- YouTube. The most popular videos in the US through the YouTube Data API (only when a key is set). Counted in views.
For agents
Connect the MCP server https://trending.openkrill.app/mcp. Call trending with a topic and a window, trending_digest with a niche for this page's daily digest, or trending_deals with search words for open deals. Or read it over plain HTTP:
curl -s 'https://tools.openkrill.app/trending/read?topic=ai&window=6h'
The answer holds the posts with link, source, publish time, count, gain per hour and heat, the rising keywords, the state of each source and why Reddit and X are not read.
Frequently asked questions
Where do the posts come from?
From Hacker News, DEV Community, GitHub, Product Hunt, Lobsters, YouTube. Each one is read through its official API or the public feed its publisher offers, every 2 hours. Nothing needs a login and no web page is scraped. YouTube is read only when an API key is set.
Why are Reddit and X missing?
Reddit is not read: its Data API terms need a separate agreement for commercial use. X is not read: the X developer terms and xAI's usage policy do not allow passing posts or Grok output on through an API.
What does the heat score mean?
Sources count in different units: points, reactions, stars, views. So each post is ranked only against the other posts of its own source, and heat is its place there from 0 to 100. A heat of 90 on GitHub and 90 on Hacker News both mean near the top of that source.
Why do the keywords show no percent yet?
A keyword's rise compares its share of titles in the last 3 days with the 3 days before. Until the store holds 6 days of posts the page shows post counts only, so no percent is ever a guess.
Is anything about me stored?
No. The page reads a store of public post titles and links. It keeps no post text, no author names and nothing about who asks.