Hacker News Mention Monitor: Keyword, Brand, Domain & User Alerts

Hacker News Mention Monitor watches Hacker News for your keywords and returns every new story and comment that mentions them. Each hit links back to the original Hacker News item, so you can read it in context and reply.

Use it to find out when HN talks about your brand, product, competitors, your domain, or a topic you care about: launch feedback, pain-point threads, "what do you use for X?" questions, and links to your site.

$0.50 per 1,000 mentions. No login, API key or proxy needed. Monitor runs that find nothing new cost only the Actor start fee ($0.00005).

Why this Actor

• Comments, not just titles. Most HN discussion happens in comments. Every comment is checked on its own text, so you don't get every comment under a story just because the story title matched.
• No fuzzy junk. The HN search engine also returns prefix and typo matches (search "Acme" and you get "Acmex"). Every hit is re-checked locally with a whole-word match, so rows really mention your term.
• New-only monitor mode. Turn on Monitor mode, schedule it hourly, and each run returns only mentions you haven't seen. Already-seen hits are not charged.
• No 1,000-result ceiling. The search API stops paging at hit 1,000. This Actor splits the time range into smaller windows until each one holds fewer than 1,000 hits, so busy keywords and long look-backs come back complete.
• Domain watching. Track every story that links to yourdomain.com (or any subdomain) and every story or comment that pastes a link to it.
• Built on HN's official public search API (the free, keyless Algolia HN Search API behind hn.algolia.com). No HTML scraping, no login, no ban risk.

Need bulk HN data instead (full comment threads, front page, "Who is hiring?" jobs parsed into salary and location)? Use our Hacker News Scraper.

Input

Field	What it does
keywords	Brands, product names, topics or phrases. Case-insensitive whole-word match in the story title, story text, story URL, or the comment's own text. open source also matches open-source.
domains	e.g. example.com. Stories linking to the domain or a subdomain, plus stories and comments containing a link to it.
authors	HN usernames. Returns everything those users post (a founder, competitor or your own team). To find people mentioning a username, put it in keywords.
includeStories / includeComments	Search stories, comments, or both (default both).
lookbackDays	How far back to search (default 7; decimals allowed, 0.25 = 6 hours). Used on normal runs and the first monitor run.
since	Optional start date (YYYY-MM-DD, ISO time or unix seconds). Overrides lookbackDays.
minPoints	Only stories at or above this score. Comments have no public score, so they are not filtered.
excludeKeywords	Drop hits that mention any of these words, e.g. to remove an unrelated meaning of your brand name.
matchMode	wholeWord (default) or substring.
maxTotalItems	Cost cap per run (default 1,000; 0 = no cap). Newest mentions come first.
onlyNew	Monitor mode: output only mentions not returned by earlier runs.
monitorKey	Name for this monitor's memory, e.g. acme-brand, so editing the input keeps the memory.

Example: brand + domain alerts, comments and stories, hourly schedule:

{
  "keywords": ["Acme", "Acme Cloud"],
  "domains": ["acme.io"],
  "excludeKeywords": ["Looney Tunes"],
  "onlyNew": true,
  "monitorKey": "acme-brand",
  "lookbackDays": 2
}

Output

One row per matching story or comment, newest first. Each row has a direct link back to Hacker News.

Field	Description
type	story or comment
hnUrl	Link to the item: https://news.ycombinator.com/item?id=<id>
matchType / matchedTerm / matchedTerms	keyword, domain or author; the term that matched (all matched terms if several did)
excerpt	~280 characters around the first match
author / authorUrl	Who posted it
createdAt / createdAtUnix	When it was posted (UTC)
storyTitle / storyHnUrl / storyId	The story (for comments: the parent story it belongs to)
linkedUrl / linkedDomain	The article the story links to (empty for Ask HN and other text posts)
parentHnUrl / parentId	For comments: the item it replies to
points / numComments	Story score and comment count (empty for comments)
storyType	show_hn, ask_hn, job or poll when it applies
text	Full plain text of the comment or story body
isNew	true in monitor mode

Example comment row:

{
  "type": "comment",
  "id": "50240994",
  "hnUrl": "https://news.ycombinator.com/item?id=50240994",
  "matchType": "domain",
  "matchedTerm": "github.com",
  "matchedTerms": ["github.com"],
  "excerpt": "https://github.com/infumap/imwebsearch . web-search routes each query to the free quota of one of a number of providers. ...",
  "author": "math",
  "createdAt": "2026-10-26T08:56:47Z",
  "storyTitle": "DeepSeek search and fetch – $0.75 per 1000 successful requests",
  "storyHnUrl": "https://news.ycombinator.com/item?id=50138531",
  "linkedUrl": "https://api-docs.deepseek.com/guides/search/",
  "parentHnUrl": "https://news.ycombinator.com/item?id=50139399",
  "points": null
}

Export as JSON, CSV, Excel, or send to Google Sheets, Slack, email or webhooks with Apify integrations.

Scheduling a monitor

1. Set your keywords / domains, turn on Monitor mode (onlyNew), and give it a monitorKey.
2. Run it once. The first run returns mentions from the look-back window and saves them as the baseline.
3. Go to Schedules > Create, pick this Actor and this input, and run it every 15 minutes, hourly or daily.
4. Each run scans from the previous run (minus a 3-hour overlap, to catch items the search index picks up late) and returns only mentions it hasn't returned before. Already-seen hits are free.
5. Add an integration (Slack, email, webhook) on the schedule to get alerts. Runs with no new mentions produce an empty dataset.

The memory lives in the named key-value store hn-mention-monitor-state in your account and is kept until you delete it. If a run stops at maxTotalItems, the leftover mentions are returned on the next run.

Pricing

• $0.50 per 1,000 mentions written to the dataset (pay-per-event), plus the Apify Actor start fee of $0.00005 per run.
• Search hits dropped by the whole-word check or excludeKeywords, and already-seen hits in monitor mode, are not charged.
• Set Max cost per run in the run options or maxTotalItems to cap spend.
• An hourly monitor for a brand with 50 mentions a day costs about $0.03 a day, plus start fees.

Limits and fair use

• Data comes from the public Algolia HN Search API, which allows about 10,000 requests per hour per IP. This Actor makes 2 requests per time window and pauses between requests. A typical monitor run makes fewer than 10.
• New HN items usually show up in search within minutes. The 3-hour overlap in monitor mode covers slower indexing.
• Items deleted on HN may drop out of search results.
• This tool returns links and excerpts for monitoring and research. Every row links back to Hacker News; please don't use it to republish HN content in bulk.

FAQ

Does it find mentions in comments under unrelated stories? Yes. That's the main point: each comment is matched on its own text.

Will "Go" match every "go" in English? Whole-word matching still matches the plain word. For short or common names, use a more specific phrase (Go 1.24, golang) or add excludeKeywords.

Can I track a competitor's launches? Add their product name and domain, set minPoints to see only stories that took off, or add the founder's username to authors.

Can I watch several brands in one run? Yes. Add several keywords and domains. A row that matches several terms appears once, with all of them in matchedTerms.

How is this different from your Hacker News Scraper? This Actor is a monitor: it re-checks matches, remembers what it already returned, and is priced for frequent scheduled runs. The Hacker News Scraper is for bulk exports: full threads, front page and parsed hiring posts.

About

Built and maintained by Tali, an AI agent, for Trevor Staubach. Code and README are AI-written and tested on live data. Found a bug or need a field? Open an issue on the Actor's Issues tab.