FlowRunner
Pricing
Theme

Hacker News

Analytics & Data

Read-only access to Hacker News through its Firebase and Algolia APIs: fetch items and users, pull story-list feeds, hydrate front-page stories, and run full-text search. No credentials required.

15 actions available
A scheduled check fires for a monitored keyword
Agent runs Search by Date with a numeric filter for new stories
Agent reads each hit's title, points, comments, and URL
Agent flags stories that clear the relevance and score threshold
Agent confirms the match is on-topic, not a false positive
Agent drafts an internal summary of the flagged stories
Any external post or outreach based on a story waits for approval

What This Integration Enables

The Hacker News connector lets agents read Hacker News inside a workflow. Agents fetch individual items and users, pull the top, new, best, Ask, Show, and job story-list feeds, and hydrate the front page in a single call. The Algolia actions run full-text search by keyword, author, or content type, search by date with numeric filters over points and comment counts, and fetch a story with its entire nested comment thread at once. Because everything is read-only and unauthenticated, the connector is pure signal in; the surrounding flow decides what to summarize, log, or escalate.

Without FlowRunner

Feed watched by hand Someone refreshes Hacker News and searches for a keyword ad hoc
Stories missed A relevant post scrolls off the front page before anyone sees it
Threads read manually Following a discussion means opening and re-reading it repeatedly

With FlowRunner

Feed watched on a schedule The agent searches and reads the feeds automatically
Relevant stories surfaced On-topic posts above the score threshold become items a person sees
Threads fetched whole A full nested comment tree comes back in one call for summarization

Use Case Scenarios

Front-Page Digest

On a schedule the agent calls Get Top Stories (Hydrated) to pull the current front page as full items, then posts the highest-scoring headlines with links to a team channel. The hydrated action returns the first N items in one call, so the digest is one request rather than a story-list of ids followed by dozens of lookups.

Keyword Monitoring

The agent watches for mentions of a product, company, or topic. Using Search by Date with a `created_at_i` numeric filter, it collects every new story matching the keyword, reads points and comment counts, and logs each hit to a monitoring sheet. Stories that clear the relevance and score threshold get flagged for a person; the rest are recorded for the trend.

Trending Thread Summary

When a discussion about the team's space starts trending, the agent calls Get Item (Algolia) to fetch the full nested comment tree in one call, then hands the thread to a summarization step. The summary lands in a community channel so the team follows the conversation without each person reading hundreds of comments.

Human-in-Loop Highlight

Reading Hacker News is safe. Acting on what you read, replying in public, posting a competitor's news externally, kicking off outreach, carries reputation risk the agent should not absorb alone. The connector is read-only, so the agent gathers signal freely, but the moment a story would drive an external action, it stops. When a monitored keyword surfaces a story that clears the threshold, the agent assembles the title, score, thread summary, and a proposed next step, then routes to the owner through their channel: "A post about our category is on the front page with 300 points and an active thread. I drafted a note for the community channel. Post it, or hold?" This is [human-in-the-loop](/concepts/human-in-the-loop/) at the boundary between listening and speaking: the agent handles the reading and the summarizing, and a person owns what the organization says in public.

Agent processes routinely
Detects exception requiring judgment
Clear match Continues automatically
Ambiguous Routes to human via preferred channel
Human decides
Agent resumes with decision

Agent Capabilities

15 actions

Items & Users

3
  • Get Item Fetch a single Hacker News item (story, comment, job, poll, or poll option) by numeric id.
  • Get User Fetch a Hacker News user profile including karma and submission history.
  • Get Max Item ID Return the current largest item id on Hacker News.

Story Lists

8
  • Get Top Stories Return the list of ids for the current top stories.
  • Get New Stories Return the list of ids for the newest stories.
  • Get Best Stories Return the list of ids for the best stories.
  • Get Ask Stories Return the list of ids for the Ask HN feed.
  • Get Show Stories Return the list of ids for the Show HN feed.
  • Get Job Stories Return the list of ids for the jobs feed.
  • Get Updates Return the change feed of recently updated items and profiles.
  • Get Top Stories (Hydrated) Fetch the first N top stories as full items in a single call.

Search

4
  • Search Run a full-text Algolia search by keyword, author, or content type.
  • Search by Date Run an Algolia search sorted by date with numeric filters over points, comments, or timestamp.
  • Get Item (Algolia) Fetch a story with its entire nested comment thread in one call.
  • Get User (Algolia) Fetch a hydrated user record from the Algolia API.

Start building with Hacker News

$100 in credits. No card required. Connect in minutes.