Hacker News logo

Hacker News

Organization
unifapi-agent
hacker-news

When a workflow needs public Hacker News data through UnifAPI — the top/new/best/ask/show/jobs story feeds, item metadata, comment trees, or user karma. Also use on "did our launch land on HN," "scan Show HN for," "read this HN thread," "Hacker News sentiment," or when another skill (launch monitor, social listening) needs the deterministic HN read path. No keyword search — scan the feeds. Connect via the `unifapi` skill first. Read-only research, never posts.

Overview

Publisherunifapi-agent
Repositoryagents
Skill namehacker-news
Stars
590
Forks
115
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by unifapi-agent on GitHub. Read the source before you install it.

Installation

Install the Hacker News AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/unifapi-agent/agents/tree/main/skills/hacker-news-agent/hacker-news
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/unifapi-agent/agents.git /tmp/agents
mkdir -p .claude/skills
cp -r /tmp/agents/skills/hacker-news-agent/hacker-news .claude/skills/hacker-news
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Hacker News in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Hacker News on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Hacker News is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

hacker-news

The deterministic read path for public Hacker News data through UnifAPI. This is a Data Skill: it does not run a marketing job on its own — it names the concrete hacker-news/... operations, response shapes, and gotchas so any launch-reception or listening workflow for dev/infra/startup categories reads from one known recipe instead of rediscovering the surface each time.

Read-only — eyes, not hands. It researches public HN data and returns cited records; it never posts stories or comments, and UnifAPI never holds HN credentials.

Use the unifapi skill for live evidence

Connect once through the shared unifapi skill (OAuth MCP), then call the operations below. Story feeds take a {feed} of top, new, best, ask, show, or jobs. Keep any billing metadata so the output can state record cost.

Response contract

Single-entity endpoints return the object in data:

json
{
  "request_id": "unif_...",
  "data": {},
  "billing": { "records_charged": 1, "balance_remaining": 99 }
}

List endpoints return an array in data plus pagination:

json
{
  "request_id": "unif_...",
  "data": [],
  "pagination": { "has_more": false, "next_cursor": null },
  "billing": { "records_charged": 1 }
}

When pagination.has_more is true, pass pagination.next_cursor as the next request's cursor (limit sets page size). Always preserve billing.

Core operations

NeedOperation
Story feed (hydrated)hacker-news/stories/{feed}/items
Story feed (ids only)hacker-news/stories/{feed}
Item (story / comment)hacker-news/items/{id}
User profile + karmahacker-news/users/{id}
Recently changedhacker-news/updates
Highest item id (scan bound)hacker-news/max-item

{feed} ∈ top · new · best · ask · show · jobs.

Need a field not listed here? Use the unifapi skill's get_operation to read the exact schema before calling — but pick the operation from this table, don't discover blind.

Workflow

The deterministic recipes. Pick the one that matches the job; each names exactly what to call.

  1. Frame the watch. Pick the feeds: show and ask for launches and questions, top and best for what's landing, new for the freshest.
  2. Scan a feed. Call hacker-news/stories/{feed}/items (hydrated items in one call — cheaper than ids-then-fetch) and match stories by title or url to your product or category. Page via next_cursor.
  3. Read the thread. On a matching item, read score (points) and descendants (comment count), then walk the kids ids with hacker-news/items/{id} to read the candid critique in the comment tree.
  4. Weight a critic. Call hacker-news/users/{id} for karma and submitted to judge how much a commenter's take carries.
  5. Watch for change. Call hacker-news/updates for recently changed items and profiles, or hacker-news/max-item to bound a fresh scan.
  6. Cite everything. Every figure ties back to the HN item it came from; report billing.records_charged (or estimate when billing metadata is absent).

Shape notes

  • HackerNewsItem — type (story · comment · job · poll · pollopt), by, time, title, url, text, score (points), descendants (comment count), kids (child comment ids, not objects), parent.
  • HackerNewsUser — id, created, karma, about, submitted (item ids).
  • HackerNewsUpdates — items and profiles that recently changed.

Gotchas

  • No keyword search. Discovery is feed-driven — scan top/new/best/ask/show/ jobs and match by title/url. Say so in the brief.
  • kids are child comment ids, not embedded objects — recurse with hacker-news/items/{id} to read the comment tree.
  • descendants is the comment count and score is the points — don't confuse them.
  • Prefer hacker-news/stories/{feed}/items over hacker-news/stories/{feed} when you want the hydrated stories, not a list of ids.
  • A low balance can silently truncate list pages: check billing.truncated_due_to_balance — when true the page is partial.

Output

Return the records the calling workflow needs, each cited to its HN item, plus a one-line cost note (records_charged). When this skill is used directly, a compact reception brief is the default:

markdown
**{Story title}** ({feed}) — {score} points, {descendants} comments. Landed: {yes/no}. Candid takes: "{quote}" (by {user}, {karma} karma). Evidence: {item URLs}. Records: ~{N}.

Related skills

  • competitor-launch-monitor (Competitive Intelligence), social-listening-brief (Social Listening) — launch-reception and listening work on top of this read path.
  • unifapi — the shared data skill: connect MCP and look up exact schemas with get_operation.

Frequently asked questions

What does the Hacker News AI skill do?

When a workflow needs public Hacker News data through UnifAPI — the top/new/best/ask/show/jobs story feeds, item metadata, comment trees, or user karma. Also use on "did our launch land on HN," "scan Show HN for," "read this HN thread," "Hacker News sentiment," or when another skill (launch monitor, social listening) needs the deterministic HN read path. No keyword search — scan the feeds. Connect via the `unifapi` skill first. Read-only research, never posts.

Why use Hacker News on TypingMind?

Because you install it once and use it with any model. Hacker News is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Hacker News in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/unifapi-agent/agents/tree/main/skills/hacker-news-agent/hacker-news. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Hacker News?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Hacker News?

As many as you like. As long as a model supports skills, you can use Hacker News with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Hacker News AI skill free?

Yes. It is published on GitHub by unifapi-agent under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇