hn-digest
Skill0 starsChanged 9 months agoMITView on GitHub
Digests Hacker News posts into a structured summary. Given a Hacker News URL (news.ycombinator.com/item?id=...), fetches the linked article and comment threads, then produces a concise summary of the article followed by an organized breakdown of community reactions (agree / disagree / different voices). Use this skill whenever the user shares a Hacker News link, asks to summarize an HN discussion, says things like "what are people saying about this HN post", "digest this HN thread", "summarize the comments on this", or pastes any news.ycombinator.com URL. Also trigger when the user references "Hacker News", "HN thread", or "HN comments" alongside a link or post ID. This skill is for link-type posts (posts that point to an external article); for text-only posts like "Ask HN" or discussion threads without an external link, skip this skill and summarize conversationally instead.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- thevibeworks/claude-reads-hntheir own version, first seen Sep 3, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- thevibeworks/claude-reads-hn★ 17same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository3 other files it gives its agents.
- alt-text-generatorSkill
- apple-store-trackerSkill
- emoji-recommenderSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 3 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 30
If the post is a text-only "Ask HN" or "Show HN" with no external link, don't use this skill.
- Very specific: line 37
Use web_fetch on the Hacker News URL to get the discussion page.
- Very specific: line 49
Use web_fetch on the external article URL.
- Very specific: line 96
Show the weight of each position: After the paraphrased text, include a rough comment count in parentheses like (~5 comments) indicating …
- Distinctive: line 102
If a topic has comments on one side but not the other, leave that cell with "—" (em dash).
- Distinctive: line 139
Notable replies matter: If a top-level comment sparks a substantive back-and-forth, fold the key takeaway from that thread into the bullet …
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
HN Digest
You turn a Hacker News post into a structured, opinionated digest: a tight summary of the original article plus an organised map of community sentiment from the comments.
When this skill applies
The user gives you a Hacker News URL (e.g. https://news.ycombinator.com/item?id=12345)
or asks you to summarize / digest an HN discussion. This skill targets
link-type posts — posts that point to an external article or blog post.
If the post is a text-only "Ask HN" or "Show HN" with no external link, don't use this skill; just respond conversationally.
Workflow
Step 1 — Fetch the HN comments page
Use web_fetch on the Hacker News URL to get the discussion page. Extract:
- Post title
- External article URL (the link the post points to)
- Point count and comment count (if visible)
- Comment threads — capture top-level comments and notable replies (replies that add substantial new information, strong counterarguments, or get significant engagement). Skip low-signal replies like "+1", "this", or single-emoji responses.
Step 2 — Fetch the linked article
Use web_fetch on the external article URL. Read the full content. If the
fetch fails or returns a paywall, note this and work with whatever content is
available from the HN discussion itself (commenters often summarize or quote
the article).
Step 3 — Write the digest
Produce the output in this structure:
📰 [Post Title]
Source: [domain of the article] HN: [link to the HN discussion] · [X points] · [Y comments]
Key Takeaways
A 100–200 word bullet-point summary of the original article. Each bullet should capture one distinct claim or insight from the article. Focus on:
- The core thesis or announcement
- The most important supporting evidence or arguments
- Why it matters (so what?)
Aim for 3–6 bullets. Write in your own words — distill, don't parrot. If the article is technical, make it accessible without dumbing it down.
Community Debate
Organize the agree and disagree comments into a markdown table grouped by topic. Each row represents one debated question or claim from the thread. This lets the reader compare opposing views at a glance.
Table format:
| Topic | 🟢 Agree | 🔴 Disagree |
| [Short topic label] | [Paraphrased position] (~N comments) 1 2 | [Paraphrased position] (~N comments) 1 |
Guidelines for the table:
- Identify 3–7 topics where commenters are clearly engaging the same question from opposite sides. The topic label should be a short, neutral phrase (e.g., "Learning curve", "Applicability beyond safety-critical systems").
- Each cell should be 1–3 sentences — tight enough to scan, detailed enough to be useful.
- Show the weight of each position. After the paraphrased text, include
a rough comment count in parentheses like
(~5 comments)indicating how many substantive comments express this view. Then list numbered markdown links to the most representative comments (up to 3). This lets the reader instantly distinguish a lone contrarian from a broad consensus. Example:Formal specs catch bugs earlier (~6 comments) [1](url) [2](url) [3](url) - If a topic has comments on one side but not the other, leave that cell with "—" (em dash). This is fine and expected.
- Row order: put the most actively debated topics first.
- The total number of filled Agree vs Disagree cells should roughly reflect the actual proportions of the thread. If the thread skews 70% supportive, the table should have more filled Agree cells than Disagree cells.
🔵 Different Voices
Bullet points for comments that take the conversation in a genuinely different direction — tangential insights, related experiences, adjacent topics, or meta-observations that don't fit the debate table. These are often the most interesting part of HN threads. Each bullet should:
- Describe the alternative angle or tangent
- Explain why it's noteworthy or what it adds to the discussion
- End with a markdown link to the HN comment
Guidelines
Proportional representation. The number of bullets in each section (Agree, Disagree, Different Voices) should roughly reflect the actual balance of the comment thread. If 70% of substantive comments agree and only a handful disagree, don't force equal bullet counts — let the digest reflect reality. Typical range is 2–7 bullets per section, but skew with the thread.
Every bullet gets a link. End each comment bullet with a markdown link to
the specific HN comment being summarised. This lets the reader click through
to the full thread if a point is interesting. Format:
[link](https://news.ycombinator.com/item?id=XXXXX)
Attribution without over-quoting. Paraphrase comments in your own words. You can reference a commenter's claimed background if relevant ("one commenter who works in chip manufacturing noted...") but don't use HN usernames.
Notable replies matter. If a top-level comment sparks a substantive back-and-forth, fold the key takeaway from that thread into the bullet rather than listing each reply separately.
Signal over noise. HN threads have a lot of low-signal comments. Your job is editorial — surface the insights that someone skimming the thread would miss, not catalog every hot take.
Tone. Be direct and slightly opinionated in your editorial voice. You're writing a digest for a busy person, not a neutral Wikipedia summary.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

