703 comments, 6 camps, 1 minute
Paste any Hacker News or Reddit thread and get a citable opinion map — who agrees, who pushes back, and which questions nobody answered.
No account needed. We only email you when ThreadMap opens up.
Why Comment Section Analysis Matters Right Now
On September 13, 2026, Dario Amodei published "We must pace the frontier" and it landed on Hacker News with 511 points and 703 comments — a total engagement of 1,917. The same day, a separate Ask HN thread titled "Can we please limit the AI news flood?" hit the front page. Two independent signals, one shared complaint: the most valuable discussion on the internet is now buried under its own volume. Nobody can read 703 comments. Nobody can cite them either, because there is no structure to cite.
Existing tools do not solve this. Hacker News sorts comments by heat or time, which tells you what is popular but not what is being argued. Reddit's top comments do the same. Kagi Summarizer and Perplexity summarize articles, not the debate underneath them. Pasting 700 comments into a chat window blows past the context limit and produces a mush of averaged opinion. Sorting is not clustering. A ranked list of comments is information; a map of camps, stances, and rebuttals is intelligence.
The timing is what makes this worth building now. Over the past thirty days there have been at least five long-form AI policy essays that each generated hundreds of comments — Amodei, Altman, LeCun, policy white papers, and open letters. Every one of those threads is a free acquisition event for a tool that can structure it. Journalists need quotes with attribution. Policy analysts need to know which camp is winning. Podcast hosts need material. All of them are already in the comment sections, and none of them have a tool that turns a thread into something they can actually use.
How It Works
Paste a thread URL
Drop in a Hacker News item page or a Reddit post. ThreadMap pulls the full comment tree through the official Firebase API — no scraping, no API key, no login required to try it.
We cluster the debate
Every comment is flattened, deduplicated, and clustered into five to eight opinion camps. Each camp gets a stance label — supporting the original piece, opposing it, or arguing a third path — plus its share of the total discussion.
Read, cite, export
You get representative quotes with usernames, permalinks, and upvote counts, a rebuttal map showing which camps argue with which, and the top unanswered questions. Copy it as Markdown or download a PDF for your editor.
What You Get
A map, not a summary
A summary flattens 703 comments into one averaged paragraph. ThreadMap keeps them apart: it shows you how many camps exist, what each one argues, and which camps are directly rebutting each other. You see the shape of the disagreement instead of losing it.
Citable by default
Every representative quote ships with the original username, a permalink to the exact comment, and its upvote count. Journalists and researchers can drop a quote straight into a draft without hunting through the thread to verify the source.
Built for 500+ comment threads
Generic summarizers choke on long threads and quietly drop the tail. ThreadMap is designed specifically for the 500-to-1000 comment range where the real debate lives, and it also surfaces the top questions that were asked but never answered.
Built on the official Hacker News API · No account required to try · Your email is never sold or shared