Hacker News new | ask | show | jobs
Show HN: Reading Tree, a weighted outline for articles instead of a summary (github.com)
2 points by ModelVoyager 84 days ago
I built this for close reading, especially philosophy chapters, long essays, and dense nonfiction. AI summaries are useful in many cases, but sometimes the source is good enough that I want to read it properly, not just get the gist. Those are exactly the cases where a summary can leave out the parts I would care about most.

Reading Tree keeps the original words in place. Every node links to the passage it covers, and every paragraph links back to the node that explains its role. Nodes are weighted by importance so you can see at a glance which parts carry the most weight.

Demo (desktop only): https://modelvoyager.github.io/ReadingTree/on-liberty-ch1-re...

GitHub: https://github.com/ModelVoyager/ReadingTree

It's an agent skill (a set of instructions and templates you upload to an AI chat) that works with Claude and ChatGPT. Only tested with Claude Opus 4.6 (extended thinking) and GPT-5.4 Thinking. Other models may not produce usable results. Free and open source. I designed the interaction model and UX, then used AI to build it out. Most of the effort went into actually reading with it and fixing what didn't work.

I've been the only user for a few weeks. If you read things where the argument structure matters and summaries feel like they lose too much, I'd like to hear whether this helps.

1 comments

That's brilliant. Very intuitive and useful.

- I'd like to see this as a hosted app versus something that has to be "installed" in a chatbot.

- It needs a text search feature for both the outline and full text. That would allow searching for text containing "government", highlight instances and seeing their context. And same for searching the outline for "government" and seeing supporting text.

This could be an equally useful paradigm for fiction and for source code. For fiction, it would be really useful if this could be trained to identify character introductions and locations and their mentions. Imagine how convenient it would be for the outline to mention a plot point about "Mary Sullivan" where the paragraph in chapter 22 only says "his mother" when talking about "John Sullivan."

Thanks, glad to hear it.

I've thought about a hosted app and it's something I'd like to do eventually. The API costs and building a reliable service make it a bigger project though, so it's not happening soon.

Text search is a good idea. It makes reading more active because you pick a concept, see where it shows up and how it fits the structure. The fiction use case you mentioned is similar in a way, finding where a character or concept appears even when the text doesn't use the exact same words. I'll look into it.