Return fetched pages as Markdown, not raw HTML#16
Merged
Merged
Conversation
A raw page is mostly chrome: scripts, nav, styling, and boilerplate that buries the text and burns the model's context. The new readable package lifts out the main content and renders it as Markdown, which the model reads far better than a wall of tags. It is pure Go over x/net/html, so tomo stays a single static binary. The fetch tool now runs HTML through it, returning the page title and the Markdown body. Other content types are still passed through as text.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The fetch tool used to hand the model a raw HTML page. Most of that is chrome (scripts, nav, styling, boilerplate) that buries the actual text and wastes context.
The new readable package extracts the main content (article, else main, else body) and renders it as Markdown: headings, paragraphs, lists, links, emphasis, inline and block code, tables, and images, with chrome dropped and whitespace collapsed. It is pure Go over golang.org/x/net/html, so tomo stays a single static binary.
fetch now runs HTML through it and returns the page title plus the Markdown body. Non-HTML content types are still passed through as text.