Skip to content

feat: Cleanup HTML page to reduce token usage #1073

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
mguella wants to merge 4 commits into
base: master
Choose a base branch
from
Open

feat: Cleanup HTML page to reduce token usage #1073

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
dc36bef
cleanup html page to reduce token usage
mguella Mar 20, 2026
5be0f5b
fix comments cleanup regexp
mguella Mar 20, 2026
43a9bb9
run prettier
mguella Mar 21, 2026
43a1c35
fix regex
mguella Mar 21, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions package.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,6 +25,7 @@
"clsx": "^2.1.0",
"drizzle-orm": "^0.40.1",
"js-tiktoken": "^1.0.21",
"jsdom": "^29.0.1",
"jspdf": "^3.0.4",
"lightweight-charts": "^5.0.9",
"lucide-react": "^0.556.0",
Expand Down Expand Up @@ -54,6 +55,7 @@
},
"devDependencies": {
"@types/better-sqlite3": "^7.6.12",
"@types/jsdom": "^28.0.1",
"@types/jspdf": "^2.0.0",
"@types/node": "^24.8.1",
"@types/pdf-parse": "^1.1.4",
Expand Down
17 changes: 16 additions & 1 deletion src/lib/agents/search/researcher/actions/scrapeURL.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,7 @@ import { ResearchAction } from '../../types';
import { Chunk, ReadingResearchBlock } from '@/lib/types';
import TurnDown from 'turndown';
import path from 'path';
import { JSDOM } from 'jsdom';

const turndownService = new TurnDown();

Expand Down Expand Up @@ -40,11 +41,25 @@ const scrapeURLAction: ResearchAction<typeof schema> = {
params.urls.map(async (url) => {
try {
const res = await fetch(url);
const text = await res.text();
let text = await res.text();

const title =
text.match(/<title>(.*?)<\/title>/i)?.[1] || `Content from ${url}`;

// if response is an html page, clean it up to reduce the amount of tokens used
if (res.headers?.get('Content-Type') === 'text/html') {
// remove comments and spaces
text = text
.replace(/<\!--[\s.]*?-->/gm, '') // comments
Comment thread
cubic-dev-ai[bot] marked this conversation as resolved.
Outdated
.replace(/^\s+|\s+$</gm, '') // head and tail spaces
.replace(/\s+</gm, '<') // spaces before tags
.replace(/>\s+/gm, '>') // spaces after tags
const dom = new JSDOM(text);
// removed unused tags
dom.window.document.querySelectorAll('script, style, template').forEach(el => el.remove());
text = dom.window.document.documentElement.outerHTML;
}

if (
!readingEmitted &&
researchBlock &&
Expand Down
Loading