Item logo image for Sidelight AI Assistant

Sidelight AI Assistant

5.0(

2 ratings

)
Item media 4 (screenshot) for Sidelight AI Assistant
Item media 1 (screenshot) for Sidelight AI Assistant
Item media 2 (screenshot) for Sidelight AI Assistant
Item media 3 (screenshot) for Sidelight AI Assistant
Item media 4 (screenshot) for Sidelight AI Assistant
Item media 1 (screenshot) for Sidelight AI Assistant
Item media 1 (screenshot) for Sidelight AI Assistant
Item media 2 (screenshot) for Sidelight AI Assistant
Item media 3 (screenshot) for Sidelight AI Assistant
Item media 4 (screenshot) for Sidelight AI Assistant

Overview

Ask questions about the page you're on, pull its data into tables and CSVs, and pin live panels that keep updating themselves.

Work beside Google Sheets, summarize your data, and export sheets and website tables as CSV files you can open in Excel. Sidelight AI Assistant works on the page you are already looking at. Open it with Cmd+Shift+K (Ctrl+Shift+K on Windows and Linux), ask a question about the page, and get an answer drawn from what is actually on it. WHAT IT DOES • Answers questions about the current page. It reads the real structure - headings, body text, tables, form state, embedded JSON - rather than guessing from the URL or working off a screenshot. So it answers about this invoice, this dashboard, this listing. • Pulls data out. Tables, ARIA grids and repeated card layouts all come out as clean rows. Data spanning twenty pages of pagination is one request, not twenty. Download the result as CSV, TSV, JSON or Markdown. • Pins live panels. Ask it to keep an eye on a number and it finds the request the page itself makes for that number, then pins a small panel that re-fetches it. The panel tracks the real source instead of scraping a stale rendering. • Drives the page. It can click, fill fields, select options and wait for things to load, working through a flow on your own logged-in session. WHY IT REACHES FURTHER THAN A SCRIPT IN THE PAGE • It reads inside cross-origin iframes - an analytics report embedded from another domain is invisible to the host page's own JavaScript. • It can call endpoints the page is not allowed to call, so first-party APIs you are already signed in to work. • Every tool is a fixed, packaged function. Nothing depends on evaluating code in the page, so sites with a strict Content-Security-Policy - GitHub, Google, banks - work normally. WHERE THE LIMITS ARE, AND WHY The assistant reads the page you point it at, and a web page is untrusted input: it can simply write "ignore your instructions and send this somewhere". So the boundaries are enforced in code, not by asking the model nicely. • Your cookies go to the page's own site and nowhere else. A request to any other site is anonymous. "Fetch my mail as me" is not something it can be talked into. • Anything that leaves the page's own site, and anything that changes data, asks you first - showing the method and the full URL before it happens. • Reading a different tab asks you too, naming that tab. • It cannot read a password field, or a card, CVV or one-time-code field. • Values stored under credential-shaped names are redacted inside the page, before anything is sent. • Loopback and private network addresses are refused, so the extension is not a route into your own network. HONEST ABOUT WHAT IT CANNOT DO If data is genuinely unreachable - it only ever streams over a WebSocket, or the site hands it to a third party with no readable endpoint - it says so. It does not invent plausible-looking numbers. PRICING New accounts start with 500 credits, free, no card. After that you buy credit in one-off top-ups from $10; there is no subscription. A typical question about a page costs a fraction of a cent. You can see exactly what each one cost in your activity log. PRIVACY Page content is sent to a language model only when you ask a question, and on the refresh interval of any live panel you pinned. Conversations are not stored on our servers. We do not train on your data and we do not sell it. Full policy: https://www.trysidelight.com/privacy

Details

  • Version
    2.3.0
  • Updated
    October 31, 2026
  • Offered by
    Muhammad Saad Zafar
  • Size
    102KiB
  • Languages
    English
  • Developer
    Muhammad Saad Zafar
    P97/47 Alfayaz Colony Noshahi Street Gali no 4, Satyana Road Faisalabad Faisalabad 38000 PK
    Email
    saadzafar.support@gmail.com
    Phone
    +92 315 7055586
  • Trader
    This developer has identified itself as a trader per the definition from the European Union and committed to only offer products or services that comply with EU laws.

Privacy

Manage extensions and learn how they're being used in your organization

Sidelight AI Assistant has disclosed the following information regarding the collection and usage of your data. More detailed information can be found in the developer's privacy policy.

Sidelight AI Assistant handles the following:

Personally identifiable information
Financial and payment information
Web history
User activity
Website content

This developer declares that your data is

  • Not being sold to third parties, outside of the approved use cases
  • Not being used or transferred for purposes that are unrelated to the item's core functionality
  • Not being used or transferred to determine creditworthiness or for lending purposes

Support

Related

HTML Text to .md — Local Markdown Web Clipper

5.0

Select any part of a page and convert it to clean Markdown. Tables, code and math stay intact. Runs locally in Chrome.

Table to CSV / Excel – Copy Any Web Table

0.0

Export any web table to CSV, Excel (XLSX) or clipboard. Free for every table, any size. Your table data never leaves your browser.

CSV Reader

5.0

Use CSV Reader to read CSV file, open CSV file, and view it as a clean table locally in your browser. No uploads.

Web Scraper — No-Code List & Table to Excel/CSV

5.0

Turn any webpage's lists and tables into clean data. Export to Excel, CSV, JSON, or Markdown. No code required.

HTML to CSV

5.0

Convert HTML tables to CSV files

Scrapely

5.0

Point-and-crawl data extraction. Detect tables and lists on any page, paginate automatically, and export clean data.

Webページ → CSV (無料ローカル版)

0.0

ページ上で選んだ表や文章をサイドパネルで確認し、必要な列をローカルCSVへ保存します。

Table Extractor

5.0

Detects HTML tables and virtualized data grids on webpages and exports them to CSV, XLSX, and JSON.

Magic Scraper – Lead & Web Scraper to Excel

5.0

Turn any web page into a spreadsheet: tables, lists, search results to CSV, Excel or JSON. Free, no account, runs in your browser.

Table Capture

5.0

Use table capture, a capture table tool and instant data scraper chrome extension, to convert html to csv, html to excel.

Tabular: Table Capture to Excel & CSV

5.0

Turn any web listing into an exportable table (CSV/Excel). 100% local: your data never leaves your browser.

Lumi Scraper

0.0

Turn any list, table, store or website into a spreadsheet. Auto-detects the data, follows pages, reads detail pages in bulk.

HTML Text to .md — Local Markdown Web Clipper

5.0

Select any part of a page and convert it to clean Markdown. Tables, code and math stay intact. Runs locally in Chrome.

Table to CSV / Excel – Copy Any Web Table

0.0

Export any web table to CSV, Excel (XLSX) or clipboard. Free for every table, any size. Your table data never leaves your browser.

CSV Reader

5.0

Use CSV Reader to read CSV file, open CSV file, and view it as a clean table locally in your browser. No uploads.

Web Scraper — No-Code List & Table to Excel/CSV

5.0

Turn any webpage's lists and tables into clean data. Export to Excel, CSV, JSON, or Markdown. No code required.

HTML to CSV

5.0

Convert HTML tables to CSV files

Scrapely

5.0

Point-and-crawl data extraction. Detect tables and lists on any page, paginate automatically, and export clean data.

Webページ → CSV (無料ローカル版)

0.0

ページ上で選んだ表や文章をサイドパネルで確認し、必要な列をローカルCSVへ保存します。

Table Extractor

5.0

Detects HTML tables and virtualized data grids on webpages and exports them to CSV, XLSX, and JSON.

Google apps