Get All Text

Read the text of the page open in the active tab, or of one part of it, into a field.

Try it

工作流 · summarize-page-llm-chain.awf 4 个步骤

Summarise this page in 3 bullets：包含 4 个步骤的工作流，从 When Started 开始，到 Display Markdown 结束。

1. 1 When Started（触发器 — 打开其文档页面） 触发器
2. 2 Get All Text（页面内操作 — 打开其文档页面） 页面内操作
3. 3 Basic LLM Chain（AI — 打开其文档页面） AI + Web LLM
4. 4 Display Markdown（页面内操作 — 打开其文档页面） 页面内操作

在 AWFlow 中打开 复制 .awf

What it does

Get All Text reads the readable text of the active tab and stores it, as one piece of text, in the field you name in Output field name.

• With DOM Selector empty, it reads the whole page body. With a selector, it reads only the first element that matches.
• By default it returns the text the way the browser lays it out, with line breaks between blocks. Turn on Visible text only to skip anything hidden on the page (hidden menus, collapsed sections, invisible elements). Each visible piece of text then comes on its own line, with extra spaces removed.
• The output holds only the new field. Turn on Include input fields to keep the fields of the incoming item next to it.

It reads the main page only: text inside embedded frames is not included.

When to use it

• Feed an article, a product page or a ticket to an AI node to summarize, classify or extract from it.
• Read just one area of a page (a price box, a comment thread) by picking it in DOM Selector.
• To read a page that is not open in a tab, use Get All Text from Link instead. To read only what the user highlighted, use Get Selected Text.

Inputs and settings

设置项	类型	作用
Output field name 必填（必填）	Text	The name of the output field.
DOM Selector	Page element	The selector of the DOM element to interact with.
Options	Group	The options for the node.
Options › Include input fields	Toggle	Whether to include the input fields in the output. 默认值：Off
Options › Visible text only	Toggle	Only return text from visible elements. 默认值：Off

• Output field name: any name you like, for example pageText. Later nodes read it with {{ $input.pageText }}.
• DOM Selector: click the crosshair button in the field, then click the element on the page; the selector is filled in for you. You can also type a CSS selector such as article or #main. The Ping button flashes the element the selector matches, so you can check it.

Outputs

• Output

One item with your field, for example { "pageText": "Welcome to …" }. The value is plain text, not HTML.

Dependencies and credentials

None. This node needs no connection and has no slots for other nodes.

Example workflow

1. Start with a Context Menu trigger so you can run the workflow from any page.
2. Add Get All Text with Output field name = pageText and Visible text only on.
3. Add a Summarization node with Text = {{ $input.pageText }}, and connect a chat model to it.
4. End with Display Markdown to show the summary on the page.

Common issues

• “No content found”. The page or the selected element had no text, or DOM Selector matched nothing. Use Ping to check the selector, and make sure the page has finished loading. Add Wait For Element before this node on pages that load content late.
• The text is missing a part you can see. That part is probably inside an embedded frame, which this node does not read. Open the frame’s address in its own tab with Tab Control and read it there.
• The text includes menus or hidden content. Turn on Visible text only, or narrow the read with DOM Selector.
• The node fails outside the extension. This node runs in the browser extension; install it or run the workflow from the extension.

Recipes with this node

Summarize the page you're reading, on your device 入门 · 5 分钟 设备端

Related nodes

• Get All HTML: the same area as HTML instead of text.
• Get Structured Data: tables and lists as rows.
• Get Selected Text: only what the user highlighted.
• Get All Text from Link: read a URL without opening it.