Read Webpage Text from a Link | Exa | BlockRun

Paste a public webpage link to read the text, title and source link available from that page. Your AI can summarize an article, explain a document or answer a question using the text it can read.

Your AI

Use with your AI

For tasks with several steps, continue in the AI you already use.

CodexClaude CodeCursorWindsurfCline

View instructions

Use Toolmu’s “Read Webpage Text from a Link” tool to read the public link in [brackets]. Ask for the link if it is missing. Summarize only the text you can read, include the original page link, and say when the page or needed text is unavailable. Do not guess or add details that are not on the page. If the text may not be up to date, say so. Before any paid request, tell the user the current price and wait for approval. Follow the Toolmu connection guide only if help connecting is needed.

Task:
Read [public webpage link] and summarize the text you can read in simple language. Include the original page link. If the page or any needed text is unavailable, say so.

Paste into your connected AI's conversation.

Connection guide

Up to 0.00396 credits / call

The final charge is based on confirmed usage. Unused reserved credits return to your available balance.

BlockRun

Read Webpage Text from a Link | Exa | BlockRun

Paste a public webpage link to read the text, title and source link available from that page. Your AI can summarize an article, explain a document or answer a question using the text it can read.

You receive

When page content is available, receive extracted text and its source details. Retrieval information can show whether the page succeeded or failed and whether the text came from cache or a crawl. Author and associated image information may also be available. Your Agent can summarize the returned text while keeping its source URL.

Start with your task

Read [public webpage link] and summarize the text you can read in simple language. Include the original page link. If the page or any needed text is unavailable, say so.

What to know

Content may come from cache, be incomplete or be unavailable. The result does not guarantee the page's latest version. Check the matching page status when provided, and keep missing text or metadata visible rather than filling it in.

Inputs & outputs

Inspect the current inputs, availability, and maximum charge before starting.

Input schema

Input schema
Field / Path	Type	Requirement	Meaning
Page extraction request`body`	`object`	Required	Required request body. Put the chosen page URL in its urls array. For the basic reading task, use one URL and omit optional extraction settings. Required fields `urls`
Page URL list`body.urls`	`array`	Required	Required URL array. The service documents 1–100 page URLs; use one public URL unless the current Tool quote and your spending authorization explicitly cover a larger request.

View raw JSON

{
  "type": "object",
  "required": [
    "body"
  ],
  "properties": {
    "body": {
      "type": "object",
      "required": [
        "urls"
      ],
      "properties": {
        "urls": {
          "type": "array",
          "description": "URLs to fetch content from (1â100). Total cost = urls.length Ã $0.002."
        }
      }
    }
  },
  "additionalProperties": false
}

Result schema

No fixed output contract is available. Field types reflect saved result evidence; individual responses can differ.

Result schema
Field / Path	Type	Meaning
Page content records`result.results`	`array`	Available extracted page records. Match each record to its source URL and retrieval status; a page that fails may have no content record.
Document identifier`result.results[].id`	`string`	Identifier supplied for the returned document; it may be the page URL. Match it to retrieval status. It is separate from the Toolmu run ID.
Source URL`result.results[].url`	`string`	Source link for the returned page. Keep this URL with any summary based on its extracted text.
Extracted page text`result.results[].text`	`string`	Available page content as text. Formatting can include Markdown or code artifacts, and the result may be partial or cached. It does not reproduce browser layout.
Associated image URL`result.results[].image`	`string`	Image associated with the page when available. This does not list all page images or guarantee a downloadable original asset.
Page title`result.results[].title`	`string`	Title provided for the returned page. Keep it with the source link when presenting the text.
Reported author`result.results[].author`	`null`	Author name when available. Preserve null or an absent value rather than inventing an author.
Page retrieval statuses`result.statuses`	`array`	Fetch information for requested URLs or document IDs. A successful overall request can still include a page error.
Requested page identifier`result.statuses[].id`	`string`	Requested URL or document ID for this status. Match it to the relevant content record without assuming array order.
Content source`result.statuses[].source`	`string`	cached indicates cached text; crawled indicates fetched page content. The label does not supply cache age or guarantee the latest version.
Page retrieval status`result.statuses[].status`	`string`	success or error for this individual page. Read any accompanying error details rather than treating request success as every-page success.
Source request identifier`result.requestId`	`string`	Identifier for the Exa contents request. It is separate from a page document ID and the Toolmu run ID.
Source processing time`result.searchTime`	`number`	Server-side processing time in milliseconds, measured at Exa’s gateway. It can exclude later stages and is not total Toolmu request time.
Estimated source costs`result.costDollars`	`object`	Estimated USD breakdown for this source request. It is not an invoice or the settled Toolmu credit charge.
Estimated total source cost`result.costDollars.total`	`number`	Estimated total source cost in USD. Keep it separate from the Toolmu quote and settled credits.
Estimated content retrieval costs`result.costDollars.contents`	`object`	Breakdown of estimated USD costs for requested content retrieval.
Estimated text extraction cost`result.costDollars.contents.text`	`number`	Estimated USD cost attributed to text extraction, not text length or a Toolmu credit count.

Additional notes

Keep text separate from interpretation

Ask the Agent to base its summary on the returned text and keep the source link. Reading a page does not verify every claim on it. If a passage is absent, state that it was not available in the returned text.

Read retrieval details with the page

Match any reported status to the requested URL or document identifier. A cache label does not provide the content's age; a crawl label does not establish when every passage was updated.

Source processing time uses milliseconds. Estimated source costs use USD and are separate from a Toolmu invoice or settled credit charge.

About Exa content retrieval

The official Exa contents documentation describes cache-first retrieval with fallback fetching when cached content is unavailable. Returned text can include Markdown or code formatting; it does not reproduce the original page layout.

Read Webpage Text from a Link | Exa | BlockRun

Read Webpage Text from a Link | Exa | BlockRun

You receive

Start with your task

What to know

Inputs & outputs

Input schema

Result schema

Keep text separate from interpretation

Read retrieval details with the page

About Exa content retrieval

Provide a public page link

Summarize an article

Answer a question from a page