Pullwire

Web Scraper

scrape

Read-only

Turn any public web page into clean, LLM-ready Markdown, raw HTML or JSON fields picked with CSS selectors. Optional JavaScript rendering, screenshots and an automatic residential-proxy fallback when a site blocks. Costs from 1 credit per result; failed items are free.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "scrape"
  }
}

Input Schema

Name	Required	Description	Default
`url`	No	The page to scrape.
`urls`	No	Several pages in one run (up to 20); one result per page.
`proxy`	No	auto: direct first, residential proxy only if the site blocks us. none: never a proxy. residential: always.	auto
`format`	No	markdown: clean, LLM-ready text of the page. html: the page HTML. json: the fields from `extract`.	markdown
`country`	No	Exit country for the residential proxy (two-letter code).
`extract`	No	For format json: field name -> CSS selector, or {selector, attribute, all}. attribute reads e.g. href/src/content ("html" = inner HTML); all returns every match.
`waitFor`	No	With renderJs: a CSS selector to wait for, or a delay in milliseconds (e.g. 2000, max 15000).
`renderJs`	No	Load the page in a headless browser first (for sites that build their content with JS).
`timeoutMs`	No	Per page, for loading (and rendering).
`screenshot`	No	With renderJs: a PNG of the viewport, returned as a signed download URL (valid 7 days).
`onlyMainContent`	No	Markdown: drop navigation, cookie banners, sidebars and footers.

Schema Changelog

Changes observed during successful MCP inspections.

First observedNov 11, 2026

TDQS

A3.8/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already state readOnlyHint, openWorldHint and destructiveHint=false, so safety is covered; the description adds genuinely new behavior: optional JS rendering, screenshot support, automatic proxy escalation, and a cost model ('1 credit per result; failed items are free'). It stops short of rate limits, concurrency, or what a batch failure looks like.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two sentences, zero filler, and the core capability is front-loaded before the modifiers and pricing note. Every clause carries information an agent cannot get elsewhere in the definition.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For an 11-parameter tool with a nested extraction schema and no output schema, the description covers what comes back (Markdown/HTML/JSON/screenshot URLs are mentioned in-schema) and how failures are billed. It leaves the batch result envelope and error semantics unstated, a modest gap.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so every parameter including the nested `extract` object is already documented in-schema. The description restates the format and proxy choices at a high level without adding syntax or edge-case meaning, so the baseline 3 applies.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description names a concrete action and resource ('Turn any public web page') and enumerates the three output shapes (Markdown, raw HTML, JSON via CSS selectors). This makes the contrast with the site-specific siblings (amazon-products, linkedin-company, zillow) reasonably clear, though it never states that distinction outright.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Usage is only implied: 'any public web page' suggests this is the general fallback when no specialized sibling applies, and 'residential-proxy fallback when a site blocks' hints at blocked-site scenarios. There is no explicit when-to-use/when-not statement, no prerequisite guidance, and no named alternative.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Pullwire tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?