MCP Tools Reference
Web Access

Web Access Tools

💡

All four tools are available on every plan, including Free.

web_search, web_browse, scrape, and web_fetch with screenshot or render_js run on our render/search cluster and draw down scraper credits — Free includes 250 MB of them, granted once and not topped up. Plain web_fetch is ordinary egress and is metered against your agent's bandwidth instead.

web_fetch

Fetch a URL through your Route6 IPv6 identity.

Any URL, on any plan. IPv4-only destinations are reached transparently via DNS64/NAT64 — you do not need to do anything, and there is no IPv6-only restriction on the free tier.

What differs by plan is volume: Free is capped at 250 MB of egress per agent per month, paid plans are unmetered.

ParameterTypeRequiredDescription
urlstringURL to fetch
method"GET" | "POST" | "PUT" | "DELETE" | "HEAD"HTTP method (default: "GET")
headersobjectCustom HTTP headers
bodystringRequest body (for POST/PUT)
max_lengthnumberMax response characters (default: 50000)
render_jsbooleanRender JavaScript via headless browser (paid)
screenshotbooleanReturn a base64 PNG screenshot instead of content (paid)

When screenshot: true, returns an image content block (base64 PNG, 1280×720).

web_search

Search the web via a metasearch API (aggregates Google, Bing, DuckDuckGo, and 70+ engines).

ParameterTypeRequiredDescription
querystringSearch query
num_resultsnumberNumber of results (default: 10, max: 50)
languagestringLanguage code (default: "en")
categoriesstringgeneral, images, news, science, or files

web_browse

Interactive headless-browser session. Navigate, click, type, scroll, and extract content.

ParameterTypeRequiredDescription
urlstringStarting URL
actionsarrayOrdered list of browser actions

Each action object has:

  • action: "click" | "type" | "scroll" | "extract" | "wait" | "navigate"
  • selector: CSS selector (for click/type/extract)
  • text: text to type (for type)
  • direction: "up" | "down" (for scroll)
  • amount: scroll amount (for scroll)
  • url: target URL (for navigate)

scrape

Scrape structured content from a URL, or manage scraper credits.

ParameterTypeRequiredDescription
urlstringURL to scrape (mutually exclusive with action)
selectorstringCSS selector for targeted extraction (used with url)
action"balance" | "topup"balance to check credits, topup to purchase (mutually exclusive with url)
pack"starter" | "pro" | "agency"Credit pack (required when action=topup)

Examples:

scrape(url=https://example.com, selector=.article-body)
scrape(action=balance)
scrape(action=topup, pack=starter)

plan_upgrade

Get a Stripe Checkout URL to upgrade your Route6 plan. No parameters.

Returns a checkout URL — direct your user there to complete payment. Available on all tiers.