Skip to main content
Glama
Website Screenshot Capture MCP Server

Website Screenshot Capture MCP Server

Capture website screenshots

capture_website_screenshots
Read-onlyIdempotent

Capture one or more webpages as PNG or JPEG screenshots, full-page or visible-window, and return image links with final URL, HTTP status, page title, and image size.

Instructions

Open each URL in a Chromium browser and save a PNG or JPEG screenshot, full page or window only. Returns one row per URL with the image link, final URL, HTTP status, page title, and image size. Pages that refuse the browser come back as free blocked rows.

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": "capture_website_screenshots"
  }
}

Input Schema

NameRequiredDescriptionDefault
urlsNoPages to capture, each starting with http:// or https://. A bare domain such as example.com gets https:// added. From Clay, put the column inside a one item list.
formatNopng (default) or jpeg. JPEG files are smaller.
delay_msNoExtra wait after the page loads and before the capture, 0 to 20000. Default 0.
full_pageNoyes captures the whole scrollable page, up to 16,000 pixels tall. no (default) captures the visible window only.
proxy_modeNoauto (default): pages load through the Apify proxy pool. none: pages load straight from the Apify server. A reliability setting only; it does not change what you pay.
wait_untilNoWhen the page counts as loaded. load (default): the page and its images loaded. domcontentloaded: the HTML is parsed, faster. networkidle: no network traffic for 0.5 s, best for pages that load content late.
jpeg_qualityNo1 to 100, used for jpeg only. Default 80.
block_trackersNoyes (default) skips analytics, ad, and session replay scripts and video or audio files, which draw nothing you need in the image and make pages slower. no loads everything, so ads appear.
hide_selectorsNoElements to hide before the capture, for example a cookie banner or a chat widget: #cookie-banner, .chat-widget.
max_charge_usdNoOptional. The run stops before a charge that would pass this amount. Summary rows are always returned.
viewport_widthNoBrowser window width, 320 to 3840. Default 1280.
timeout_secondsNoLongest wait for one page to load, 10 to 120. Default 60. A page still loading at this point is captured as it stands and marked partial.
viewport_heightNoBrowser window height, 240 to 2160. Default 800. With Full page on, the image is as tall as the page.
scroll_to_bottomNoyes scrolls down the page before the capture so images that load on scroll appear, then returns to the top. Default no.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv1.0.0

TDQS

A4/5.0
Behavior4/5

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

Annotations already declare readOnlyHint, idempotentHint, openWorldHint and destructiveHint=false, so the safety profile is covered. Beyond that, the description adds genuinely useful behavior: the browser is Chromium, the return row fields (image link, final URL, HTTP status, page title, image size), and that inaccessible pages yield 'free blocked rows'. It stops short of mentioning billing/concurrency behavior, which matters for a paid scraping run.

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?

Three tight sentences: action and format first, return contract second, edge-case failure behavior last. No filler, no repetition of the title, and the most decision-relevant facts are front-loaded.

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?

With 14 parameters and no output schema, the description does the important extra work of enumerating the returned row fields and the blocked-row case. What remains unstated is cost/charging behavior beyond the max_charge_usd parameter and any concurrency limits, which a caller running many URLs would want to know.

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% and every one of the 14 parameters carries a detailed description, including defaults, ranges and enum meanings. The description adds essentially no parameter-level detail beyond restating 'full page or window only' and the PNG/JPEG choice, so the baseline of 3 for schema-driven documentation applies.

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

Purpose5/5

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

The description states a specific verb and resource ('Open each URL in a Chromium browser and save a PNG or JPEG screenshot') and immediately narrows scope with 'full page or window only'. It also discloses the output shape and the failure mode, so an agent knows exactly what the tool produces without opening the schema.

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 implied by the description rather than stated: it is clear this is a screenshot-capture tool, but there is no explicit when-to-use/when-not-to-use guidance and no named alternative (none exist in the provided sibling list). The 'pages that refuse the browser come back as blocked rows' note gives partial context about expected failures but not about when to prefer or avoid this tool.

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

Deploy Server

Other Tools