Screenshots of any public web page as PNG, JPEG, WebP, PDF or a scrolling video.
Scored across 2 tools
The two tools have entirely distinct purposes: list_devices only enumerates viewport presets, while take_screenshot performs the rendering. There is no realistic way to confuse them.
Both names follow a clean verb_noun snake_case pattern (list_devices, take_screenshot) with a single, consistent convention and no deviation.
Two tools is minimal, but the domain is narrow — one rendering operation plus a lookup helper for its argument — so each tool earns its place. It is slightly thin rather than genuinely under-scoped.
The core capability (rendering to image/PDF/scrolling video) and the preset enumeration are both present, covering the stated purpose. Minor gaps like quota/account inspection are not strictly necessary but would round out the surface.
List the device presets accepted by the viewport_device argument of take_screenshot. Needs no access key.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint, idempotentHint and destructiveHint=false, so the safety profile is covered. The description adds a genuinely non-obvious trait beyond them — 'Needs no access key' — which tells the agent this call is unauthenticated and safe to make freely. It does not describe the shape of the returned list.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two short sentences, no filler. The purpose is front-loaded and the auth note is a compact trailing clause.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a no-argument, read-only enumerator with no output schema, the description supplies enough: what it lists, why it exists, and that no credentials are required. Only the exact return format (plain string array vs. objects with labels) is left unspecified.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Zero parameters, so the baseline is 4. The description still earns its place by explaining the semantics of the tool's output — that the returned presets are the valid domain for take_screenshot's viewport_device argument — which is information no schema provides here.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Names a specific verb and resource ('List the device presets') and pins down exactly which values it returns — the ones accepted by take_screenshot's viewport_device argument. The sibling relationship is explicit, so an agent can distinguish it from take_screenshot immediately.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The tie to 'the viewport_device argument of take_screenshot' makes the usage context clear: call this to discover valid values before screenshotting. It lacks an explicit instruction (e.g. 'call before take_screenshot') or any stated exclusions, so it stops short of a 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Render a public web page and return it as an image (or a PDF, or a video of the page scrolling). Uses one screenshot of the account quota; cache hits and failed renders are free. Prefer jpeg or webp to keep the image small. A result over 6 MB comes back as a download link (valid for about a day) instead of inline data.
| Name | Required | Description | Default |
|---|---|---|---|
| url | Yes | Full address of the page, starting with http:// or https://. | |
| cache | No | Reuse a stored copy when the same request was made before. | |
| delay | No | Extra seconds to wait after the page has loaded. | |
| format | No | File format of the result. mp4, webm and gif record a video of the page scrolling from top to bottom. | jpeg |
| selector | No | Capture only the first element matching this CSS selector. | |
| block_ads | No | Block requests to ad networks. | |
| dark_mode | No | Ask the page for its dark theme. | |
| full_page | No | Capture the whole page, not just the first screen. | |
| wait_until | No | Page event that marks the page as loaded. | load |
| viewport_width | No | Width of the browser window in pixels. | |
| viewport_device | No | Emulate a device preset (see list_devices). | |
| viewport_height | No | Height of the browser window in pixels. | |
| block_cookie_banners | No | Hide cookie consent banners. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already cover the safety profile (readOnly, non-destructive, open-world), yet the description adds genuinely useful operational detail: quota consumption per screenshot, free cache/failed renders, the 6 MB threshold that switches to a download link, and that link's ~1-day validity. It stops short of stating auth requirements or rate limits.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Four short sentences, front-loaded with the core action and output types, then quota, format preference, and size handling. No sentence is redundant and each carries distinct information.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a 13-parameter tool with full schema coverage and no output schema, the description covers the key runtime facts an agent needs (cost, size cap, format choice, download fallback). It leaves auth/permission expectations for private pages and any rate limiting unstated.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so the baseline is 3, but the description adds a tradeoff not present in the schema: jpeg/webp keep the image small. That gives the agent selection logic for the format enum beyond the schema's mechanical enumeration.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb and resource ('Render a public web page') and enumerates the output modes (image, PDF, scrolling video), which immediately distinguishes it from the only sibling, list_devices. The 'public' qualifier also frames the scope precisely.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Offers concrete guidance: prefer jpeg or webp for smaller images, and the note that cache hits and failed renders don't consume quota helps the agent reason about retries. The 'public web page' wording implicitly excludes authenticated pages, but there is no explicit when-not statement or routing to alternatives.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
list_devicestake_screenshotScreenshot any website as PNG, JPEG, WebP, PDF or scrolling MP4/GIF video, with public file links.
Pixel-perfect webpage screenshots rendered in a real browser, full-page or viewport, via one POST.
Screenshots and clean markdown of any public web page, rendered in a real Chrome browser.
Capture website screenshots, run bulk captures, and manage screenshot schedules and history.
Add one secure layer between your agents and this server.
