agentleFS
DevinoSolutionsDevinoSolutions/claude-plugins/skills/capturing-web-pages/SKILL.md

capturing-web-pages

Skill0 starsChanged 9 days agoMITView on GitHub

Captures a web page with upAPI's headless browser: a PNG or JPEG screenshot of the viewport or the full page, or a print-quality PDF of a URL or of raw HTML with paper size, margins and page ranges. Use when the user asks 'take a screenshot of this site', 'capture our pricing page on mobile width', 'save this page as a PDF', 'render this HTML invoice to PDF', or 'how does this page look'. Read-only; each capture costs 20 units of upAPI quota.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Every tool in this skill comes from the upapi MCP server, the upAPI connector, and is written here as upapi:<tool>.

    line 13
  2. Very specific:

    Tools listed by name (the plugin's ?tools=claude surface): call them directly, for example upapi:screenshot_post.

    line 15
  3. Distinctive:

    Call the tool once. Each call costs 20 units, so do not take several variations unless the user asked for them.

    line 29
  4. Very specific:

    If the screenshot looks unfinished (blank areas, spinners), offer one retry with waitUntil: "networkidle" or a delayMs of 500 to 2000.

    line 33

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md
<!-- Generated from plugins/upapi/skills/capturing-web-pages/SKILL.md by scripts/sync-skills.mjs. Edit the plugin copy, then run the script. -->

upAPI: capturing web pages

Render a page in a real browser and hand the user the image or PDF they asked for.

Connecting and calling the tools

Every tool in this skill comes from the upapi MCP server, the upAPI connector, and is written here as upapi:<tool>.

  • Tools listed by name (the plugin's ?tools=claude surface): call them directly, for example upapi:screenshot_post.
  • Only upapi:search_ops and upapi:call_op listed (the plain URL's compact table): call upapi:call_op with the slug and input, for example { "slug": "screenshot.post", "input": { "url": "https://example.com" } }.
  • No upapi tools at all: upAPI is not connected yet. Follow the setup-upapi skill.

A failed operation comes back as an error result whose text starts with its code, for example INVALID_INPUT:.

Tools you will use

  • upapi:screenshot_post (slug screenshot.post, 20 units): renders a public page in headless Chromium. Input: url (required), fullPage (default false), format (png default, or jpeg with quality 1 to 100), width (320 to 3840, default 1280), height (320 to 2160, default 800), waitUntil (load default, domcontentloaded, or networkidle for lazy-rendered apps), delayMs (0 to 5000), blockAds. Returns image as base64, plus format, width, height, bytes and finalUrl.
  • upapi:html_to_pdf_post (slug html-to-pdf.post, 20 units): renders a page or a raw HTML document to PDF. Input: exactly one of url or html (up to 2 MB, with absolute or inline assets), plus format (A4 default, Letter, Legal), landscape, margin (top, right, bottom, left as CSS lengths), printBackground (default true), scale (0.1 to 2), pageRanges (for example 1-5, 8), media (screen default, or print for the site's print stylesheet). Returns pdf as base64, plus pages and bytes.

Workflow

  1. Confirm what to capture: the URL (or the HTML), and for a screenshot the width (for example 390 for a phone, 1280 for a laptop) and whether the whole page or just the first screen. For a PDF, the paper size and orientation if the user cares.
  2. Call the tool once. Each call costs 20 units, so do not take several variations unless the user asked for them.
  3. Deliver the result:
    • Claude Code: decode the base64 field and write it to a file the user names, or a clear default such as screenshot.png or page.pdf in the working directory, then give the path. Never print the base64 into the chat.
    • claude.ai, Claude Desktop, Cowork: describe the capture (size in pixels or page count, bytes, finalUrl). Offer to save it as a file if the app can create one; do not paste the base64 string.
  4. If the screenshot looks unfinished (blank areas, spinners), offer one retry with waitUntil: "networkidle" or a delayMs of 500 to 2000.

Rules

  • Capture only public pages the user asked for. URLs resolving to private, loopback, link-local or cloud-metadata addresses are refused with INVALID_INPUT; do not try to work around it.
  • quality is only for jpeg; sending it with png is refused.
  • On RATE_LIMITED, wait the seconds the message gives before one retry. On QUOTA_EXCEEDED, stop and tell the user.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.