Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    CurlShot Screenshots

    1

    Screenshot any web page from your agent: PNG, JPEG, WebP, PDF or scrolling video

    Free

    3 uses1.0 (1 review)
    CurlShot Screenshots

    CurlShot Screenshots

    Example session with this skill installed

    Take a full-page screenshot of https://example.com in dark mode and save it as a PNG. Use a desktop viewport of 1440x900.

    • Read your context and instructions
    • Compiled the curlshot screenshots
    • Generated the UI component

    I have captured the full-page screenshot of example.com using the CurlShot API. The request included dark mode enabled and a specific desktop viewport.

    Screenshot Capture Results

    ParameterValue
    URLhttps://example.com
    FormatPNG
    ModeDark Mode
    Viewport1440 x 900
    Full PageEnabled
    StatusSuccess
    File Savedscreenshot_example_dark.png
    
    # Command executed
    curl -sS -G "https://curlshot.com/api/v1/screenshot" \
      -H "X-Access-Key: $CURLSHOT_API_KEY" \
      --data-urlencode "url=https://example.com" \
      -d format=png \
      -d full_page=true \
      -d dark_mode=true \
      -d viewport_width=1440 \
      -d viewport_height=900 \
      --output screenshot_example_dark.png
    

    Next steps

    • Verify the screenshot_example_dark.png file in your current directory.
    • To capture a mobile version, try the viewport_device=iPhone 13 option.
    • Use block_cookie_banners=true if the site displays intrusive overlays.

    screenshot_example_dark.png

    TSX · React component

    Generated

    Example file from a real run - the skill writes it into your workspace.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    CurlShot is a screenshot API for developers and AI agents. This free skill teaches your agent to call it: one HTTP request turns a URL, HTML or Markdown into a PNG, JPEG, WebP, PDF or a scrolling video (MP4, WebM, GIF). Pages render in a real browser.

    What it does

    • Screenshot any public URL as PNG, JPEG or WebP
    • Full-page capture: lazy images are loaded on the way down
    • Phones, tablets and desktops from ready-made device presets, or your own window size
    • Clean captures: cookie banners and ads can be blocked before the capture
    • One element by CSS selector, dark mode, retina pixel density
    • Your own HTML or Markdown rendered to an image or PDF (social cards, reports)
    • Scrolling video of the page as MP4, WebM or GIF

    How it works

    1. Set CURLSHOT_API_KEY (a free key with 100 screenshots a month, no card: https://curlshot.com/register).
    2. Ask your agent for a screenshot. It sends one curl request to https://curlshot.com/api/v1/screenshot and saves the file.
    3. On an error the API returns JSON with error_code and error_message; a misspelled option is rejected with a hint, never ignored.

    MCP alternative

    Agents with MCP support can use the hosted server instead: https://curlshot.com/api/mcp (streamable HTTP), tools take_screenshot and list_devices.

    Good to know

    Public pages work out of the box. For pages behind a login, send cookies or headers with the request. Local and private network addresses are refused: send the HTML instead. One successful render is one screenshot; failed renders and cache hits are free.

    Docs: https://curlshot.com/docs

    Requires CURLSHOT_API_KEY. Does not support localhost or private network URLs. Non-public pages must be sent as raw HTML.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    1.0
    1 review
    5
    0
    4
    0
    3
    0
    2
    0
    1
    1

    This skill has been used 3 times.

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    • Free with an Agensi account

    Requires

    Curlshot
    curl
    CurlShot API key (free plan available)

    Permissions

    Terminal / Shell
    Network Access

    Allowed Hosts

    curlshot.com
    Listed1 month ago

    What's inside

    Frequently Asked Questions

    This skill automates the complex process of capturing web content, handling difficult tasks like bypassing cookie banners, rendering JavaScript-heavy pages, and simulating different devices without requiring you to manage your own headless browser infrastructure.

    While it works via standard HTTP requests and `curl`, it is specifically optimized for Model Context Protocol (MCP) environments, allowing AI agents to seamlessly trigger web captures as a native tool during a conversation.

    You will receive the pre-configured tool definitions for your agent and documentation on integrating your CurlShot API key; however, please note that an active CurlShot API subscription is required and sold separately from the skill itself.

    Yes, you can send raw HTML or Markdown strings directly to the skill to generate styled images or PDFs, which is perfect for capturing private content that the API cannot access via a public URL.

    The skill supports several high-quality formats including PNG, JPEG, and WebP for images, standard PDF for documents, and MP4, WebM, or GIF for scrolling video captures.

    New in APIs & Backend