设计与多媒体

Website Screenshot API - Full-Page PNG of Any URL

试用

Capture a full-page PNG screenshot of any public URL, returned inline as a base64 data:image/png URI ready to drop into an <img> tag. One endpoint, three capture tiers - normal and advanced at 1 credit, ultra for the hardest sites at 5. Only a successful capture is billed. Returns url, image, image_format, mode and image_bytes.

它能做什么

Give any public URL and get back a full-page PNG screenshot of the page, returned inline as a base64 URI ready to drop into an tag or decode to a file. Every capture is the full page, top to bottom.

技能文档

Website Screenshot via Scavio

Give any public URL and get back a full-page PNG screenshot of the page, returned inline as a base64 data:image/png URI ready to drop into an `` tag or decode to a file. Every capture is the full page, top to bottom.

When to trigger

Use this skill when the user asks to:

  • Screenshot a web page or capture a full-page image of a URL
  • Grab a visual snapshot / preview / thumbnail of a website
  • Turn a page into an image for a report, a slide, an email or a PDF
  • Feed a rendered page to a vision model or an OCR step as a PNG
  • Visually diff or archive how a page looks right now

Setup

Get a free API key at scavio.dev (50 free credits to get started, no card required):

export SCAVIO_API_KEY=sk_live_your_key

Every request is a POST with a JSON body and:

Authorization: Bearer $SCAVIO_API_KEY

Endpoints

Base URL: https://api.scavio.dev.

EndpointCreditsWhat it returns
POST /api/v1/extract/screenshot1-5A full-page PNG of the URL as a base64 data:image/png URI, plus its format, the capture tier used, and the image size in bytes

Parameters

ParameterTypeDefaultDescription
urlstringrequiredThe page to capture. http(s) only; a bare host like example.com is upgraded to https. Internal / loopback hosts are rejected. Always captured full-page
modestringnormalCapture tier: normal, advanced, or ultra

Capture tiers and their cost:

modeCreditsUse it for
normal1Most sites - the default
advanced1Pages that load their content a moment after the first response, so the capture waits longer before shooting
ultra5The hardest, most heavily protected sites that the lower tiers cannot reach

Start at normal and only step up if the returned image is blank or incomplete - the higher tiers cost more and are wasted on pages that do not need them.

Scope notes

  • The capture is always the full page, not just the visible viewport - long pages come back as tall images.
  • The image is returned inline as a data:image/png;base64,... URI, not a hosted link. There is no URL to expire; you hold the bytes.
  • Only a successful capture is billed. A page that cannot be captured (a 422) or a target that is not found (a 404) costs nothing.
  • image_bytes is the size of the decoded PNG, so you can budget before decoding.

Examples

import base64
import requests

BASE = "https://api.scavio.dev"
# Your key from https://scavio.dev. Load it from your environment or secret
# store in real code - keep it out of source control.
API_KEY = "sk_your_key_here"
HEADERS = {"Authorization": f"Bearer {API_KEY}"}

# 1. Capture a full-page screenshot (normal tier, 1 credit)
shot = requests.post(f"{BASE}/api/v1/extract/screenshot", headers=HEADERS,
    json={"url": "https://example.com"}).json()

data = shot["data"]
print(data["mode"], data["image_format"], data["image_bytes"], "bytes")

# 2. The image is a data:image/png;base64 URI - strip the prefix, decode it,
#    and write the PNG to disk.
b64 = data["image"].split(",", 1)[1]
with open("example.png", "wb") as f:
    f.write(base64.b64decode(b64))

# 3. A tougher, heavily protected site - step up the capture tier (ultra, 5 credits)
hard = requests.post(f"{BASE}/api/v1/extract/screenshot", headers=HEADERS,
    json={"url": "https://www.nike.com", "mode": "ultra"}).json()
print(hard["credits_used"], "credits used")

curl:

curl -s https://api.scavio.dev/api/v1/extract/screenshot \
  -H "Authorization: Bearer $SCAVIO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"url":"https://example.com","mode":"normal"}'

Response shape

Every response uses the envelope { data, response_time, credits_used, credits_remaining }.

{
  "data": {
    "url": "https://example.com/",
    "image": "data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAA7E...",
    "image_format": "png",
    "mode": "normal",
    "image_bytes": 16917
  },
  "response_time": 4255,
  "credits_used": 1,
  "credits_remaining": 4574
}

Guardrails

  • Cost tracks the tier: normal and advanced are 1 credit, ultra is 5. Only a successful capture is billed.
  • The capture is always the full page. If the user only wants the fold, crop the returned PNG yourself.
  • Never fabricate the image, its size or its contents. Only report what the API returned.
  • The image URI can be large. Decode it to a file rather than pasting the base64 into a prompt.

Failure handling

  • 400 means an invalid or missing url, or an unreachable internal/loopback host. Fix and retry.
  • 401 means the API key is invalid or missing. Check SCAVIO_API_KEY.
  • 404 means the target page was not found.
  • 422 means the page cannot be captured - try a higher mode, or accept that the site blocks capture.
  • 429 means rate or usage limit exceeded. Wait before retrying. See rate limits.
  • 502 / 503 mean the source is temporarily unavailable - wait a few seconds and retry.
  • If SCAVIO_API_KEY is not set, prompt the user to export it before continuing.

相关技能

把自然语言描述转为结构化 JSON,并由 mcp-diagram-generator MCP 服务生成 Draw.io、Mermaid 或 Excalidraw 图表文件。

作者 nssa.io1.0k 次安装47 星标

以 AI 机器人身份加入视频会议,提供语音、虚拟形象与屏幕共享四种模式。

作者 johnpatternai22 次安装8 星标

诊断生产力系统反复失效的根因,给出最小干预——容量测算、瓶颈定位、可靠的本地记录。

作者 Iván855 次安装69 星标

从 AdMapix API 拉取广告创意、应用、榜单和收入预估等数据,原样返回结构化 JSON。

作者 fly0pants4.3k 次安装296 星标

在本地磁盘以分类纯 Markdown 文件保存需要长期留存的事实,与智能体内置记忆并存。

作者 Iván559 次安装18 星标

scavio-ai 的更多技能

浏览全部技能

Pull a Facebook page's profile, posts, reels and photos, one post with its comments, a single reel/video with downloadable URLs, a public group and its posts, an event, and hashtag posts. 11 endpoints, 1 credit each, structured JSON.

作者 scavio-ai3 次安装

Search the Meta Ad Library by keyword or Facebook Page id and walk full cursor pagination through every ad, with creative, run dates, platforms and political spend. 3 endpoints, 1 credit per page.

作者 scavio-ai3 次安装

Search live or SOLD eBay listings, read a listing in full, and look up a seller's profile card as structured JSON. 3 endpoints, 1 credit each. Sold-listing search is the price-research feature eBay itself hides.

作者 scavio-ai1 次安装