SkylineWebcams Live Webcams Scraper

Refresh a travel-content catalogue of skylinewebcams live webcams by country. Export camera names, short descriptions, exact camera pages, displayed host credits and short-lived observed HLS manifest URLs when the public page exposes them.

The Actor returns one camera per dataset row and a country-grouped JSON file. It does not record video, download snapshots, verify playback, or guarantee permanent stream links. A missing or unavailable stream does not discard otherwise useful catalogue metadata.

Who is this for?

• Travel-content teams refreshing destination camera catalogues.
• Developers maintaining country-indexed webcam discovery lists.
• Data teams comparing dated catalogue snapshots in their own storage.

Why use this Actor?

Country codes, stable page links and observation timestamps make repeat exports easier to join. Displayed host credits remain attached to their camera instead of being lost during manual copying. A small manifest request distinguishes an observed HLS response from a URL that was merely present in HTML.

This is a catalogue product, not a licensed media feed or weather service. It collects only public SkylineWebcams pages, without customer credentials or paid proxy configuration.

What data is extracted?

Field	Meaning
countryCode	ISO 3166-1 alpha-2 code, for example IT
countryName	Normalized country name
sourceCountryName	Country/territory label displayed by the source
cameraId	Public player identifier, nullable
name, description	Displayed camera heading and short attraction description
pageUrl	Exact public English camera-detail page
hostCredits	Displayed host labels and source credit links; empty if absent
streamUrl	Observed short-lived HLS manifest URL, nullable
streamStatus	manifest_available, manifest_unavailable, or not_exposed
streamObservedAt, scrapedAt	ISO observation and extraction timestamps
attributionNote	Reminder that credits do not establish reuse permission

manifest_available means the fetched response begins with #EXTM3U. It does not mean media segments played, that a camera is currently live, or that the URL will work from your network.

How to get started

1. Open the Actor and select country codes, initially IT.
2. Set the global maximum records and per-country maximum.
3. Start the run.
4. Open the camera dataset or the country-grouped JSON link in Output.
5. Export JSON, CSV or Excel through Apify's dataset tools.

For country discovery:

{"countries":["IT","MT"],"maxItems":20,"maxItemsPerCountry":10}

For exact camera pages, clear the country selection:

{
  "countries":[],
  "startUrls":[{"url":"https://www.skylinewebcams.com/en/webcam/italia/lazio/roma/fontana-di-trevi.html"}],
  "maxItems":1,
  "maxItemsPerCountry":1
}

Input parameters

Input	Default	Behavior
countries	["IT"]	Up to 10 supported uppercase ISO country codes, deduplicated in order
startUrls	[]	Up to 50 exact HTTPS English camera-detail pages; mutually exclusive with nonempty countries
maxItems	10	Global unique camera-row limit, 1–100
maxItemsPerCountry	10	Country-row limit across either route, 1–20

The schema lists supported countries. This reflects the current public directory, not coverage of every country worldwide. GB is the United Kingdom code. The source's Zanzibar category maps to TZ (Tanzania); it does not cover all Tanzanian cameras. Country codes are ISO 3166-1 alpha-2, not subdivision codes from ISO 3166-2.

For startUrls, only url is consumed. Custom method, headers, payload and userData are ignored and not applied to requests, even if a hosted MCP client exposes those standard request-list options. Camera pages use the Actor's own GET requests.

Earlier countries can consume the global limit. Set a per-country cap for a more balanced collection. Limits are row counts, not counts of working streams. Duplicate URLs are removed; unavailable-stream records still count. No keyword, city, location search, scene filter or unlimited mode is implemented.

Output example

A real camera-page result has this structure; the expiring URL token is omitted here:

{
  "countryCode":"IT",
  "countryName":"Italy",
  "sourceCountryName":"Italy",
  "cameraId":"286",
  "name":"Trevi Fountain - Rome",
  "description":"View of the Trevi Fountain in Rome",
  "pageUrl":"https://www.skylinewebcams.com/en/webcam/italia/lazio/roma/fontana-di-trevi.html",
  "hostCredits":[{"name":"Webcam host","creditUrl":"https://www.skylinewebcams.com/click.php?l=..."}],
  "streamUrl":"https://hd-auth.skylinewebcams.com/live.m3u8?a=...",
  "streamStatus":"manifest_available",
  "streamObservedAt":"2026-10-03T14:00:00.000Z",
  "scrapedAt":"2026-10-03T14:00:00.000Z",
  "attributionNote":"Displayed host credits only; this record does not establish reuse permission or a complete legal attribution requirement."
}

Values and availability change. Some host images have only the generic label “Webcam host”; the Actor preserves that label rather than inventing a company name.

Country-grouped JSON

The default key-value store's COUNTRIES record contains:

[{"countryCode":"IT","countryName":"Italy","webcams":[{"name":"Trevi Fountain - Rome","pageUrl":"..."}]}]

Each webcam contains the full dataset fields; this shortened illustration shows the grouping only. The observation summary reports saved rows, country codes and manifest availability. Both JSON files are included in the camera-record price. If a later page fails, existing rows and the group file are partial; check run status before treating them as a complete export.

How much does it cost to export SkylineWebcams camera catalogues?

Pay-per-event pricing charges a $0.005 start fee per run, plus each unique camera catalogue record saved. Unavailable-stream metadata records are also charged; failed pages and duplicates are not.

Plan tier	Per camera record
FREE	$0.0092
BRONZE	$0.008
SILVER	$0.00624
GOLD / PLATINUM / DIAMOND	$0.0048

BRONZE run totals including the start fee: 1 camera — $0.013; 10 cameras — $0.085; 100 cameras — $0.805. Your applicable plan tier determines the per-record price. Use Apify's maximum charge per run to control spending; a billing cap can stop collection before the requested row limit.

Integrations and recurring refreshes

Save the input as an Apify Task and attach an Apify schedule for recurring exports. Join snapshots by pageUrl or cameraId in a spreadsheet, database or warehouse. Compare names, descriptions and credit links, not expiring stream tokens as permanent identifiers.

Connect completion webhooks to your ingestion pipeline, or use Apify integrations to send dataset rows to Google Sheets. Retain full JSON for nested host credits and country grouping. This Actor does not itself store historical changes, send change alerts or monitor cameras between runs.

API examples

Use a managed Apify token; do not put credentials into Actor input.

curl -X POST "https://api.apify.com/v2/acts/automation-lab~skyline-tourist-webcam-directory/run-sync-get-dataset-items" \
  -H "Authorization: Bearer $APIFY_TOKEN" -H "Content-Type: application/json" \
  -d '{"countries":["IT"],"maxItems":10,"maxItemsPerCountry":10}'

JavaScript with apify-client:

const client = new ApifyClient({token: process.env.APIFY_TOKEN});
const run = await client.actor('automation-lab/skyline-tourist-webcam-directory').call({countries:['IT'], maxItems:10});
const {items} = await client.dataset(run.defaultDatasetId).listItems({limit:10});

Python with apify-client:

import os
from apify_client import ApifyClient
client = ApifyClient(os.environ['APIFY_TOKEN'])
run = client.actor('automation-lab/skyline-tourist-webcam-directory').call(run_input={'countries':['IT'], 'maxItems':10})
items = client.dataset(run['defaultDatasetId']).list_items(limit=10).items

Use with MCP

Select the Actor-specific endpoint and authenticate your client according to Apify's MCP documentation:

$claude mcp add --transport http apify "https://mcp.apify.com?tools=automation-lab/skyline-tourist-webcam-directory"

For Claude Desktop, Cursor, and VS Code clients supporting HTTP MCP:

{"mcpServers":{"apify":{"url":"https://mcp.apify.com?tools=automation-lab/skyline-tourist-webcam-directory"}}}

Example prompt: “Export ten Italian SkylineWebcams camera catalogue records and summarize the displayed host credits and stream observation statuses.” Country and row-limit fields determine the collection; prompt wording does not add unsupported search filters.

Discover actual tool names and schemas with scoped tools/list. Actor selection also exposes run/dataset helpers; it is not exactly one tool. Start an authorized run once and retain its ID, status and storage IDs (metadata, not result rows). The server waits at most waitSecs 0–45 seconds (default 30), so completion is not guaranteed. Configure the client request timeout above the chosen wait plus transport margin, for example 40 seconds for a 30-second wait, within the total deadline below. If a client timeout occurs before a run ID is known, report that run creation/status is uncertain; do not blindly restart. If the ID is known, recover/check that same run. If it is nonterminal, check that same run with bounded 2/4/8-second backoff capped at 10 seconds and a 120-second deadline; do not restart to poll after a client timeout. At the deadline report pending with the run ID.

After success, use get-dataset-items with the returned defaultDatasetId, explicit limit: 20, offset: 0 and a fields projection naming countryCode, countryName, name, description, pageUrl, hostCredits, streamUrl, streamStatus, streamObservedAt, scrapedAt and attributionNote. Use the field-list encoding and dataset-ID argument exposed by scoped tools/list. Retrieve subsequent untransformed pages by source offset, at most 100 rows and 64 KiB of serialized content admitted into model context. Enforce that byte ceiling host-side, disclose omitted/oversized rows and the continuation offset, and keep full exports outside model context. If your client cannot intercept oversized responses, do not claim a hard byte guarantee. Terminal failures and any existing rows must be reported as partial.

Limits and reliability

Country listings are finite, changing source pages. The Actor reads their current server-rendered camera cards; it does not invent pagination or guarantee exhaustive source coverage. It uses sequential HTTP requests with at most three attempts for temporary errors or empty responses. A stable HTTP 4xx or an unrecognizable page fails the run rather than producing a misleading empty success.

A missing HLS source is not_exposed; a failed or non-HLS manifest observation is manifest_unavailable. Those are useful availability observations, not proof of a page-fetch failure. URLs may expire quickly, require source headers or be unavailable from another geography. No media segments, snapshots, weather or coordinates are collected.

Legality, responsible use and attribution

Use data only where you have permission and comply with source terms and applicable law. Public accessibility does not grant a licence to rebroadcast, record, hotlink or redistribute streams. Displayed host credits are factual source labels, not a complete determination of required attribution. Obtain permission separately for your intended media use.

The Actor does not use AI to process your input or generate output. Extraction runs on Apify; HTTP requests go to SkylineWebcams and its HLS service. Dataset and grouped files remain in your Apify storage under your account retention settings; delete them there when no longer needed. No cross-run cache is maintained. For help, open an issue on the Actor's Apify page.

This independently operated Actor is not affiliated with or endorsed by SkylineWebcams. The icon is an original generated illustration, not a source logo.

Failed operations send sanitized diagnostic input, exceptions and Actor/build/run IDs to our private GlitchTip service for repair. Secret fields and URL queries are removed; reports are retained for 30 days.

FAQ and troubleshooting

Can I get six to ten working streams for every country? No. Row limits cap catalogue records, while the source controls country coverage and stream availability. Manifest observations are not playback verification.

Can I provide country or city pages in startUrls? No. Use country codes for discovery; explicit URLs must be English camera-detail pages. Clear countries when using them.

Why did a stream URL stop working? The anonymous source token can expire or depend on network/session conditions. Refresh the catalogue and treat URLs as short-lived observations.

Why did the run fail with some rows saved? A source page failed or changed after earlier rows were saved. Inspect logs and run status; rows are partial and should not be presented as a complete catalogue.

Related automation

For a separate destination-data workflow, OpenStreetMap POI Extractor exports map points of interest, not webcam streams. It is not a substitute or fallback for this source.