Quick start

Go from an empty terminal to an Actor running on Scrape.ly, with results you can download, in about ten minutes. The whole flow:

npm install -g scrapely-cli
export SCRAPELY_API_URL=https://api.scrape.ly SCRAPELY_WEB_URL=https://my.scrape.ly
scrapely login
scrapely create my-first-actor --template crawlee-cheerio-js && cd my-first-actor
scrapely push
scrapely call -i '{"startUrls":[{"url":"https://crawlee.dev"}],"maxRequestsPerCrawl":10}' -o

You write code locally and the platform builds and runs it. The templates build on the Scrape.ly base images (scrapely/actor-node, scrapely/actor-python), which already contain the SDK, so you don't add it to package.json or requirements.txt, and you don't need Docker.

1. Install the CLI

The CLI is published as scrapely-cli and installs the scrapely executable.

npm install -g scrapely-cli
scrapely --version

Or skip the global install and use npx scrapely-cli <command>.

2. Log in

export SCRAPELY_API_URL=https://api.scrape.ly
export SCRAPELY_WEB_URL=https://my.scrape.ly
scrapely login
scrapely info

On an interactive terminal, login asks how to sign in:

The CLI checks the token against GET /v2/users/me and saves it to ~/.scrapely/credentials.json. scrapely info shows who you're logged in as. For CI, use an API key. See Authentication.

3. Create an Actor from a template

scrapely create my-first-actor --template crawlee-cheerio-js
cd my-first-actor

• The name needs at least three characters: lowercase letters, digits, and hyphens, not starting or ending with a hyphen.
• create with no arguments asks for the name, use case, language, and template.
• create copies the template into my-first-actor/, sets the Actor name in package.json and .actor/scrapely.json, makes a first Git commit, and runs npm install. Skip the last two with --skip-git-init and --skip-dependency-install.

The templates come from the platform catalog (scrapely templates ls); sources are in scrapely-app/actor-templates.

crawlee-playwright-js and crawlee-puppeteer-js still work as aliases for crawlee-playwright and crawlee-puppeteer-chrome. Browser templates launch Chromium in the container; to offload it, turn on the Scrape.ly browser for the Actor later.

4. Write your Actor

src/main.js reads the input, sets up the platform proxy, and runs the crawler. src/routes.js handles each page; this version also stores the first heading and the link count:

import { CheerioCrawler } from '@crawlee/cheerio';
import { Actor } from 'scrapely';

import { router } from './routes.js';

await Actor.init();

const { startUrls = ['https://crawlee.dev'], maxRequestsPerCrawl = 100 } = (await Actor.getInput()) ?? {};

const proxyConfiguration = await Actor.createProxyConfiguration({ checkAccess: true });

const crawler = new CheerioCrawler({
    proxyConfiguration,
    maxRequestsPerCrawl,
    requestHandler: router,
});

await crawler.run(startUrls);

await Actor.exit();

5. Push and build

scrapely push

The first push creates the Actor in your account, uploads the source as version 0.0 with build tag latest, writes the Actor ID into .actor/scrapely.json, then builds and waits. A failed build prints its log and exits non-zero. Later pushes update the same Actor and always rebuild.

6. Run it

scrapely call -i '{"startUrls":[{"url":"https://crawlee.dev"}],"maxRequestsPerCrawl":10}' -o

call starts a run on the platform and streams the log until it finishes. -o prints the dataset items when it succeeds.

7. Get the results

scrapely runs ls --desc          # this Actor's runs, newest first
scrapely runs info <runId>       # shows the run's Dataset ID
scrapely datasets get-items <datasetId> --format csv > results.csv

--format accepts json (default), jsonl, csv, html, rss, xml, and xlsx, written to stdout. To browse the log and dataset in the dashboard, open https://my.scrape.ly/actors/<actorId>, or use scrapely push --open.

Next steps

• Actor development academy: a longer, guided course.
• Actors: how builds, versions, runs, input, and output fit together.
• JavaScript SDK or Python SDK: everything Actor can do.
• Publishing: list the Actor in the Store and set its pricing.
• CLI reference: every command used here, and the rest.

Troubleshooting