Migrate from Firecrawl to Skryp

# The official Firecrawl Python SDK, unchanged, pointed at Skryp: only the API URL and the key differ.
# pip install firecrawl-py    export SKRYP_API_KEY=...
import os

from firecrawl import Firecrawl

app = Firecrawl(api_key=os.environ["SKRYP_API_KEY"], api_url=os.environ.get("SKRYP_API_URL", "https://api.skryp.dev"))

doc = app.scrape("https://quotes.toscrape.com/", formats=["markdown", "links"])
print(f"scrape: {doc.metadata.title!r}, {len(doc.markdown)} characters of Markdown, {len(doc.links)} links")

site = app.map("https://books.toscrape.com/", limit=20)
print(f"map: {len(site.links)} URLs, first {site.links[0].url}")

found = app.search("books to scrape", limit=3)
print(f"search: {[r.url for r in found.web]}")

crawl = app.crawl("https://quotes.toscrape.com/", limit=3, poll_interval=2)
print(f"crawl: {crawl.status}, {len(crawl.data)} pages")

batch = app.batch_scrape(["https://quotes.toscrape.com/page/1/", "https://quotes.toscrape.com/page/2/"], formats=["markdown"], poll_interval=2)
print(f"batch scrape: {batch.status}, {len(batch.data)} pages")

usage = app.get_credit_usage()
print(f"credit usage: {usage.remaining_credits} credits remaining")

// The official Firecrawl JavaScript SDK, unchanged, pointed at Skryp: only the API URL and the key differ.
// npm install @mendable/firecrawl-js    export SKRYP_API_KEY=...
import Firecrawl from "@mendable/firecrawl-js";

const app = new Firecrawl({ apiKey: process.env.SKRYP_API_KEY, apiUrl: process.env.SKRYP_API_URL ?? "https://api.skryp.dev" });

const doc = await app.scrape("https://quotes.toscrape.com/", { formats: ["markdown"] });
console.log(`scrape: ${JSON.stringify(doc.metadata.title)}, ${doc.markdown.length} characters of Markdown`);

const site = await app.map("https://books.toscrape.com/", { limit: 20 });
console.log(`map: ${site.links.length} URLs`);

const found = await app.search("books to scrape", { limit: 3 });
console.log(`search: ${found.web.map((r) => r.url).join(", ")}`);

const crawl = await app.crawl("https://quotes.toscrape.com/", { limit: 3, pollInterval: 2 });
console.log(`crawl: ${crawl.status}, ${crawl.data.length} pages`);

# The rest of the Firecrawl v2 surface through the official SDK: other formats, page actions, extract, cancel.
# pip install firecrawl-py    export SKRYP_API_KEY=...
import os
import re

from firecrawl import Firecrawl

app = Firecrawl(api_key=os.environ["SKRYP_API_KEY"], api_url=os.environ.get("SKRYP_API_URL", "https://api.skryp.dev"))
URL = "https://quotes.toscrape.com/"


def check(name, fn):
    try:
        print(f"ok    {name}: {fn()}")
    except Exception as e:  # report and carry on, so one gap does not hide the others
        print(f"FAIL  {name}: {type(e).__name__}: {str(e)[:160]}")


check("html", lambda: f"{len(app.scrape(URL, formats=['html']).html)} characters")
check("rawHtml", lambda: f"{len(app.scrape(URL, formats=['rawHtml']).raw_html)} characters")
check("images", lambda: f"{len(app.scrape('https://books.toscrape.com/', formats=['images']).images)} image URLs")
check("screenshot", lambda: app.scrape(URL, formats=["screenshot"]).screenshot[:40] + "…")
check("json with a prompt", lambda: app.scrape(URL, formats=[{"type": "json", "prompt": "The first quote's text and author"}]).json)
def page_two_by_click():
    doc = app.scrape(
        "https://quotes.toscrape.com/js/",
        formats=["markdown"],
        actions=[{"type": "wait", "milliseconds": 1000}, {"type": "click", "selector": "li.next a"}, {"type": "wait", "milliseconds": 1000}],
    )
    return f"after clicking Next the browser is on {doc.metadata.url}"


check("actions", page_two_by_click)
check("extract", lambda: app.extract(urls=[URL], prompt="The authors of the first three quotes").data)


def country():
    doc = app.scrape("https://ipinfo.io/json", formats=["markdown"], location={"country": "DE"}, proxy="enhanced")
    exit_ = doc.metadata.egress  # Skryp adds the verified exit to Firecrawl's metadata
    seen = re.search(r'"country": "(\w\w)"', doc.markdown).group(1)
    return f"exit {exit_['country'].upper()} (verified: {exit_['verified']}), and ipinfo.io saw a request from {seen}"


check("location.country + proxy=enhanced", country)


def start_and_cancel():
    job = app.start_crawl("https://books.toscrape.com/", limit=50)
    return f"crawl {job.id} cancelled: {app.cancel_crawl(job.id)}"


check("cancel crawl", start_and_cancel)