The full data collection
pipeline in one place.

import alterlab

client = alterlab.Client(api_key="YOUR_KEY")

# Step 1: Map competitor site to discover all URLs
map_result = client.map("https://competitor.com", max_urls=2000)
blog_urls = [u for u in map_result.urls if "/blog/" in u]

# Step 2: Crawl all blog pages for full content
crawl = client.crawl(
    urls=blog_urls,
    formats=["markdown"],
    max_concurrent=10,
)

# Step 3: Search for competitor brand mentions
mentions = client.search(
    "competitor brand name site:reddit.com",
    num_results=20,
    scrape_results=True,
)

print(f"Crawled {len(crawl.results)} blog posts")
print(f"Found {len(mentions.organic_results)} brand mentions")

import alterlab

client = alterlab.Client(api_key="YOUR_KEY")

# Step 1: Map the full site structure
site_map = client.map("https://yoursite.com", max_urls=5000)
product_pages = [u for u in site_map.urls if "/products/" in u]

# Step 2: Search for keywords you want to rank for
serp = client.search(
    "best web scraping api 2026",
    engine="google",
    num_results=10,
    scrape_results=True,
)

# Step 3: Crawl your product pages to compare content
your_pages = client.crawl(
    urls=product_pages[:50],
    formats=["markdown", "metadata"],
)

# Analyze content gaps vs top-ranking pages
top_competitor_content = [r.content for r in serp.organic_results]
your_content = [p.markdown for p in your_pages.results]

import alterlab

client = alterlab.Client(api_key="YOUR_KEY")

# Step 1: Find authoritative sources via search
sources = client.search(
    "machine learning documentation tutorials",
    num_results=5,
)
source_domains = list({r.domain for r in sources.organic_results})

# Step 2: Map each source domain for full URL inventory
all_urls = []
for domain in source_domains:
    result = client.map(f"https://{domain}", max_urls=500)
    doc_urls = [u for u in result.urls if "/docs/" in u or "/guide" in u]
    all_urls.extend(doc_urls)

# Step 3: Crawl everything into Markdown
corpus = client.crawl(
    urls=all_urls,
    formats=["markdown"],
    max_concurrent=20,
)

print(f"Knowledge base: {len(corpus.results)} documents")
print(f"Total size: {sum(len(p.markdown) for p in corpus.results):,} chars")