Firecrawl Map Capabilities

The Firecrawl map endpoint is the discovery-only counterpart to crawl: it enumerates candidate URLs on a site without retrieving full page bodies.

Role in the Ingestion Pipeline

Use map to understand the URL surface of a domain before spending crawl credits.
Use it when you need to choose bounds, refine include/exclude patterns, or estimate the scope of a docs section.
It complements firecrawl-crawling-capabilities rather than replacing it.

What Map Returns

In the vault's mental model, map returns URL discovery results only:

candidate URLs
enough structure to decide what to crawl next
no markdown body extraction

That makes map the lowest-cost recon step in the Firecrawl family represented here.

Practical Use

Start with map when the site structure is unknown.
Promote promising sections into a bounded crawl.
Use scrape when you already know the exact page you want.

Why Keep It Separate from Crawl

map answers: "What is here?"
crawl answers: "Fetch these pages and give me content."
scrape answers: "Fetch this one page now."

That distinction matters operationally because discovery, retrieval, and synthesis have different cost and risk profiles.

Known Parameters and Limits

The local pipeline spec does not provide the same level of parameter detail for map that it provides for scrape and crawl. Treat this note as a conceptual reference first, not an exhaustive parameter sheet.

Caveats

This rewrite is based on the pipeline spec's description of POST /v2/map as "Discover all URLs" plus surrounding ingestion design, not a live endpoint walkthrough.
If the vault begins using map programmatically, this note should be extended with concrete request and response examples from the implementation path.