firecrawl-search
firecrawl search
Via the Firecrawl MCP (preferred)
This plugin bundles the Firecrawl MCP server. When it's connected, prefer the native firecrawl_search tool over the CLI (and firecrawl_search_feedback to report result quality) — it needs no local install and authenticates through the plugin's one-time browser sign-in. The firecrawl CLI commands below are the fallback for when the MCP isn't connected.
Web search with optional content scraping. Returns search results as JSON, optionally with full page content.
Structured data via Alexandria
Signed-in search also returns matching Alexandria providers (official APIs, licensed publishers, Firecrawl indexes) in data.tools when you add "sources": ["web", "alexandria"] to firecrawl_search. A match is a lead, not data: read its contract with firecrawl_find_tools, then execute it with firecrawl_scrape and an alexandria body. Discovery is free; execution is billed at the capability's price. The full flow, the terms handling, and the CLI equivalents live in firecrawl-alexandria. Use it when the user needs records, listings, series, or a dataset rather than page content.
When to use
- You don't have a specific URL yet
- You need to find pages, answer questions, or discover sources
- First step in the workflow escalation pattern: search → scrape → map → crawl → interact