firecrawl-search

Search the web and extract full page content as JSON via the Firecrawl CLI.

Updated Jul 16, 2026
One-click install
npx skills add https://github.com/The-Focus-AI/walking-thoughts --skill firecrawl-search-the-focus-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: firecrawl-search
Source: https://github.com/The-Focus-AI/walking-thoughts/tree/main/.agents/skills/firecrawl-search
Command: npx skills add https://github.com/The-Focus-AI/walking-thoughts --skill firecrawl-search-the-focus-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Finding relevant pages and reading their full content usually requires separate search and scraping steps. This Skill combines web search with optional full-page markdown extraction in a single Firecrawl CLI command, returning structured JSON instead of bare snippets. ## Core Features & Use Cases - Web search with scraping: Run queries across web, images, and news sources, optionally scraping full page content for each result in one pass. - Developer-focused search: Query a dedicated index of GitHub issues, merged pull requests, READMEs, and documentation using the developer category or the firecrawl developer command. - Filtered and time-bounded queries: Narrow results by category, country, location, or recency (hour, day, week, month, year). - Use Case: When debugging an error message, run a developer-category search to surface matching GitHub issues and merged PRs with the exact matched passages, then read them with jq without re-scraping. ## Quick Start Ask the agent to search the web for your topic with firecrawl and save the results as JSON, for example: search for recent news about a library release and scrape the full article content.

Frequently Asked Questions about firecrawl-search

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I search the web and get full page content in one command?

Run firecrawl search with the --scrape flag to fetch full page content for every result alongside the search hits. Write output to a file with -o and --json, then extract fields with jq instead of re-scraping URLs.

How to search GitHub issues and pull requests for a programming error?

Use firecrawl search with --categories developer, or the dedicated firecrawl developer command, to query an index of GitHub issues, merged PRs, READMEs, and docs. Results include matched passages you can read directly with jq.

Can I filter Firecrawl search results by date or news sources?

Yes. Use --sources to restrict results to web, images, or news, and --tbs with values like qdr:d or qdr:w to limit results to the past day or week. You can also filter by country code and location.

When should I use firecrawl search instead of firecrawl scrape?

Use search when you do not have a specific URL and need to discover pages or answer a question. Use scrape when you already know the exact URL. Search is the first step in the search, scrape, map, crawl escalation pattern.

Why should search results be written to a file instead of stdout?

Writing results to a .firecrawl/ directory with -o keeps large JSON payloads out of the context window. You can then query the file with jq to extract only the URLs, titles, or passages you need.