DEV Community

Cover image for How Can You Install Firecrawl CLI and Use Firecrawl CLI
Wanda
Wanda

Posted on • Originally published at apidog.com

How Can You Install Firecrawl CLI and Use Firecrawl CLI

#ai #cli #tutorial #webscraping

Firecrawl CLI is a unified terminal tool for developers and AI agents to scrape, search, map, crawl, and automate browsers on any website. Output formats include clean markdown, JSON, screenshots, and more—written directly to your filesystem. You can run Firecrawl CLI instantly with npx firecrawl (no install required) or install globally, then connect to Claude Code, Cursor, or OpenCode using firecrawl init for immediate agent integration.

Try Apidog today

Firecrawl CLI is essential if you need reliable, real-time web data without fragile custom scripts or blocked requests. It unifies scraping, search, site mapping, recursive crawling, and cloud browser sessions into a single terminal-native tool. Outputs are tailored for LLMs, keeping token counts low and context precise. Agents like Claude Code, Cursor, and OpenCode use Firecrawl CLI daily to fetch fresh content from JavaScript-heavy or protected sites where traditional tools fail.

đź’ˇ Tip: Before starting with Firecrawl CLI, download Apidog for free. Use it to visually test and debug Firecrawl API endpoints, including keys, custom params, and responses. This saves significant time when setting up or troubleshooting agent integrations.

With Firecrawl CLI, you set up your environment, install and authenticate, explore core commands, and integrate with agents. It handles concurrency, rate limits, and local caching automatically. Fine-tune outputs and efficiency with flags like format selectors and wait timers.

What Firecrawl CLI Delivers vs. Traditional Tools

  • Native JavaScript rendering: Uses cloud browsers to fetch content from complex or protected sites.
  • LLM-optimized output: Returns markdown stripped of boilerplate, minimizing token usage.
  • Filesystem-first: Writes files locally for easy bash-powered search and agent ingestion.
  • Unified commands: Combines scrape, search, map, crawl, and browser automation—no need for separate libraries or proxies.
  • Efficient flags: Options like --only-main-content yield cleaner, cheaper results.

Prepare Your Environment

  1. Check Node.js (required ≥18): 
   node --version
Enter fullscreen mode Exit fullscreen mode

Update via your package manager or nvm if needed.

  1. Create a workspace directory: 
   mkdir firecrawl-cli-projects && cd firecrawl-cli-projects
Enter fullscreen mode Exit fullscreen mode

Keep outputs organized and ready for git versioning.

  1. (Optional) Disable telemetry: 
   export FIRECRAWL_NO_TELEMETRY=1
Enter fullscreen mode Exit fullscreen mode

Fastest Install: Init Method for Agents

Install, authenticate, and add agent skills in one command:

npx -y firecrawl-cli@latest init --all --browser
Enter fullscreen mode Exit fullscreen mode
  • Logs you in via browser, stores your API key, and sets up Claude Code, Cursor, and compatible agents.
  • Restart your agent afterward to detect the new CLI capability.

Permanent Install: Global NPM

For ongoing use across projects:

npm install -g firecrawl-cli
firecrawl --version
Enter fullscreen mode Exit fullscreen mode
  • Firecrawl CLI now runs instantly from any directory.

Authenticate and Check Configuration

  1. Authenticate: 
   firecrawl login
Enter fullscreen mode Exit fullscreen mode

Or set your API key manually:

   export FIRECRAWL_API_KEY=fc-your-key-here
Enter fullscreen mode Exit fullscreen mode
  1. Check CLI status: 
   firecrawl --status
   firecrawl view-config
Enter fullscreen mode Exit fullscreen mode
  • Shows credits, concurrency, and auth state.
  • Switch accounts: firecrawl logout then login again.
  • For self-hosted use: --api-url http://localhost:3002

Scraping Content

Extract content from any URL:

firecrawl scrape https://example.com --only-main-content
Enter fullscreen mode Exit fullscreen mode
  • Outputs clean markdown.
  • Save to file: 
  firecrawl scrape https://example.com --only-main-content -o output.md
Enter fullscreen mode Exit fullscreen mode

Request multiple formats:

firecrawl scrape https://example.com --format markdown,json,html,links,images --pretty
Enter fullscreen mode Exit fullscreen mode

Take screenshots:

firecrawl scrape https://example.com --screenshot
Enter fullscreen mode Exit fullscreen mode

or

firecrawl scrape https://example.com --full-page-screenshot
Enter fullscreen mode Exit fullscreen mode

Handle slow loaders:

firecrawl scrape https://example.com --wait-for 5000
Enter fullscreen mode Exit fullscreen mode

Precise filtering:

firecrawl scrape https://docs.example.com --include-tags main,article --exclude-tags nav,footer,script
Enter fullscreen mode Exit fullscreen mode

Benchmark performance:

firecrawl scrape https://example.com --timing
Enter fullscreen mode Exit fullscreen mode

Performing Web Search

Search and scrape top results:

firecrawl search "latest AI agent benchmarks" --scrape --limit 8 --scrape-formats markdown
Enter fullscreen mode Exit fullscreen mode
  • Filter by recency: --tbs qdr:w
  • Filter by location or source type as needed.
  • Combine with browser sessions for deeper validation.

Mapping Websites

Discover URLs before deep extraction:

firecrawl map https://example.com -o sitemap.json
Enter fullscreen mode Exit fullscreen mode
  • Outputs structured lists. Chain filtered URLs into scrape or crawl commands.
  • Honors robots.txt and polite crawling by default.

Recursive Crawling

Crawl entire sites:

firecrawl crawl https://example.com --wait --progress -o crawl-output.json
Enter fullscreen mode Exit fullscreen mode
  • Controls for depth, max pages, and concurrency.
  • Real-time progress reporting.

Browser Automation

Start cloud browser sessions for interactive flows:

firecrawl browser launch-session
Enter fullscreen mode Exit fullscreen mode
  • Returns a session ID. Then: 
  firecrawl browser execute "open https://news.ycombinator.com" --session <id>
  firecrawl browser execute "click .titleline > a" --session <id>
  firecrawl browser execute "scrape" --session <id>
Enter fullscreen mode Exit fullscreen mode
  • Supports navigation, clicks, typing, and extraction after JS interactions.
  • Close sessions when done.

Advanced Configuration

Set persistent options:

firecrawl config --api-url https://your-custom-endpoint --concurrency 5
Enter fullscreen mode Exit fullscreen mode
  • Adjust output format, headers, and concurrency globally.
  • Export your API key in your shell profile for seamless sessions.

Integrate with AI Coding Agents

  • Install the skill once: 
  npx -y firecrawl-cli@latest init --all
Enter fullscreen mode Exit fullscreen mode
  • Agents auto-discover Firecrawl CLI.
  • CLI returns local file paths, not direct content—keeping agent context lean and efficient.

Troubleshooting

  • Auth issues: Re-run firecrawl login.
  • Rate limits: Lower concurrency or check your dashboard.
  • Empty results (JS-heavy sites): Increase --wait-for or use --only-main-content.
  • Diagnostics: Use --timing.
  • Switch keys: firecrawl logout and re-authenticate.

Best Practices

  • Always use --only-main-content for clean markdown.
  • Use descriptive filenames and dedicated folders.
  • Test on small scopes before full crawls.
  • Build pipelines: search → map → crawl.
  • Version-control output directories.
  • Monitor weekly credit usage for cost efficiency.

Enhance Firecrawl CLI Workflows with Apidog

Download Apidog for free and import Firecrawl endpoints (scrape, search, crawl, etc.) into collections. Apidog visualizes requests, stores your Firecrawl API key as a variable, mocks responses, and runs automated tests. Debug advanced Firecrawl CLI options or payloads before running in terminal. Combining Firecrawl CLI with Apidog ensures reliable, validated workflows.

Conclusion

You now have actionable steps to install, authenticate, scrape, search, map, crawl, and automate browser sessions with Firecrawl CLI. Use the init command, test a scrape, and iterate—Firecrawl CLI rewards careful flag usage and experimentation with superior results.

Download Apidog for free to supercharge your Firecrawl CLI testing and API validation. Set up Firecrawl CLI, leverage its full capabilities, and unlock real-time web automation.

Additional resources

Top comments (0)