### Quick Setup Firecrawl CLI

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-cli/rules/install.md

Installs the Firecrawl CLI globally, authenticates via browser, and installs all skills. This command is safe to re-run to update or fix the installation.

```bash
npx -y firecrawl-cli@1.16.2 init -y --browser
```

--------------------------------

### Install Skills and MCP with `firecrawl setup`

Source: https://context7.com/firecrawl/cli/llms.txt

Use `firecrawl setup` to install Firecrawl agent skills or the MCP server into detected editors. Skills enable AI agents to use the CLI and SDK. You can install globally or for specific agents like Cursor or Claude Code.

```bash
# Install skills to all detected AI agents (Cursor, Claude Code, Windsurf, etc.)
firecrawl setup skills
```

```bash
# Install skills to a specific agent only
firecrawl setup skills --agent cursor
```

```bash
firecrawl setup skills --agent claude-code
```

```bash
# Install globally (user-level, default)
firecrawl setup skills --global
```

```bash
# Install MCP server into all detected editors
firecrawl setup mcp
```

--------------------------------

### Initialize Firecrawl CLI with Browser Support

Source: https://context7.com/firecrawl/cli/llms.txt

Performs a full setup including installation, authentication via browser OAuth, and installation of agent skills. The `-y` flag bypasses confirmation prompts.

```bash
npx -y firecrawl-cli@latest init -y --browser
```

--------------------------------

### Setup Firecrawl Skills

Source: https://github.com/firecrawl/cli/blob/main/README.md

Manually installs Firecrawl skills for AI coding agents. Use --agent <agent> to scope installation to a specific editor.

```bash
firecrawl setup skills
```

--------------------------------

### Setup Firecrawl MCP Server

Source: https://github.com/firecrawl/cli/blob/main/README.md

Installs the Firecrawl MCP server into supported AI coding editors.

```bash
firecrawl setup mcp
```

--------------------------------

### AI Workflow: Claude Demo

Source: https://github.com/firecrawl/cli/blob/main/README.md

Launch the Claude AI workflow for a demonstration on a given URL. The Claude CLI needs to be installed.

```bash
firecrawl x claude demo https://resend.com
```

--------------------------------

### Interactive Website Download

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-download/SKILL.md

Initiates an interactive wizard to guide you through downloading a website, automatically selecting formats, screenshots, and paths.

```bash
firecrawl download https://docs.example.com
```

--------------------------------

### Comprehensive Website Download Configuration

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-download/SKILL.md

A full example combining multiple options for website download: including specific paths, excluding others, focusing on main content, capturing screenshots, and skipping confirmation.

```bash
firecrawl download https://docs.example.com \
  --include-paths "/features,/sdks" \
  --exclude-paths "/zh,/ja" \
  --only-main-content \
  --screenshot \
  -y
```

--------------------------------

### Start a Website Crawl

Source: https://github.com/firecrawl/cli/blob/main/README.md

Initiate a crawl of a website. The command returns a job ID, which can be used to check status or wait for completion.

```bash
firecrawl crawl https://example.com
```

--------------------------------

### Quick Start Firecrawl Command

Source: https://github.com/firecrawl/cli/blob/main/README.md

Executes a basic Firecrawl command. The CLI will prompt for authentication if it's not already configured.

```bash
firecrawl https://example.com
```

--------------------------------

### Manually Install Firecrawl CLI

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-cli/rules/install.md

Installs a specific version of the Firecrawl CLI globally using npm.

```bash
npm install -g firecrawl-cli@1.16.2
```

--------------------------------

### Verify Firecrawl CLI Installation and Auth

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-cli/rules/install.md

Performs a real request to ensure the CLI is installed, authenticated, and output works correctly. Creates a directory and scrapes a webpage, saving the output to a markdown file.

```bash
mkdir -p .firecrawl
firecrawl scrape "https://firecrawl.dev" -o .firecrawl/install-check.md
```

--------------------------------

### AI Workflow: Claude Competitor Analysis

Source: https://github.com/firecrawl/cli/blob/main/README.md

Launch the Claude AI workflow for competitor analysis on a given URL. Ensure the Claude CLI is installed.

```bash
firecrawl x claude competitor-analysis https://firecrawl.dev
```

--------------------------------

### Start AI Agent Job

Source: https://context7.com/firecrawl/cli/llms.txt

Launches an AI agent to perform a task, returning a job ID immediately. The job runs asynchronously.

```bash
firecrawl agent "Find the pricing plans for Firecrawl"
```

--------------------------------

### Example Vitest Test for executeScrape

Source: https://github.com/firecrawl/cli/blob/main/src/__tests__/README.md

This example demonstrates how to write a Vitest test for the `executeScrape` command. It mocks the Firecrawl client and verifies that the `scrape` method is called with the correct parameters and that the response is handled properly. Ensure `setupTest()` and `teardownTest()` are used in `beforeEach`/`afterEach` hooks.

```typescript
import { describe, it, expect, vi, beforeEach } from 'vitest';
import { executeScrape } from '../../commands/scrape';
import { getClient } from '../../utils/client';
import { setupTest, teardownTest } from '../utils/mock-client';

// Mock the client module
vi.mock('../../utils/client', async () => {
  const actual = await vi.importActual('../../utils/client');
  return {
    ...actual,
    getClient: vi.fn(),
  };
});

describe('executeScrape', () => {
  let mockClient: any;

  beforeEach(() => {
    setupTest();
    mockClient = { scrape: vi.fn() };
    vi.mocked(getClient).mockReturnValue(mockClient);
  });

  it('should call scrape with correct parameters', async () => {
    mockClient.scrape.mockResolvedValue({ markdown: '# Test' });

    await executeScrape({ url: 'https://example.com' });

    expect(mockClient.scrape).toHaveBeenCalledWith('https://example.com', {
      formats: ['markdown'],
    });
  });
});
```

--------------------------------

### Install Firecrawl CLI Globally

Source: https://context7.com/firecrawl/cli/llms.txt

Installs the Firecrawl CLI globally using npm. This command makes the `firecrawl` command available in your terminal.

```bash
npm install -g firecrawl-cli
```

--------------------------------

### Get All URLs from a Site

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-map/SKILL.md

Retrieve all URLs from a given website, with a limit on the number of URLs returned. The output is formatted as JSON and saved to a file.

```bash
firecrawl map "<url>" --limit 500 --json -o .firecrawl/urls.json
```

--------------------------------

### Start Asynchronous Website Crawl

Source: https://context7.com/firecrawl/cli/llms.txt

Initiates an asynchronous crawl job for an entire website, returning a job ID immediately. The job status is 'processing'.

```bash
# Start crawl (returns job ID immediately)
firecrawl crawl https://example.com
```

--------------------------------

### Crawl Website and Wait for Completion

Source: https://context7.com/firecrawl/cli/llms.txt

Starts a website crawl and waits for the job to complete before returning. Useful for sequential processing.

```bash
# Wait for completion (blocks until done)
firecrawl crawl https://example.com --wait
```

--------------------------------

### Basic Firecrawl CLI Commands

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-cli/SKILL.md

Examples of using `firecrawl search` and `firecrawl scrape` with output redirection. Ensure URLs are quoted to prevent shell interpretation issues.

```bash
firecrawl search "react hooks" -o .firecrawl/search-react-hooks.json --json
```

```bash
firecrawl scrape "<url>" -o .firecrawl/page.md
```

--------------------------------

### AI Workflow: Codex Competitor Analysis (Coming Soon)

Source: https://github.com/firecrawl/cli/blob/main/README.md

Placeholder for the upcoming Codex AI workflow to perform competitor analysis. Requires the Codex CLI to be installed.

```bash
firecrawl x codex competitor-analysis https://crawlee.dev
```

--------------------------------

### AI Workflow: Claude QA

Source: https://github.com/firecrawl/cli/blob/main/README.md

Execute the Claude AI workflow for question answering on a specific URL. Ensure the Claude CLI is installed.

```bash
firecrawl x claude qa https://myapp.com
```

--------------------------------

### AI Agent with Schema from File

Source: https://context7.com/firecrawl/cli/llms.txt

Uses a JSON schema defined in a local file to guide the AI agent's data extraction. Specify the schema file with `--schema-file`.

```bash
firecrawl agent "Extract product catalog" \
  --schema-file ./product-schema.json \
  --wait -o products.json --pretty
```

--------------------------------

### Crawl with Custom Scrape Options

Source: https://context7.com/firecrawl/cli/llms.txt

Starts a crawl with custom scraping options applied per page, such as specifying formats and content extraction rules. Options can be provided as a JSON string.

```bash
# Crawl with custom scrape options per page
firecrawl crawl https://example.com \
  --scrape-options '{"formats":["markdown"],"onlyMainContent":true}' \
  --wait --limit 100
```

--------------------------------

### Run Deep Research Workflow

Source: https://github.com/firecrawl/cli/blob/main/src/commands/experimental/README.md

Starts a deep research workflow on a given topic. The `--depth` option controls the number of sources consulted, and `-o` specifies the output format. Use `-y` to auto-approve tool permissions.

```bash
firecrawl claude deep-research
```

```bash
firecrawl claude deep-research "AI code generation landscape" --depth exhaustive -y
```

```bash
firecrawl claude deep-research "WebSocket alternatives" --depth quick -o markdown -y
```

--------------------------------

### Get Markdown and Links

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-scrape/SKILL.md

This command extracts both markdown content and links from a URL. The output is saved as a JSON file containing both formats.

```bash
firecrawl scrape "<url>" --format markdown,links -o .firecrawl/page.json
```

--------------------------------

### AI Workflow: OpenCode Deep Research (Coming Soon)

Source: https://github.com/firecrawl/cli/blob/main/README.md

Placeholder for the upcoming OpenCode AI workflow for deep research. Requires the OpenCode CLI to be installed.

```bash
firecrawl x opencode deep-research "browser automation frameworks"
```

--------------------------------

### Wait for AI Agent Job Completion

Source: https://context7.com/firecrawl/cli/llms.txt

Starts an AI agent job and waits for it to complete before returning the results. Use the `--wait` flag.

```bash
firecrawl agent "Extract all product names and prices" --wait
```

--------------------------------

### AI Workflow: Claude Lead Research

Source: https://github.com/firecrawl/cli/blob/main/README.md

Use the Claude AI workflow to conduct research on a specified company or entity. The Claude CLI must be installed.

```bash
firecrawl x claude lead-research "Vercel"
```

--------------------------------

### Submitting Search Feedback

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-cli/SKILL.md

Example of sending structured feedback for a search result to potentially refund credits and improve search quality. The `--missing-content` field is particularly valuable for identifying expected but absent information.

```bash
SEARCH_ID=$(jq -r '.id' .firecrawl/search-react-hooks.json)

firecrawl search-feedback "$SEARCH_ID" \
  --rating good \
  --valuable-sources '[{"url":"https://react.dev/reference/react/hooks","reason":"Authoritative"}]' \
  --missing-content '[{"topic":"useDeferredValue example"},{"topic":"Server Components hooks"}]' \
  --query-suggestions "Boost react.dev for react-hooks queries" \
  --silent &
```

--------------------------------

### Wait for Crawl Completion

Source: https://github.com/firecrawl/cli/blob/main/README.md

Start a crawl and automatically wait for it to finish. The '--progress' flag shows a progress indicator during the wait.

```bash
firecrawl crawl https://example.com --wait
```

```bash
firecrawl crawl https://example.com --wait --progress
```

--------------------------------

### AI Workflow: Claude SEO Audit

Source: https://github.com/firecrawl/cli/blob/main/README.md

Run the Claude AI workflow to perform an SEO audit on a given website. Requires the Claude CLI to be installed.

```bash
firecrawl x claude seo-audit https://example.com
```

--------------------------------

### Troubleshoot Firecrawl CLI Version Check

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-cli/rules/install.md

Checks the Firecrawl CLI version using npx, useful if the 'firecrawl' command is not found after installation.

```bash
npx firecrawl-cli@1.16.2 --version
```

--------------------------------

### Agent with Structured Output Schema

Source: https://github.com/firecrawl/cli/blob/main/README.md

Define a JSON schema using the `--schema` option to guide the agent in extracting data into a specific structured format. This ensures consistent output.

```bash
firecrawl agent "Extract company info" --schema '{"type":"object","properties":{"name":{"type":"string"},"employees":{"type":"number"}}}'
```

--------------------------------

### Search and Scrape Content

Source: https://github.com/firecrawl/cli/blob/main/README.md

Perform a search and enable scraping to get the full content of the results. Output can be saved to a JSON file. Use when you need the actual text from web pages.

```bash
firecrawl search "firecrawl documentation" --scrape --scrape-formats markdown --json -o results.json
```

--------------------------------

### Experimental: Bulk Site Download with Options

Source: https://github.com/firecrawl/cli/blob/main/README.md

Perform a bulk site download with additional options like generating screenshots, limiting the number of pages, and automatically approving tool permissions.

```bash
firecrawl x download https://docs.firecrawl.dev --screenshot --limit 20 -y
```

--------------------------------

### Experimental: Bulk Site Download

Source: https://github.com/firecrawl/cli/blob/main/README.md

Use the experimental `download` command to save an entire website as local files. This combines `map` and `scrape` functionalities.

```bash
firecrawl x download https://docs.firecrawl.dev
```

--------------------------------

### Website Download with Multiple Formats and Screenshots

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-download/SKILL.md

Downloads a website in multiple formats (markdown and links) along with screenshots. Each page will have its corresponding files (e.g., index.md, links.txt, screenshot.png).

```bash
firecrawl download https://docs.example.com --format markdown,links --screenshot --limit 20 -y
```

--------------------------------

### Update Firecrawl CLI

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-cli/rules/install.md

Updates the globally installed firecrawl-cli package to the latest version.

```bash
npm update -g firecrawl-cli
```

--------------------------------

### Check Firecrawl CLI Status

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-cli/rules/install.md

Verifies the current status of the Firecrawl CLI installation and configuration.

```bash
firecrawl --status
```

--------------------------------

### Website Download with Screenshots and Limit

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-download/SKILL.md

Downloads a website including screenshots, limiting the download to 20 pages. The `-y` flag skips the confirmation prompt.

```bash
firecrawl download https://docs.example.com --screenshot --limit 20 -y
```

--------------------------------

### Experimental: Bulk Site Download with Path Inclusion

Source: https://github.com/firecrawl/cli/blob/main/README.md

Download a site while specifying which paths to include, useful for targeting specific sections of a website. The `-y` flag auto-approves tool permissions.

```bash
firecrawl x download https://docs.firecrawl.dev --include-paths "/features,/sdks" -y
```

--------------------------------

### Bulk Site Download with `firecrawl x download`

Source: https://context7.com/firecrawl/cli/llms.txt

Use `firecrawl x download` for bulk downloading entire websites. It supports interactive wizards or non-interactive flags for specifying limits, include/exclude paths, content filtering, and output formats like HTML or screenshots. Files are organized locally by URL structure.

```bash
firecrawl x download https://docs.firecrawl.dev
```

```bash
firecrawl x download https://docs.firecrawl.dev \
  --limit 50 \
  --include-paths "/features,/sdks" \
  --only-main-content \
  -y
```

```bash
firecrawl x download https://docs.firecrawl.dev --screenshot --limit 20 -y
```

```bash
firecrawl x download https://docs.example.com \
  --exclude-paths "/zh,/ja,/fr" \
  --html \
  --limit 200 \
  -y
```

```bash
# Files are organized by URL structure:
# .firecrawl/docs.firecrawl.dev/index.md
# .firecrawl/docs.firecrawl.dev/features/scrape/index.md
# .firecrawl/docs.firecrawl.dev/sdks/python/index.md
```

--------------------------------

### Run Competitor Analysis Workflow

Source: https://github.com/firecrawl/cli/blob/main/src/commands/experimental/README.md

Initiates an interactive competitor analysis workflow. Use the `-y` flag to auto-approve tool permissions.

```bash
firecrawl claude competitor-analysis
```

```bash
firecrawl claude competitor-analysis https://firecrawl.dev -y
```

```bash
firecrawl claude competitor-analysis https://firecrawl.dev \
  --competitors "apify,scrapy,crawlee" \
  -o json -y
```

--------------------------------

### Agent with Schema from File

Source: https://github.com/firecrawl/cli/blob/main/README.md

Load a JSON schema for structured output from a file using the `--schema-file` option. This is convenient for complex or reusable schemas. Use `--wait` to ensure completion.

```bash
firecrawl agent "Extract product data" --schema-file ./product-schema.json --wait
```

--------------------------------

### File Output and Naming Conventions

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-cli/SKILL.md

Illustrates the default output directory (`.firecrawl/`) and common naming conventions for search and scrape results. It also suggests using tools like `wc`, `head`, and `grep` for incremental file processing instead of reading entire files at once.

```bash
wc -l .firecrawl/file.md && head -50 .firecrawl/file.md
```

```bash
grep -n "keyword" .firecrawl/file.md
```

--------------------------------

### Get Recent News

Source: https://github.com/firecrawl/cli/blob/main/README.md

Retrieve news articles from a specific time frame, such as the past week. Useful for staying updated on current events.

```bash
firecrawl search "AI startups funding" --sources news --tbs qdr:w --limit 15
```

--------------------------------

### Basic Web Search

Source: https://github.com/firecrawl/cli/blob/main/README.md

Perform a basic web search using the `search` command followed by your query.

```bash
firecrawl search "firecrawl"
```

--------------------------------

### Run SEO Audit Workflow

Source: https://github.com/firecrawl/cli/blob/main/src/commands/experimental/README.md

Initiates an interactive SEO audit workflow for a website.

```bash
firecrawl claude seo-audit
```

--------------------------------

### CI/CD: Use Self-Hosted Instance

Source: https://github.com/firecrawl/cli/blob/main/README.md

Configure the CLI to use a self-hosted Firecrawl instance by setting the API URL environment variable and then scrape content to a Markdown file.

```bash
export FIRECRAWL_API_URL=${{ secrets.FIRECRAWL_API_URL }}
firecrawl scrape https://example.com -o output.md
```

--------------------------------

### Website Download Filtering by Included Paths

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-download/SKILL.md

Downloads a website, but only includes pages that match the specified paths, such as '/features' and '/sdks'.

```bash
firecrawl download https://docs.example.com --include-paths "/features,/sdks"
```

--------------------------------

### Get Sitemap URLs Only

Source: https://github.com/firecrawl/cli/blob/main/README.md

Use the '--sitemap only' option to retrieve only URLs found in the website's sitemap. Useful for focusing on indexed content.

```bash
firecrawl map https://example.com --sitemap only
```

--------------------------------

### Load Scrape Options from File

Source: https://context7.com/firecrawl/cli/llms.txt

Loads custom scraping options for a crawl from a JSON file, allowing for complex configurations to be reused.

```bash
# Load scrape options from file
firecrawl crawl https://example.com \
  --scrape-options-file ./crawl-options.json \
  --wait
```

--------------------------------

### Firecrawl CLI Authentication Methods

Source: https://github.com/firecrawl/cli/blob/main/README.md

Demonstrates various ways to authenticate with the Firecrawl CLI, including interactive login, browser login, API key, and environment variables.

```bash
firecrawl
firecrawl login
firecrawl login --api-key fc-your-api-key
export FIRECRAWL_API_KEY=fc-your-api-key
firecrawl scrape https://example.com --api-key fc-your-api-key
```

--------------------------------

### Website Download Excluding Specific Paths

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-download/SKILL.md

Downloads a website while skipping pages that match the provided exclusion paths, useful for avoiding translated or irrelevant sections.

```bash
firecrawl download https://docs.example.com --exclude-paths "/zh,/ja,/fr,/es,/pt-BR"
```

--------------------------------

### Handle Paths with Spaces

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-parse/SKILL.md

Demonstrates how to parse a local file with a name containing spaces by quoting the file path.

```bash
firecrawl parse "./My Doc.pdf" -o .firecrawl/mydoc.md
```

--------------------------------

### Run Lead Research Workflow

Source: https://github.com/firecrawl/cli/blob/main/src/commands/experimental/README.md

Executes a lead research workflow for pre-meeting intelligence. It can optionally research a specific person and use provided context. Use `-y` to auto-approve tool permissions.

```bash
firecrawl claude lead-research
```

```bash
firecrawl claude lead-research "Stripe" --person "Patrick Collison" --context "partnership call" -y
```

--------------------------------

### Configuration Management with `firecrawl config` / `view-config`

Source: https://context7.com/firecrawl/cli/llms.txt

Manage Firecrawl configuration using `config` and `view-config`. Set API URLs (including self-hosted), authenticate via browser, view current settings, log out, and export API keys to `.env` files.

```bash
# Configure API URL (e.g., self-hosted)
firecrawl config --api-url https://firecrawl.mycompany.com
```

```bash
firecrawl config --api-url http://localhost:3002 --api-key fc-xxx
```

```bash
# Login via browser
firecrawl config --browser
```

```bash
# View current configuration and auth status
firecrawl view-config
```

```bash
# Logout (clears stored credentials)
firecrawl logout
```

```bash
# Pull API key into .env file
firecrawl env
```

```bash
firecrawl env --file .env.local --overwrite
```

--------------------------------

### Discover All URLs on a Website

Source: https://github.com/firecrawl/cli/blob/main/README.md

Use the 'map' command to quickly list all URLs on a given website without scraping their content. Useful for site audits or index generation.

```bash
firecrawl map https://example.com
```

```bash
firecrawl map https://example.com --json
```

```bash
firecrawl map https://example.com --search "blog"
```

```bash
firecrawl map https://example.com --limit 500
```

--------------------------------

### Submit Bad Search Feedback (Quick Form)

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-search/SKILL.md

Use a simplified format for bad search results, repeating `--missing-content` or using comma-separated topics. `--silent &` runs the command in the background without blocking.

```bash
firecrawl search-feedback "$SEARCH_ID" \
  --rating bad \
  --missing-content "official api reference: missing v2 endpoints" \
  --missing-content "code examples in python" \
  --silent &
```

--------------------------------

### Interact with a Page using Natural Language Prompts

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-interact/SKILL.md

Executes instructions provided in natural language to interact with the last scraped page. Use for actions like clicking buttons, filling forms, or extracting specific data.

```bash
firecrawl interact --prompt "Click the login button"
```

```bash
firecrawl interact --prompt "Fill in the email field with test@example.com"
```

```bash
firecrawl interact --prompt "Extract the pricing table"
```

--------------------------------

### Generate AI Summary from Local File

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-parse/SKILL.md

Generates an AI-powered summary of a local PDF file and saves it to a markdown file. The -S flag enables summary generation.

```bash
firecrawl parse ./paper.pdf -S -o .firecrawl/paper-summary.md
```

--------------------------------

### Crawl with Live Progress Updates

Source: https://context7.com/firecrawl/cli/llms.txt

Initiates a website crawl and waits for completion, displaying live progress updates on stderr. Shows page count and current status.

```bash
# Wait with live progress counter
firecrawl crawl https://example.com --wait --progress
```

--------------------------------

### Ask a Question About Page Content

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-scrape/SKILL.md

Use this command to ask a specific question about the content of a webpage. This feature consumes 5 credits and returns a targeted answer.

```bash
firecrawl scrape "https://example.com/pricing" --query "What is the enterprise plan price?"
```

--------------------------------

### Full Domain Crawl with Subdomains and File Output

Source: https://context7.com/firecrawl/cli/llms.txt

Perform a comprehensive crawl of a domain, including all subdomains, and save the results to a JSON file. Includes progress and wait options.

```bash
firecrawl crawl https://example.com \
  --crawl-entire-domain \
  --allow-subdomains \
  --wait --progress \
  --pretty -o crawl-results.json
```

--------------------------------

### Scrape URL and Ask a Question

Source: https://context7.com/firecrawl/cli/llms.txt

Scrapes a URL and then uses AI to answer a specific question about the page content. Requires the content to be scraped first.

```bash
firecrawl scrape https://docs.example.com -Q "What authentication methods are supported?"
```

--------------------------------

### Raw Markdown Output

Source: https://github.com/firecrawl/cli/blob/main/README.md

Request raw markdown content from a URL.

```bash
firecrawl https://example.com --format markdown
```

--------------------------------

### View Current Configuration

Source: https://github.com/firecrawl/cli/blob/main/README.md

Display the current FireCrawl configuration, including authentication status and credentials location.

```bash
firecrawl view-config
```

--------------------------------

### CI/CD: Set API Key and Crawl

Source: https://github.com/firecrawl/cli/blob/main/README.md

In a CI/CD environment, set the Firecrawl API key using an environment variable and then perform a crawl, waiting for completion and saving to a JSON file.

```bash
export FIRECRAWL_API_KEY=${{ secrets.FIRECRAWL_API_KEY }}
firecrawl crawl https://docs.example.com --wait -o docs.json
```

--------------------------------

### Timing Information for Parsing

Source: https://context7.com/firecrawl/cli/llms.txt

Measures and displays the time taken to parse a local document. Use the `--timing` flag to enable this.

```bash
firecrawl parse ./large-doc.pdf --timing
```

--------------------------------

### AI Workflow: Claude Shopping Search

Source: https://github.com/firecrawl/cli/blob/main/README.md

Utilize the Claude AI workflow for shopping-related queries, such as finding the best products. Requires the Claude CLI.

```bash
firecrawl x claude shop "best mechanical keyboard for developers"
```

--------------------------------

### Authenticate Firecrawl CLI with Browser

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-cli/rules/install.md

Initiates the authentication process for the Firecrawl CLI by opening a browser for OAuth. Credentials are then securely stored.

```bash
firecrawl login --browser
```

--------------------------------

### Map All URLs on a Site

Source: https://context7.com/firecrawl/cli/llms.txt

Discovers all URLs on a website without downloading page content. Outputs URLs one per line.

```bash
# List all URLs (one per line, like curl output)
firecrawl map https://docs.firecrawl.dev
```

--------------------------------

### Search Multiple Source Types

Source: https://context7.com/firecrawl/cli/llms.txt

Searches across multiple source types simultaneously, such as web, news, and images.

```bash
# Multiple source types
firecrawl search "machine learning tutorials" --sources web,news,images
```

--------------------------------

### Scrape URL with Timing Diagnostics

Source: https://context7.com/firecrawl/cli/llms.txt

Scrapes a URL and provides detailed timing diagnostics for the scraping process. This helps in analyzing performance bottlenecks.

```bash
firecrawl scrape https://example.com --timing
```

--------------------------------

### Parse PDF and Ask a Question

Source: https://context7.com/firecrawl/cli/llms.txt

Parses a local PDF file and allows you to ask a specific question about its content, receiving a concise answer.

```bash
firecrawl parse ./annual-report.pdf -Q "What was the total revenue in Q4?"
```

--------------------------------

### Configure Firecrawl CLI for Self-Hosted Instance

Source: https://context7.com/firecrawl/cli/llms.txt

Logs into a self-hosted or local Firecrawl instance by specifying the API URL and API key.

```bash
firecrawl login --api-url http://localhost:3002 --api-key fc-xxx
```

--------------------------------

### Scrape URL and Output as HTML

Source: https://context7.com/firecrawl/cli/llms.txt

Scrapes a URL and returns the raw HTML content. Use the `-H` flag as a shorthand.

```bash
firecrawl scrape https://example.com --html
```

```bash
firecrawl scrape https://example.com -H
```

--------------------------------

### Parallel QA Testing with Cloud Browser

Source: https://github.com/firecrawl/cli/blob/main/src/commands/experimental/README.md

Initiate parallel QA testing using Firecrawl's cloud browser. This workflow simulates a QA team to find bugs by interacting with the site. Use the -y flag to auto-approve permissions.

```bash
firecrawl claude qa
```

```bash
firecrawl claude qa https://myapp.com -y
```

```bash
firecrawl claude qa https://myapp.com --focus forms -y
```

--------------------------------

### JSON Output with Multiple Formats

Source: https://github.com/firecrawl/cli/blob/main/README.md

Request multiple data formats (e.g., markdown, links, images) and receive them in a single JSON output.

```bash
firecrawl https://example.com --format markdown,links,images
```

--------------------------------

### Processing Output Files

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-cli/SKILL.md

Demonstrates how to use `jq` to extract specific data like URLs or titles and URLs from JSON output files generated by Firecrawl.

```bash
jq -r '.data.web[].url' .firecrawl/search.json
```

```bash
jq -r '.data.web[] | "\(.title): \(.url)"' .firecrawl/search.json
```

--------------------------------

### Check Credit Usage

Source: https://github.com/firecrawl/cli/blob/main/README.md

Monitor your Firecrawl credit consumption with the `credit-usage` command. Use `--json` and `--pretty` for structured, readable output.

```bash
firecrawl credit-usage
```

```bash
firecrawl credit-usage --json --pretty
```

--------------------------------

### Crawl a Docs Section

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-crawl/SKILL.md

Use this command to crawl a specific section of a website, like the /docs directory. It limits the crawl to 50 pages and waits for completion, saving the output to a JSON file.

```bash
firecrawl crawl "<url>" --include-paths /docs --limit 50 --wait -o .firecrawl/crawl.json
```

--------------------------------

### Pretty JSON Search Output

Source: https://github.com/firecrawl/cli/blob/main/README.md

Format the output of a web search as pretty-printed JSON by including the `--pretty` flag.

```bash
firecrawl search "AI data tools"
```

--------------------------------

### Find GitHub Repositories

Source: https://github.com/firecrawl/cli/blob/main/README.md

Filter search results to find specific categories like GitHub repositories. Useful for developers looking for code or projects.

```bash
firecrawl search "web data library" --categories github --limit 20
```

--------------------------------

### Basic Markdown Extraction

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-scrape/SKILL.md

Use this command for basic markdown extraction from a single URL. The output is saved to a specified file.

```bash
firecrawl scrape "<url>" -o .firecrawl/page.md
```

--------------------------------

### Scrape URL with Multiple Output Formats and Pretty Printing

Source: https://context7.com/firecrawl/cli/llms.txt

Scrapes a URL and outputs the content in multiple specified formats (e.g., markdown, links, images). The `--pretty` flag formats the JSON output for readability.

```bash
firecrawl scrape https://example.com --format markdown,links,images --pretty
```

--------------------------------

### Search and Scrape Results

Source: https://github.com/firecrawl/cli/blob/main/README.md

Combine web search with scraping by adding the `--scrape` flag. You can also specify desired output formats for scraped content using `--scrape-formats`.

```bash
firecrawl search "firecrawl tutorials" --scrape
```

```bash
firecrawl search "API documentation" --scrape --scrape-formats markdown,links
```

--------------------------------

### Interact with Live Page Using Natural Language

Source: https://context7.com/firecrawl/cli/llms.txt

Performs actions on a live web page based on natural language commands. This command assumes a previous scrape ID is active.

```bash
firecrawl interact "Click the Pro plan tab"
```

```bash
firecrawl interact "Extract the pricing table as a structured list"
```

```bash
firecrawl interact "What is the price of the Enterprise plan?"
```

--------------------------------

### Scrape URL with Firecrawl CLI

Source: https://github.com/firecrawl/cli/blob/main/README.md

Basic usage of the scrape command to extract content from a URL. Outputs markdown by default. Use -o to save to a file.

```bash
firecrawl https://example.com
firecrawl scrape https://example.com
firecrawl https://example.com -o output.md
```

--------------------------------

### Firecrawl CLI Manual Login Options

Source: https://context7.com/firecrawl/cli/llms.txt

Provides different methods for logging into the Firecrawl CLI. Options include interactive login, browser-based OAuth, or direct API key input.

```bash
firecrawl login
```

```bash
firecrawl login --method browser
```

```bash
firecrawl login --api-key fc-YOUR-KEY
```

--------------------------------

### Scrape and Interact with a Page

Source: https://github.com/firecrawl/cli/blob/main/README.md

Scrape a web page and then interact with the live browser session using natural language commands or code. Requires a prior scrape operation.

```bash
firecrawl scrape https://example.com
```

```bash
firecrawl interact "Click the pricing tab"
```

```bash
firecrawl interact "Fill in the email field with test@example.com"
```

```bash
firecrawl interact "Extract the pricing table"
```

--------------------------------

### Scrape Markdown and Take Screenshot

Source: https://github.com/firecrawl/cli/blob/main/README.md

Combine `--format markdown` with `--screenshot` to obtain both the cleaned markdown content and a base64 encoded screenshot of the page.

```bash
firecrawl https://example.com --format markdown --screenshot
```

--------------------------------

### Convert Scraped Content to PDF

Source: https://github.com/firecrawl/cli/blob/main/README.md

Scrape content from a URL and convert it to a PDF document using `pandoc`.

```bash
firecrawl https://example.com | pandoc -o document.pdf
```

--------------------------------

### Save URL List to File

Source: https://context7.com/firecrawl/cli/llms.txt

Saves the discovered list of URLs from a website to a specified text file.

```bash
# Save URL list to file
firecrawl map https://example.com -o urls.txt
```

--------------------------------

### Full Website Crawl with Depth Limit

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-crawl/SKILL.md

Perform a full crawl of a website up to a maximum link depth of 3. This command displays progress and saves the results to a JSON file.

```bash
firecrawl crawl "<url>" --max-depth 3 --wait --progress -o .firecrawl/crawl.json
```

--------------------------------

### Scrape After JS Rendering

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-scrape/SKILL.md

Use this command when you need to wait for JavaScript to render the page content before scraping. Specify the waiting time in milliseconds. The output is saved to a file.

```bash
firecrawl scrape "<url>" --wait-for 3000 -o .firecrawl/page.md
```

--------------------------------

### Search and Scrape Full Content

Source: https://context7.com/firecrawl/cli/llms.txt

Searches the web and then scrapes the full content from the search results, saving them to a JSON file.

```bash
# Search and scrape full content from results
firecrawl search "firecrawl documentation" --scrape --scrape-formats markdown --json -o results.json
```

--------------------------------

### Submit Search Quality Feedback (Partial)

Source: https://context7.com/firecrawl/cli/llms.txt

Submits partial feedback for a search, specifying valuable sources and missing content. Useful for detailed quality reporting.

```bash
# Rate as partial with details
firecrawl search-feedback search_abc123 \
  --rating partial \
  --valuable-sources "https://example.com/docs,https://example.com/api" \
  --missing-content "pricing tiers, rate limits" \
  --query-suggestions "Try adding site:docs.example.com to narrow results"
```

--------------------------------

### Extract Links and Process with jq

Source: https://github.com/firecrawl/cli/blob/main/README.md

Extract only the links from a URL and process them using the `jq` command-line JSON processor.

```bash
firecrawl https://example.com --format links | jq '.links[].url'
```

--------------------------------

### Submit Good Search Feedback

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-search/SKILL.md

Send feedback for useful search results, including valuable sources and specific missing content. Use `--silent &` to run in the background and suppress output.

```bash
firecrawl search-feedback "$SEARCH_ID" \
  --rating good \
  --valuable-sources '[{"url":"https://react.dev/reference/react/hooks","reason":"Most authoritative"}]' \
  --missing-content '[
    {"topic":"useDeferredValue","description":"No example of useDeferredValue with Suspense"},
    {"topic":"useTransition","description":"No coverage of useTransition for routing"}
  ]' \
  --query-suggestions "Boost react.dev for queries about react hooks" \
  --silent &
```

--------------------------------

### AI Summary of Local Document

Source: https://context7.com/firecrawl/cli/llms.txt

Generates an AI-powered summary of a local document, such as a PDF. Use the `-S` flag for summarization.

```bash
firecrawl parse ./whitepaper.pdf -S
```

--------------------------------

### JSON Search Output with ID

Source: https://context7.com/firecrawl/cli/llms.txt

Performs a search and outputs results in JSON format, including a search ID for feedback. Use --pretty for readable JSON.

```bash
# JSON output with search ID for feedback
firecrawl search "React Server Components" --json --pretty
```

--------------------------------

### Parse Local File to Markdown

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-parse/SKILL.md

Converts a local PDF file into a markdown file. Ensure the output directory exists.

```bash
mkdir -p .firecrawl

firecrawl parse ./paper.pdf -o .firecrawl/paper.md
```

--------------------------------

### Interact with Explicit Prompt

Source: https://context7.com/firecrawl/cli/llms.txt

Executes a specific, explicit prompt against the live web page in an interactive session. Use the `-p` flag.

```bash
firecrawl interact -p "Fill the email field with test@example.com and click Submit"
```

--------------------------------

### JSON Output for Interactive Session

Source: https://context7.com/firecrawl/cli/llms.txt

Requests the output of an interactive session command to be formatted as JSON. Use the `--json` flag.

```bash
firecrawl interact "Get the page title" --json
```

--------------------------------

### Extract Data with Cost Limit

Source: https://github.com/firecrawl/cli/blob/main/README.md

Perform an agent task with a maximum credit limit. Ensures cost control for data extraction.

```bash
firecrawl agent "Get all blog post titles and dates" --urls https://blog.example.com --max-credits 100 --wait
```

--------------------------------

### Execute Code in Browser Sandbox (Python)

Source: https://github.com/firecrawl/cli/blob/main/README.md

Run Python code within the browser sandbox during an interactive session. Requires the `--python` flag.

```bash
firecrawl interact "Print the page title" --python
```

```bash
firecrawl interact -c "print(await page.title())" --python
```

--------------------------------

### Ask a Question About a Local File

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-parse/SKILL.md

Extracts information by asking a specific question about the content of a local PDF file. The answer is saved to a markdown file.

```bash
firecrawl parse ./paper.pdf -Q "What are the main conclusions?" \
  -o .firecrawl/paper-qa.md
```

--------------------------------

### Scrape with Profile Persistence

Source: https://github.com/firecrawl/cli/blob/main/README.md

Scrape a page and persist browser state using a profile. Allows maintaining login sessions across multiple scrapes.

```bash
firecrawl scrape "https://app.example.com/login" --profile my-app
```

```bash
firecrawl interact "Fill in email and click login"
```

```bash
firecrawl scrape "https://app.example.com/dashboard" --profile my-app
```

```bash
firecrawl interact "Extract the dashboard data"
```

--------------------------------

### Parse HTML to HTML Output

Source: https://context7.com/firecrawl/cli/llms.txt

Converts a local HTML file and outputs it in HTML format. Use the `-H` flag for this specific output.

```bash
firecrawl parse ./page.html -H
```

--------------------------------

### Execute Precise Code in Browser Session

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-interact/SKILL.md

Runs specific code commands within the live browser session for fine-grained control. Requires specifying the code and its language.

```bash
firecrawl interact --code "agent-browser click @e5" --language bash
```

```bash
firecrawl interact --code "agent-browser snapshot -i" --language bash
```

--------------------------------

### Deep Crawl with High Limits and Progress

Source: https://github.com/firecrawl/cli/blob/main/README.md

Perform extensive crawls by setting a high `--limit` for pages and `--max-depth` for crawl depth. Use `--wait` and `--progress` to monitor the process.

```bash
firecrawl crawl https://example.com --limit 1000 --max-depth 10 --wait --progress
```

--------------------------------

### Parse Document with Multiple Output Formats

Source: https://context7.com/firecrawl/cli/llms.txt

Extracts information from a local document in multiple specified formats, such as markdown and links.

```bash
firecrawl parse ./contract.docx --format markdown,links
```

--------------------------------

### AI Workflow: Claude Natural Language Prompt

Source: https://github.com/firecrawl/cli/blob/main/README.md

Execute an AI workflow using Claude by providing a natural language instruction without specifying a workflow name. This allows for flexible, ad-hoc tasks.

```bash
firecrawl x claude "scrape the firecrawl docs and summarize"
```

--------------------------------

### Firecrawl Search with Scraping

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-cli/SKILL.md

Perform a web search for a given query and scrape the content of the top results. The --limit option controls the number of results to fetch.

```bash
firecrawl search "query" --scrape --limit 3
```

--------------------------------

### Search Research Topics

Source: https://github.com/firecrawl/cli/blob/main/README.md

Use this command to research a topic, specifying time filters and the maximum number of results. Useful for finding recent information.

```bash
firecrawl search "React Server Components" --tbs qdr:m --limit 10
```

--------------------------------

### Focus Extraction on Specific URLs

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-agent/SKILL.md

Extracts data from a website, focusing the agent's attention on the provided URLs. The `--wait` flag is used to wait for the task to complete, and `-o` specifies the output file.

```bash
firecrawl agent "get feature list" --urls "<url>" --wait -o .firecrawl/features.json
```

--------------------------------

### Scrape URL with Multiple Formats and Pretty JSON Output

Source: https://github.com/firecrawl/cli/blob/main/README.md

Extracts content in multiple formats (e.g., markdown, links, images) and saves it as pretty-printed JSON to a specified file.

```bash
firecrawl https://example.com --format markdown,links,images
firecrawl https://example.com --format json -o data.json --pretty
```

--------------------------------

### Submit Search Quality Feedback (Good)

Source: https://context7.com/firecrawl/cli/llms.txt

Submits positive feedback for a previous search using its ID. The first submission per search ID refunds 1 credit.

```bash
# Rate a search as good
firecrawl search-feedback search_abc123 --rating good
```

--------------------------------

### Sitemap Only Discovery

Source: https://context7.com/firecrawl/cli/llms.txt

Configures the site mapping process to exclusively use the sitemap for URL discovery.

```bash
# Sitemap only
firecrawl map https://example.com --sitemap only
```

--------------------------------

### AI Agent Webhook on Completion

Source: https://context7.com/firecrawl/cli/llms.txt

Configures a webhook to be called upon completion of an AI agent job. Specify the webhook URL with `--webhook`.

```bash
firecrawl agent "Scrape all job listings from careers page" \
  --urls https://example.com/careers \
  --webhook https://myapp.com/hooks/agent \
  --wait
```

--------------------------------

### Scrape a URL for Markdown Content

Source: https://context7.com/firecrawl/cli/llms.txt

Scrapes a single URL and outputs the content in markdown format to standard output. This is the default behavior when a URL is provided as the first argument.

```bash
firecrawl scrape https://example.com
```

```bash
firecrawl https://example.com
```

--------------------------------

### Run Firecrawl CLI Tests

Source: https://github.com/firecrawl/cli/blob/main/src/__tests__/README.md

Commands to execute tests for the Firecrawl CLI. Use `pnpm test:run` for a single execution, `pnpm test:watch` for continuous testing, and `pnpm test:ui` to run tests with a user interface.

```bash
# Run tests once
pnpm test:run
```

```bash
# Run tests in watch mode
pnpm test:watch
```

```bash
# Run tests with UI
pnpm test:ui
```

--------------------------------

### JSON Output with Pretty Printing

Source: https://github.com/firecrawl/cli/blob/main/README.md

Generate JSON output from a FireCrawl command and format it for readability.

```bash
firecrawl https://example.com --format links --pretty
```

--------------------------------

### Execute Node.js/Playwright Code in Interactive Session

Source: https://context7.com/firecrawl/cli/llms.txt

Runs Node.js code using Playwright against the live web page. Use the `-c` flag for code execution.

```bash
firecrawl interact -c "await page.title()"
```

```bash
firecrawl interact -c "const price = await page.locator('.price').textContent(); console.log(price)"
```

--------------------------------

### AI Workflow: Claude Deep Research

Source: https://github.com/firecrawl/cli/blob/main/README.md

Initiate the Claude AI workflow for deep research on a specific topic. This command requires the Claude CLI.

```bash
firecrawl x claude deep-research "RAG pipeline data ingestion tools"
```

--------------------------------

### Scrape URL for AI-Generated Summary

Source: https://context7.com/firecrawl/cli/llms.txt

Scrapes a URL and uses AI to generate a concise summary of the content. This is useful for quickly understanding the main points of a page.

```bash
firecrawl scrape https://blog.example.com -S
```

--------------------------------

### JSON Output for Site Map

Source: https://context7.com/firecrawl/cli/llms.txt

Generates a JSON output of all discovered URLs on a website, useful for programmatic processing.

```bash
# JSON output
firecrawl map https://example.com --json --pretty
```

--------------------------------

### Execute Code in Browser Sandbox (Node.js)

Source: https://github.com/firecrawl/cli/blob/main/README.md

Run Node.js code within the browser sandbox during an interactive session. Useful for advanced automation.

```bash
firecrawl interact -c "await page.title()"
```

--------------------------------

### Basic Firecrawl Search

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-search/SKILL.md

Perform a basic web search and save the results to a JSON file. Use this for general information retrieval.

```bash
firecrawl search "your query" -o .firecrawl/result.json --json
```

--------------------------------

### Checking Firecrawl Credit Usage

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-cli/SKILL.md

Commands to display current credit usage. The `--json` and `--pretty` flags can be used to format the output as JSON and save it to a file.

```bash
firecrawl credit-usage
```

```bash
firecrawl credit-usage --json --pretty -o .firecrawl/credits.json
```

--------------------------------

### Execute Python/Playwright Code in Interactive Session

Source: https://context7.com/firecrawl/cli/llms.txt

Runs Python code using Playwright against the live web page. Use `--python` and `-c` for execution.

```bash
firecrawl interact -c "print(await page.title())" --python
```

```bash
firecrawl interact --python -c '
items = await page.query_selector_all(".plan-name")
for item in items:
    print(await item.inner_text())
'
```

--------------------------------

### Focus Agent on Specific URLs

Source: https://github.com/firecrawl/cli/blob/main/README.md

Direct the AI agent to focus its extraction efforts on a provided list of URLs using the `--urls` option. This improves relevance and efficiency.

```bash
firecrawl agent "Get the main features listed" --urls https://example.com/features
```

--------------------------------

### Scrape URL for Raw HTML

Source: https://github.com/firecrawl/cli/blob/main/README.md

Retrieves the raw HTML content of a webpage using the --html or -H flag.

```bash
firecrawl https://example.com --html
firecrawl https://example.com -H
```

--------------------------------

### Take Screenshot of a URL

Source: https://context7.com/firecrawl/cli/llms.txt

Takes a screenshot of a given URL. Use `--full-page-screenshot` to capture the entire scrollable page.

```bash
firecrawl scrape https://example.com --screenshot
```

```bash
firecrawl scrape https://example.com --full-page-screenshot
```

--------------------------------

### Include Subdomains in URL Discovery

Source: https://github.com/firecrawl/cli/blob/main/README.md

Discover URLs across all subdomains of a given website by using the '--include-subdomains' flag. Useful for comprehensive site mapping.

```bash
firecrawl map https://example.com --include-subdomains --limit 1000
```

--------------------------------

### Submit Search Quality Feedback (Bad) with JSON

Source: https://context7.com/firecrawl/cli/llms.txt

Submits negative feedback for a search in JSON format, specifying missing content details. Saves output to a file.

```bash
# JSON output and save to file
firecrawl search-feedback search_abc123 --rating bad \
  --missing-content '[{"topic":"auth","description":"OAuth2 flow missing"}]' \
  --json --pretty -o feedback.json
```

--------------------------------

### Execute Bash Command in Interactive Sandbox

Source: https://context7.com/firecrawl/cli/llms.txt

Runs a bash command within the interactive session's sandbox environment. Use `--bash` and `-c`.

```bash
firecrawl interact -c "agent-browser snapshot" --bash
```

--------------------------------

### Structured Extraction from Schema File

Source: https://context7.com/firecrawl/cli/llms.txt

Scrapes a URL and extracts data into a structured JSON file using a schema defined in a separate file. The output is saved to `product.json`.

```bash
firecrawl scrape https://shop.example.com \
  --format json \
  --schema-file ./product-schema.json \
  -o product.json --pretty
```

--------------------------------

### Scrape Page for Interactive Session

Source: https://context7.com/firecrawl/cli/llms.txt

Initiates a scrape of a given URL, which is automatically saved for subsequent interactive sessions. This is the first step before using `firecrawl interact`.

```bash
firecrawl scrape https://app.example.com/pricing
```

--------------------------------

### Scrape with JavaScript Wait Time

Source: https://github.com/firecrawl/cli/blob/main/README.md

Employ the `--wait-for` option to specify a delay in milliseconds, allowing JavaScript-rendered content to load before scraping.

```bash
firecrawl https://spa-app.com --wait-for 3000
```

--------------------------------

### Find a Specific Page on a Large Site

Source: https://github.com/firecrawl/cli/blob/main/skills/firecrawl-map/SKILL.md

Use this command to find a specific page within a large website by providing a search query. The results are saved to a text file.

```bash
firecrawl map "<url>" --search "authentication" -o .firecrawl/filtered.txt
```

--------------------------------

### SEO Audit with Keywords

Source: https://github.com/firecrawl/cli/blob/main/src/commands/experimental/README.md

Perform an SEO audit on a given URL, specifying target keywords. The -y flag automatically approves tool permissions.

```bash
firecrawl claude seo-audit https://example.com --keywords "scraping,api,web data" -y
```