crawlfix

External SEO and AI-search auditor. Scan sites, get prioritized fix prompts for your coding agent. Zero dependencies, Node 18+.

# One-off, no install:
npx crawlfix scan example.com

# Or install globally:
npm install -g crawlfix
crawlfix scan example.com

Quick start

crawlfix login                 # device-flow login, opens your browser
crawlfix scan example.com      # run an audit, prints summary on completion
crawlfix issues <auditId>      # list issues for that audit
crawlfix fix <issueId>         # get an AI fix prompt to paste into your agent

Commands

Command	What it does
crawlfix login [--email <email>]	Authenticate this machine via device-authorization-grant flow.
crawlfix logout	Clear locally stored credentials.
crawlfix whoami	Show the currently logged-in account and server URL.
crawlfix scan <url>	Run an audit; waits for completion and prints a summary.
crawlfix issues <auditId>	List all issues for an audit, with severity / category filters.
crawlfix fix <issueId>	Get an AI-generated fix prompt for one issue. (paid)
crawlfix history	List your recent audits.
crawlfix verify <issueId>	Re-crawl to confirm a fix worked.
crawlfix export <auditId> --format pdf	Download an audit report as JSON, HTML, or PDF. (paid)
crawlfix compare <auditId> <url>	Side-by-side audit against a competitor URL. (paid)
crawlfix monitor <subcommand>	Schedule recurring scans and alerts. (paid)
crawlfix policy generate ...	Generate a privacy / cookie / terms / AUP policy document. (paid)
crawlfix policy list [--domain X]	List policy documents you've already generated.
crawlfix policy install ...	Print a policy-install prompt for your coding agent.
crawlfix banner snippet ...	Print the consent-banner <script> tag for your site.
crawlfix banner install ...	Print a consent-banner install prompt for your coding agent.
crawlfix analytics snippet --site-id X	Print the analytics <script> tag for your site. (paid)
crawlfix analytics install --site-id X	Print an analytics install prompt for your coding agent.
crawlfix help [command]	Show this list, or detailed help for one command.
crawlfix --version	Print the installed CLI version.

crawlfix scan <url>

crawlfix scan example.com
crawlfix scan https://example.com --json       # pipe-friendly raw JSON
crawlfix scan example.com --no-wait            # submit and return immediately
crawlfix scan example.com --timeout 300        # max seconds to wait (default 180)

crawlfix issues <auditId>

crawlfix issues aud_abc123
crawlfix issues aud_abc123 --severity high
crawlfix issues aud_abc123 --category SEO
crawlfix issues aud_abc123 --json

crawlfix fix <issueId>

crawlfix fix iss_xyz789                       # prints prompt in a box
crawlfix fix iss_xyz789 --copy                # also copies to clipboard
crawlfix fix iss_xyz789 --save                # write .md in cwd, auto-named
crawlfix fix iss_xyz789 --save ./fixes/       # auto-named inside ./fixes/
crawlfix fix iss_xyz789 --out ./fix.md        # exact file path

--save writes a labeled markdown file with YAML front-matter (issue id, severity, category, page URL, scan id, timestamp), an H1 title, and the fix prompt. Hand the file straight to your coding agent instead of copy-pasting out of the terminal. Auto-named files look like crawlfix-fix-<issueId>-<host>-<YYYYMMDD>.md. Set CRAWLFIX_OUT_DIR to change the default directory for bare --save.

Paid feature. If you are on the free tier, the CLI opens a checkout page in your browser and waits for payment to confirm before retrying the call.

crawlfix monitor

crawlfix monitor add example.com
crawlfix monitor list
crawlfix monitor pause example.com
crawlfix monitor resume example.com
crawlfix monitor remove example.com

Embeddable widgets

Crawlfix ships three drop-in widgets your AI agent can install for you without ever opening the dashboard: a consent banner, a privacy-friendly analytics beacon, and on-demand policy documents (privacy / cookie / terms / AUP).

Each widget has two CLI surfaces:

• snippet prints the <script> tag (and marketplace JS URL) you drop into your site's HTML.
• install prints a framework-specific "install this widget" prompt you can paste into your coding agent (Claude Code, Cursor, etc.). Add --copy to put it on your clipboard or --save to write a .md file.

# Consent banner
crawlfix banner snippet --site-id site_abc --privacy-url https://x.com/privacy
crawlfix banner install --framework next-app --copy

# Analytics (paid)
crawlfix analytics snippet --site-id site_abc
crawlfix analytics install --site-id site_abc --framework sveltekit --save

# Policies
crawlfix policy generate --domain example.com --type privacy \
  --business-name "Acme" --jurisdictions gdpr-eu,ccpa --save
crawlfix policy list --domain example.com
crawlfix policy install --type privacy --framework next-app --copy

Supported --framework values: next-app, nuxt, sveltekit, astro, plain-html.

Environment variables

Variable	Purpose
CRAWLFIX_SERVER	Override the backend URL. Default: https://crawlfix.ai.
CRAWLFIX_TOKEN	Bearer token (for CI / scripting). Takes priority over ~/.crawlfix/creds.
CRAWLFIX_OUT_DIR	Default directory for crawlfix fix --save when no path is given.
CRAWLFIX_DEBUG	Set to 1 to print stack traces on errors.
NO_COLOR	Disable ANSI colors.
FORCE_COLOR	Force ANSI colors on (1 or true).

Credentials are stored at ~/.crawlfix/credentials.json with mode 0600 (POSIX). On Windows the file is created in %USERPROFILE%\.crawlfix\.

Local development

cd cli
npm link                       # symlinks `crawlfix` globally
crawlfix --version
crawlfix login --server http://localhost:7238
crawlfix scan example.com

Or run the entry directly without linking:

node cli/bin/crawlfix.js scan example.com

There is a tiny smoke test:

node cli/scripts/smoke-test.mjs

Publishing to npm

cd cli
npm login                      # one-time, prompts for an npm account
npm publish                    # publishes 'crawlfix' to the public registry

# After publish, anywhere on the internet:
npx crawlfix scan example.com

Support

Issues, bug reports, or licensing questions: Michael@faction.chat

License

MIT. See LICENSE.