Test Your Localhost App With AI — Before You Deploy

Traditional hosted testing tools share a major bottleneck: they require a public URL.

To crawl your application, generate test cases, or run an accessibility audit, most platforms demand you deploy first. You push code to a staging box, wait for the CI/CD pipeline to finish, and then — finally — trigger your tests. This “deploy-to-test” loop is slow. It forces you to move code away from your local machine before you know if it’s actually broken.

AegisRunner changes this workflow. Our AI regression testing tool closes the gap between local development and cloud-scale automation. You can now point our autonomous AI crawler directly at localhost with a single CLI command. No public deploys. No staging URLs. No wait times.

One command, your localhost, real tests

Stop waiting for staging environments to spin up. Install the AegisRunner CLI and start testing the code running on your laptop right now. The CLI has zero dependencies and requires only Node 18+.

Get started in seconds

Run these lines to connect your local dev server to our cloud AI:

npm install -g @aegisrunner/cli
export AEGIS_TOKEN=aegis_xxxxxxxx     # Your project's CI token

aegis scan --tunnel --port 3000 --watch

The —tunnel flag opens a secure connection between your machine and the AegisRunner cloud. The —port 3000 flag tells the AI where your app is running. Finally, —watch streams the progress directly to your terminal.

Our AI doesn’t just ping your home page. It discovers every interactive element, submits forms, and maps every flow on your local branch. It identifies functional regressions, accessibility (A11y) violations, and SEO issues before the code ever leaves your machine.

Secure by design: the outbound-only tunnel

Security is the primary concern when connecting local environments to cloud services. AegisRunner uses a reverse HTTP relay architecture to keep your network protected.

Why the tunnel is safe

• Outbound-only connection: Your machine dials out to AegisRunner over HTTPS. You never open an inbound port on your router or firewall.
• Firewall & VPN compatible: Because the connection is outbound, it works seamlessly behind corporate firewalls, VPNs, and NAT.
• Ephemeral lifecycle: The tunnel exists only while the command is running. The moment the scan finishes or you kill the process, the connection terminates.
• Scoped subdomains: Every tunnel receives a unique, temporary subdomain, so root-relative links like /dashboard or /settings stay scoped strictly to your application.

This keeps your local environment invisible to the public internet while remaining fully accessible to our AI regression testing tool.

Authenticated testing: scanning behind the login

Most modern web applications are hidden behind a login screen, and a crawler that can’t authenticate is useless for testing the core of your app. AegisRunner handles authentication automatically by mapping your credentials to the real fields on your login form.

Handing credentials to the CLI

You can pass authentication details directly through the CLI without leaving sensitive passwords in your shell history:

# Pass the password from an env var via stdin (never in argv or history)
printf %s "$DEV_PW" | aegis scan --tunnel --port 3000 \
  --username dev@example.com --password-stdin --watch

Our AI analyzes your login page, identifies the correct input fields, and authenticates the session. If you have complex roles (e.g. Admin vs. Viewer), use the —role flag to scan as a specific stored role from your project settings — so you can verify that a new feature on your local branch doesn’t break the experience for a given permission set.

Build the pre-deploy loop

The ultimate goal of local testing is to stop bugs from reaching shared environments. By wiring the AegisRunner CLI into your local workflow, you create a “pre-deploy loop.”

Block bad pushes with status codes

The CLI exits with a standard status code, so you can use it to block a git push or a deployment script when the AI finds a regression:

# Prevent deployment if the local scan fails
aegis scan --tunnel --port 3000 --watch || exit 1

Export to clean Playwright scripts

AegisRunner doesn’t lock you into a proprietary format. Every test case generated during a local scan exports to production-ready Playwright TypeScript:

1. Run a local scan via the CLI.
2. Review the generated test suite in the AegisRunner dashboard.
3. Export the suite to Playwright.
4. Run npx playwright test locally to debug or customize the scripts.

That is the best of both worlds: autonomous regression testing that produces standard, maintainable code.

Beyond localhost: reaching private staging

The same CLI and tunnelling technology reaches any environment the cloud can’t see.

• IP-restricted staging: If your staging box is restricted to specific IPs, use the tunnel — or add our static egress IPs to your allowlist.
• Air-gapped environments: For highly sensitive networks, run a self-hosted runner. It sits inside your network, claims jobs over an outbound connection, and executes scans locally without traffic ever leaving your infrastructure. See Testing behind a firewall.

Start testing before you deploy

Setup takes minutes, not weeks. Point our AI regression testing tool at your local dev server today and catch regressions before your users do.

1. Install: npm install -g @aegisrunner/cli
2. Scan: aegis scan —tunnel —port [your-port] —watch
3. Review: watch the AI discover pages and generate tests live.

Start for free — no credit card required.