Creator plans in preparation · Scanning and payments are not open
A subtitle tool, a member site, your first small service.
Understand security risks in plain language, get fix tasks for your coding AI and check again after changes.
Website you entered:
Only the homepage will be checked, not any subpage in the address you paste.
Free checks need no sign-in or DNS setup. Once available, confirm permission and complete the CAPTCHA to start.
View sample report without signing in Contact us to enable this test addressAnonymous reports are available only in the original browser for up to 24 hours. Sign in only if you want to save them.
Start with your own tool · Plain-language Chinese reports · Clear tested and untested coverage
This is the planned workflow. Confirm permission for a free check; website control verification is required only for deep checks.
01
Confirm permission for the free check, without DNS setup or a verification file. Deep checks require separate website control verification.
02
See what data or costs could be affected, what we observed and what remains unconfirmed.
03
Copy a task with fix goals and acceptance checks, then review the changes. Aegis does not deploy them for you.
04
Compare before-and-after evidence. Failed checks or changed scope are marked unconfirmed.
Planned checks depend on the supported platform and the information you authorize.
Observable HTTPS, response settings and public exposure. This is not a full penetration test.
Identify suspected secrets and mask evidence. Public-use keys are not automatically classified as leaks.
The planned deep check reviews limited snapshots and dependencies without running your code or arbitrary commands.
Review authorized settings. Without backend evidence, we do not claim private data or login permissions have passed.
Results are planned to use LabSpace Intelligence. Self-service results are not a human audit; human review is quoted separately.
Initial plan proposal · Not available for purchase · No subscription required
Understand risks observable from the URL
NT$0
Main plan · In preparation, not yet for sale
For small tools preparing to go public
NT$499per check
When you need an engineer's judgment
Quoted by scope
Charges start only after deep-check and retest acceptance. Supported platforms, project limits and data handling are confirmed before purchase. Results are not a security guarantee.
Explain who could access what data or misuse which services, with traceable evidence.
Package fix goals and acceptance checks for your coding AI.
Distinguish observations, inferences and untested areas. A single score cannot hide the unknown.
A working website does not prove that its data and permissions are protected. Separate public settings, account access and transaction flows to understand what remains untested.
These observations cover one limited response from the public homepage. The check does not read the page body, sign in or modify data. No findings do not prove safety; if no cookie is returned, login-cookie protection cannot be determined.
Checking whether one member can read another member's data requires authorization, a dedicated test environment and synthetic accounts. Homepage headers cannot answer this question; this deeper assessment is still planned.
Determining whether private videos are exposed or paid API secrets are leaked requires authorized settings or a limited source snapshot. The current free check does not inspect these inputs or verify file-access permissions.
Order isolation, server-side pricing and duplicate-charge behavior need an isolated environment and test data. They are outside the seven URL observations and are not yet available for self-service assessment.
Illustrative example, not a measured report for any website
Suppose a homepage response lacks nosniff: first verify correct Content-Type values for HTML, CSS, JavaScript and uploaded files, then configure X-Content-Type-Options: nosniff. Do not disable existing protections just to pass a check.
For verification, confirm that the original URL returns the header and that the page and its resources still load correctly. This verifies the setting only, not fixes for XSS, authentication or authorization.
Aegis is designed for individuals building websites and small tools, with or without AI, to understand security risks, actionable fixes and retest results before launch.
Initial checks and report saving have been verified on controlled UAT test sites. General public website scanning and live payments are not yet available; entering a URL does not mean a scan has started.
Aegis is designed to check websites and tools you build, rather than protect your computer from malware. The current controlled HTTP engine checks response headers and cookie settings only. It is not a full penetration test or a guarantee of website security.
The planned offering is a free initial check and a NT$499 one-time deep check with two same-scope retests within 14 days. Human assistance is quoted separately. These are planned terms; purchasing is not open and full coverage is not yet available.
Current controlled-case reports include findings, original evidence, plain-language impact and fix tasks for your development AI, with tested, skipped and failed coverage. Explanations support Traditional Chinese and English while evidence stays in its original form. Reports are not human-reviewed.
The planned free check covers HTTPS, security headers and cookie settings on the public homepage. Confirm that you have permission; no verification file or DNS setup is required. Deep checks still require website control verification. Public scanning is not available yet. Do not submit third-party websites, passwords, keys or personal data lists.
No. Once available, paste your website URL, confirm permission and complete the CAPTCHA to run a check without registering. Reports remain in the original browser for up to 24 hours. Sign in only to save a report or use deep checks. Deep checks also require website control verification. Public checks are not available yet.
AI code review can suggest fixes, but deployed responses and behavior still need verification. Aegis aims to connect reproducible observations, repair tasks and rechecks. The current free check covers only seven external observations and does not replace code review or deeper authorization tests.
These seven checks cannot determine launch readiness on their own. Authentication, member-data isolation, source code, databases, dependencies and other pages are outside the free check. No findings do not mean that these areas passed.
The free check is designed to require a URL, authorization confirmation and a CAPTCHA, without registration or DNS setup. Sign in only to save the report. Deeper assessments separately confirm site control, test accounts and supported scope. General public access is not yet open.
Interested in the pilot or need help? Leave a message. We will send an acknowledgment and our team will respond.
You can also email[email protected]
