Fix/deceptive url security

src/components/ProfileSnippet.tsx

@@ -10,6 +10,7 @@ import { useAbilities } from '@app/hooks/abilities/useAbilities'
 import { useBeautifiedName } from '@app/hooks/useBeautifiedName'
 import { useEnsAvatar } from '@app/hooks/useEnsAvatar'
 import { useRouterWithHistory } from '@app/hooks/useRouterWithHistory'
+import { isDeceptiveUrl, getUrlDisplayText } from '@app/utils/security/validateUrl'
 
 import { useTransactionFlow } from '../transaction-flow/TransactionFlowProvider'
 import { NameAvatar } from './AvatarWithZorb'
@@ -180,7 +181,8 @@ export const getUserDefinedUrl = (url?: string) => {
   if (url.startsWith('http://') || url.startsWith('https://')) {
     return url
   }
-  return ``
+  // Don't add protocol to URLs that don't have one
+  return undefined
 }
 
 export const ProfileSnippet = ({
@@ -304,11 +306,23 @@ export const ProfileSnippet = ({
               </Typography>
             )}
             {url && (
-              <a href={url} data-testid="profile-snippet-url" target="_blank" rel="noreferrer">
-                <Typography color="blue" id="profile-url">
-                  {url?.replace(/http(s?):\/\//g, '').replace(/\/$/g, '')}
+              isDeceptiveUrl(url) ? (
+                <Typography
+                  color="greyPrimary"
+                  id="profile-url"
+                  data-testid="profile-snippet-url-disabled"
+                  style={{ textDecoration: 'line-through', cursor: 'not-allowed' }}
+                  title="This URL contains deceptive patterns and has been disabled for security"
+                >
+                  {getUrlDisplayText(url)}
                 </Typography>
-              </a>
+              ) : (
+                <a href={url} data-testid="profile-snippet-url" target="_blank" rel="noreferrer">
+                  <Typography color="blue" id="profile-url">
+                    {getUrlDisplayText(url)}
+                  </Typography>
+                </a>
+              )
             )}
           </LocationAndUrl>
         )}

src/components/pages/profile/ProfileButton.tsx

@@ -33,6 +33,7 @@ import { getSocialData } from '@app/utils/getSocialData'
 import { makeEtherscanLink, shortenAddress } from '@app/utils/utils'
 import { getVerifierData } from '@app/utils/verification/getVerifierData'
 import { isVerificationProtocol } from '@app/utils/verification/isVerificationProtocol'
+import { isDeceptiveUrl } from '@app/utils/security/validateUrl'
 
 const StyledAddressIcon = styled(DynamicAddressIcon)(
   ({ theme }) => css`
@@ -88,6 +89,10 @@ export const SocialProfileButton = ({
   )
 
   if (!socialData) return null
+
+  // Check if the URL is deceptive
+  const isDeceptive = socialData.type === 'link' && isDeceptiveUrl(socialData.urlFormatter)
+
   return (
     <VerificationBadge
       isVerified={isVerified}
@@ -101,9 +106,15 @@ export const SocialProfileButton = ({
         inline
         data-testid={`social-profile-button-${iconKey}`}
         value={socialData.value}
-        {...(socialData.type === 'link'
+        {...(socialData.type === 'link' && !isDeceptive
           ? { as: 'a' as const, link: socialData.urlFormatter }
           : { as: 'button' as const })}
+        style={isDeceptive ? {
+          textDecoration: 'line-through',
+          opacity: 0.5,
+          cursor: 'not-allowed'
+        } : {}}
+        title={isDeceptive ? 'This URL contains deceptive patterns and has been disabled for security' : undefined}
       >
         {socialData.value}
       </RecordItem>
@@ -230,17 +241,24 @@ export const OtherProfileButton = ({
       if (!decodedContentHash) return {}
       const _link = getContentHashLink({ name, chainId, decodedContentHash })
       if (!_link) return {}
+      // Check if contenthash link is deceptive
+      if (isDeceptiveUrl(_link)) return {}
       return {
         as: 'a',
         link: _link,
       } as const
     }
+    // Check if regular URL is deceptive
+    if (isDeceptiveUrl(value)) return {}
     return {
       as: 'a',
       link: value,
     } as const
   }, [isLink, type, value, name, chainId])
 
+  // Determine if the URL is deceptive for styling
+  const isDeceptive = isLink && !linkProps.link && !linkProps.as
+
   return (
     <RecordItem
       {...linkProps}
@@ -259,6 +277,12 @@ export const OtherProfileButton = ({
         )
       }
       data-testid={`other-profile-button-${iconKey}`}
+      style={isDeceptive ? {
+        textDecoration: 'line-through',
+        opacity: 0.5,
+        cursor: 'not-allowed'
+      } : {}}
+      title={isDeceptive ? 'This URL contains deceptive patterns and has been disabled for security' : undefined}
     >
       {formattedValue}
     </RecordItem>

src/utils/security/validateUrl.test.ts

@@ -0,0 +1,111 @@
+import { describe, it, expect } from 'vitest'
+import { isDeceptiveUrl, getSafeUrl, getUrlDisplayText } from './validateUrl'
+
+describe('isDeceptiveUrl', () => {
+  it('should return false for valid URLs', () => {
+    expect(isDeceptiveUrl('https://google.com')).toBe(false)
+    expect(isDeceptiveUrl('https://app.ens.domains')).toBe(false)
+    expect(isDeceptiveUrl('http://example.com')).toBe(false)
+  })
+
+  it('should return true for URLs with @ symbol (userinfo)', () => {
+    expect(isDeceptiveUrl('https://google.com@evil.com')).toBe(true)
+    expect(isDeceptiveUrl('https://user@evil.com')).toBe(true)
+    expect(isDeceptiveUrl('https://user:pass@evil.com')).toBe(true)
+  })
+
+  it('should return true for URLs with Unicode spaces', () => {
+    // EM SPACE
+    expect(isDeceptiveUrl('https://google.com\u2003@evil.com')).toBe(true)
+    // ZERO WIDTH SPACE
+    expect(isDeceptiveUrl('https://google.com\u200b@evil.com')).toBe(true)
+    // NO-BREAK SPACE
+    expect(isDeceptiveUrl('https://google.com\u00a0@evil.com')).toBe(true)
+  })
+
+  it('should return true for javascript: and data: URIs', () => {
+    expect(isDeceptiveUrl('javascript:alert(1)')).toBe(true)
+    expect(isDeceptiveUrl('Javascript:alert(1)')).toBe(true)
+    expect(isDeceptiveUrl('data:text/html,<script>alert(1)</script>')).toBe(true)
+    expect(isDeceptiveUrl('DATA:text/html,<h1>test</h1>')).toBe(true)
+  })
+
+  it('should return true for file:// protocol', () => {
+    expect(isDeceptiveUrl('file:///etc/passwd')).toBe(true)
+    expect(isDeceptiveUrl('FILE:///c:/windows/system.ini')).toBe(true)
+  })
+
+  it('should return true for URLs with null bytes', () => {
+    expect(isDeceptiveUrl('https://example.com\0/evil')).toBe(true)
+  })
+
+  it('should return true for URLs with multiple slashes after protocol', () => {
+    expect(isDeceptiveUrl('https:///evil.com')).toBe(true)
+    expect(isDeceptiveUrl('http:////evil.com')).toBe(true)
+  })
+
+  it('should return true for punycode domains', () => {
+    expect(isDeceptiveUrl('https://xn--e1afmkfd.xn--p1ai')).toBe(true)
+    expect(isDeceptiveUrl('https://xn--80ak6aa92e.com')).toBe(true)
+  })
+
+  it('should return true for mixed Cyrillic and Latin characters', () => {
+    // аррӏе.com (Cyrillic 'a' and Latin)
+    expect(isDeceptiveUrl('https://аррӏе.com')).toBe(true)
+  })
+
+  it('should return true for invalid URLs', () => {
+    expect(isDeceptiveUrl('not-a-url')).toBe(true)
+    expect(isDeceptiveUrl('ftp://example.com')).toBe(true)
+  })
+
+  it('should handle undefined and empty strings', () => {
+    expect(isDeceptiveUrl(undefined)).toBe(false)
+    expect(isDeceptiveUrl('')).toBe(false)
+  })
+})
+
+describe('getSafeUrl', () => {
+  it('should return the URL if it is safe', () => {
+    expect(getSafeUrl('https://google.com')).toBe('https://google.com')
+    expect(getSafeUrl('http://example.com')).toBe('http://example.com')
+  })
+
+  it('should return undefined for deceptive URLs', () => {
+    expect(getSafeUrl('https://google.com@evil.com')).toBeUndefined()
+    expect(getSafeUrl('javascript:alert(1)')).toBeUndefined()
+    expect(getSafeUrl('file:///etc/passwd')).toBeUndefined()
+  })
+
+  it('should return undefined for URLs without protocol', () => {
+    expect(getSafeUrl('google.com')).toBeUndefined()
+    expect(getSafeUrl('www.example.com')).toBeUndefined()
+  })
+
+  it('should handle undefined', () => {
+    expect(getSafeUrl(undefined)).toBeUndefined()
+  })
+})
+
+describe('getUrlDisplayText', () => {
+  it('should return hostname and path without protocol', () => {
+    expect(getUrlDisplayText('https://google.com')).toBe('google.com')
+    expect(getUrlDisplayText('https://example.com/path')).toBe('example.com/path')
+    expect(getUrlDisplayText('http://test.com/foo/bar')).toBe('test.com/foo/bar')
+  })
+
+  it('should remove trailing slashes', () => {
+    expect(getUrlDisplayText('https://google.com/')).toBe('google.com')
+    expect(getUrlDisplayText('https://example.com/path/')).toBe('example.com/path')
+  })
+
+  it('should handle invalid URLs gracefully', () => {
+    expect(getUrlDisplayText('not-a-url')).toBe('not-a-url')
+    expect(getUrlDisplayText('google.com')).toBe('google.com')
+  })
+
+  it('should handle undefined and empty strings', () => {
+    expect(getUrlDisplayText(undefined)).toBe('')
+    expect(getUrlDisplayText('')).toBe('')
+  })
+})

src/utils/security/validateUrl.ts

@@ -0,0 +1,117 @@
+/**
+ * Validates URLs for security issues and deceptive patterns
+ * Returns true if URL should be considered invalid/unsafe
+ */
+export const isDeceptiveUrl = (url: string | undefined): boolean => {
+  if (!url) return false
+
+  try {
+    // 1. Block userinfo pattern (@ symbol abuse like https://user@evil.com)
+    if (url.includes('@')) {
+      return true
+    }
+
+    // 2. Block Unicode spaces and invisible characters that could hide @ or other deceptive patterns
+    const suspiciousChars = [
+      '\u2003', // EM SPACE
+      '\u200B', // ZERO WIDTH SPACE
+      '\u00A0', // NO-BREAK SPACE
+      '\u2000', '\u2001', '\u2002', '\u2004', '\u2005', // Various Unicode spaces
+      '\u2006', '\u2007', '\u2008', '\u2009', '\u200A',
+      '\uFEFF', // ZERO WIDTH NO-BREAK SPACE
+      '\u202E', // RIGHT-TO-LEFT OVERRIDE
+      '\u202D', // LEFT-TO-RIGHT OVERRIDE
+    ]
+
+    if (suspiciousChars.some((char) => url.includes(char))) {
+      return true
+    }
+
+    // 3. Block data: and javascript: URIs (XSS attempts)
+    const lowerUrl = url.toLowerCase()
+    if (lowerUrl.startsWith('data:') || lowerUrl.startsWith('javascript:')) {
+      return true
+    }
+
+    // 4. Block file:// protocol (local file access)
+    if (lowerUrl.startsWith('file://')) {
+      return true
+    }
+
+    // 5. Block URLs with null bytes and control characters
+    if (url.includes('\0') || /[\x00-\x1F\x7F]/.test(url)) {
+      return true
+    }
+
+    // 6. Block multiple slashes after protocol (parser confusion)
+    if (url.match(/^https?:\/\/\/+/)) {
+      return true
+    }
+
+    // Parse URL for additional checks
+    const parsed = new URL(url)
+
+    // 7. Check for punycode domains (potential homograph attacks)
+    // xn-- prefix indicates punycode encoding
+    if (parsed.hostname.startsWith('xn--')) {
+      return true
+    }
+
+    // 8. Optional: Block common URL shorteners (uncomment if desired)
+    // const urlShorteners = ['bit.ly', 'tinyurl.com', 'goo.gl', 't.co', 'ow.ly', 'short.link']
+    // if (urlShorteners.includes(parsed.hostname.toLowerCase())) {
+    //   return true
+    // }
+
+    // 9. Check for mixed scripts in domain (e.g., Latin + Cyrillic)
+    // This helps detect homograph attacks
+    const hasCyrillic = /[\u0400-\u04FF]/.test(parsed.hostname)
+    const hasLatin = /[a-zA-Z]/.test(parsed.hostname)
+    if (hasCyrillic && hasLatin) {
+      return true
+    }
+
+    return false
+  } catch {
+    // If URL parsing fails, consider it invalid
+    return true
+  }
+}
+
+/**
+ * Sanitizes URL for safe display
+ * Returns a safe version of the URL or undefined if invalid
+ */
+export const getSafeUrl = (url: string | undefined): string | undefined => {
+  if (!url) return undefined
+
+  // If URL is deceptive, return undefined (don't make it clickable)
+  if (isDeceptiveUrl(url)) {
+    return undefined
+  }
+
+  // Ensure URL has a protocol
+  if (!url.startsWith('http://') && !url.startsWith('https://')) {
+    // Don't add protocol to already invalid URLs
+    return undefined
+  }
+
+  return url
+}
+
+/**
+ * Get display text for a URL (removes protocol for cleaner display)
+ */
+export const getUrlDisplayText = (url: string | undefined): string => {
+  if (!url) return ''
+
+  try {
+    const parsed = new URL(url)
+    // Return hostname + path without protocol
+    return `${parsed.hostname}${parsed.pathname === '/' ? '' : parsed.pathname}`
+      .replace(/\/$/, '') // Remove trailing slash
+  } catch {
+    // If parsing fails, return the original URL
+    return url.replace(/^https?:\/\//, '').replace(/\/$/, '')
+  }
+}