Pentesting Web checklist
- Large: a whole company with multiple domains
- Medium: a single domain
- Small: a single website
- Subdomains recursive search
- Try to locate
/robots.txt,/crossdomain.xml/clientaccesspolicy.xml/sitemap.xmland/.well-known/ - Review comments on source code (Burp Engagement Tools)
- Locate admin and login panel
- Check ICMP packets allowed
- Study site structure
- Make a list with all possible test cases
- Understand the business area and what their customer needs
- Get a list of every asset (all_subdomains.txt, live_subdomains.txt, waybackurls.txt, hidden_directories.txt, nmap_results.txt, GitHub_search.txt, altdns_subdomain.txt, vulnerable_links.txt, js_files.txt)
- Duplicate registration (try with uppercase, [email protected], dots in name, etc)
- Overwrite existing user (existing user takeover)
- Username uniqueness
- Weak password policy (user=password, password=123456,111111,abcabc,qwerty12)
- Weak registration implementation or allows disposable email addresses
- Fuzz after user creation to check if any folder have been overwritten or created with your profile name
- Add only spaces in password
- Long password (>200) leads to DoS
- Corrupt authentication and session defects: Sign up, don't verify, request change password, change, check if account is active.
- Try to re-register repeating same request with same password and different password too
- If JSON request, add comma {“email”:“[email protected]”,”[email protected]”,“token”:”xxxxxxxxxx”}
- Lack of confirmation -> try to register with company email.
- Check OAuth with social media registration
- Check state parameter on social media registration
- Try to capture integration url leading integration takeover
- Check redirections in register page after login
- Rate limit on account creation
- XSS on name or email
- Username enumeration
- Resilience to password guessing
- Account recovery function
- "Remember me" function
- Impersonation function
- Unsafe distribution of credentials
- Fail-open conditions
- Multi-stage mechanisms
- Auto-complete testing
- Lack of password confirmation on change email, password or 2FA (try change response)
- Weak login function over HTTP and HTTPS if both are available
- User account lockout mechanism on brute force attack
- In OTP check guessable codes and race conditions
- OTP, check response manipulation for bypass
- OTP, try bruteforce
- Browser cache weakness (eg Pragma, Expires, Max-age)
- After register, logout, clean cache, go to home page and paste your profile url in browser, check for "login?next=accounts/profile" for open redirect or XSS with "/login?next=javascript:alert(1);//"
- Session handling
- Test tokens for meaning
- Test tokens for predictability
- Insecure transmission of tokens
- Disclosure of tokens in logs
- Mapping of tokens to sessions
- Session termination
- Session fixation
- Cookie scope
- Decode Cookie (Base64, hex, URL etc.)
- Cookie expiration time
- Check HTTPOnly and Secure flags
- Use same cookie from a different effective IP address or system
- Access controls
- Effectiveness of controls using multiple accounts
- Insecure access control methods (request parameters, Referer header, etc)
- Check for concurrent login through different machine/IP
- Weak generated security questions
- Path traversal on cookies
- Reuse cookie after session closed
- Logout and click browser "go back" function (Alt + Left arrow)
- 2 instances open, 1st change or reset password, refresh 2nd instance
- With privileged user perform privileged actions, try to repeat with unprivileged user cookie.
- Find parameter with user id and try to tamper in order to get the details of other users
- Change email id and update with any existing email id. Check if its getting validated on server or not.
- Check any new email confirmation link and what if user doesn't confirm.
- CSV import/export: Command Injection, XSS, macro injection
- Imagetragick in picture profile upload
- Account deletion option and try to reactivate with "Forgot password" feature
- Try bruteforce enumeration when change any user unique parameter.
- Check application request re-authentication for sensitive operations
- Try parameter pollution to add two values of same field
- Check different roles policy
- Invalidate session on Logout and Password reset
- Uniqueness of forget password reset link/code
- Reset links expiration time
- Find user id or other sensitive fields in reset link and tamper them
- Request 2 reset passwords links and use the older
- Check if many requests have sequential tokens
- Use [email protected]_collab.net and analyze the callback
- Host header injection for token leakage
- Add X-Forwarded-Host: evil.com to receive the reset link with evil.com
- Email crafting like [email protected]@target.com
- IDOR in reset link
- Capture reset token and use with other email/userID
- No TLD in email parameter
- User carbon copy [email protected]%0a%0dcc:[email protected]
- Long password (>200) leads to DoS
- No rate limit, capture request and send over 1000 times
- Check encryption in reset password token
- Token leak in referer header
- Append second email param and value
- Understand how token is generated (timestamp, username, birthdate,...)
- Response manipulation
- Fuzz all request parameters (if got user, add headers to fuzzer)
- Identify all reflected data
- RCE via Referer Header
- SQL injection via User-Agent Header
- Arbitrary redirection
- Stored attacks
- OS command injection
- Script injection
- File inclusion
- SMTP injection
- Native software flaws (buffer overflow, integer bugs, format strings)
- SOAP injection
- LDAP injection
- SSI Injection
- XPath injection
- Code Injection (<h1>six2dez</h1> on stored param)
- xmlrpc.php DOS and user enumeration
- HTTP dangerous methods OPTIONS PUT DELETE
- Insecure deserialization
- Access custom pages like /whatever_fake.php (.aspx,.html,.etc)
- Add multiple parameters in GET and POST request using different values
- Add "[]", "]]", and "[[" in cookie values and parameter values to create errors
- Generate error by giving input as "/~randomthing/%s" at the end of URL
- Use Burp Intruder "Fuzzing Full" List in input to generate error codes
- Try different HTTP Verbs like PATCH, DEBUG or wrong like FAKE
- Identify the logic attack surface
- Test transmission of data via the client
- Test for reliance on client-side input validation
- Thick-client components (Java, ActiveX, Flash)
- Multi-stage processes for logic flaws
- Handling of incomplete input
- Trust boundaries
- Transaction logic
- Implemented CAPTCHA in email forms to avoid flooding
- Tamper product id, price or quantity value in any action (add, modify, delete, place, pay...)
- Tamper gift or discount codes
- Reuse gift codes
- Try parameter pollution to use gift code two times in same request
- Try stored XSS in non-limited fields like address
- Check in payment form if CVV and card number is in clear text or masked
- Check if is processed by the app itself or sent to 3rd parts
- IDOR from other users details ticket/cart/shipment
- Check PRINT or PDF creation for IDOR
- Check unsubscribe button with user enumeration
- Parameter pollution on social media sharing links
- Change POST sensitive requests to GET
- Segregation in shared infrastructures
- Segregation between ASP-hosted applications
- Web server vulnerabilities
- Dangerous HTTP methods
- Proxy functionality
- Check for internal numeric IP's in request
- Check for external numeric IP's and resolve it
- Check the existence of alternative channels (www.web.com vs m.web.com)
- Send old captcha value.
- Send old captcha value with old session ID.
- Request captcha absolute path like www.url.com/captcha/1.png
- Remove captcha with any adblocker and request again
- Change from POST to GET
- Remove captcha parameter
- Convert JSON request to normal
- Try header injections
- X-XSS-Protection
- Strict-Transport-Security
- Content-Security-Policy
- Public-Key-Pins
- X-Frame-Options
- X-Content-Type-Options
- Referer-Policy
- Cache-Control
- Expires
Last modified 5mo ago