Security/Features/PasswordManagerImprovements: Difference between revisions

Jump to navigation Jump to search

Security/Features/PasswordManagerImprovements (view source)

Revision as of 23:13, 8 January 2015

720 bytes removed ,  8 January 2015
no edit summary
No edit summary
No edit summary
 
(6 intermediate revisions by the same user not shown)
Line 11: Line 11:
|Feature users and use cases=* See https://www.usenix.org/conference/usenixsecurity14/technical-sessions/presentation/silver
|Feature users and use cases=* See https://www.usenix.org/conference/usenixsecurity14/technical-sessions/presentation/silver
* Includes improvements to the Password Manager to prevent attacks such as "Lupin" in the paper [http://arxiv.org/abs/1309.1416 "Automated Password Extraction Attack on Modern Password Managers"].  When a user is on an insecure network and makes any http request, a mitm could add login iframes to the http request.  The mitm can then add javascript to the responses from the login pages that send the attacker the autocompleted passwords from password manager.
* Includes improvements to the Password Manager to prevent attacks such as "Lupin" in the paper [http://arxiv.org/abs/1309.1416 "Automated Password Extraction Attack on Modern Password Managers"].  When a user is on an insecure network and makes any http request, a mitm could add login iframes to the http request.  The mitm can then add javascript to the responses from the login pages that send the attacker the autocompleted passwords from password manager.
|Feature requirements=i) Preventing active network attacks:
|Feature requirements=1) Preventing active network attacks:
*# Do not autofill the username/password stored in password manager on any pages. Provide alternative UX - <b>UX help needed</b>.  If automatically filling the password must be an option, then at least do not fill in the following cases.  (Note this does not secure the user from xss attacks, third party javascript, etc.)
* [https://bugzilla.mozilla.org/show_bug.cgi?id=1118511 Bug 1118511] Do not autofill the username/password stored in password manager on any pages. Provide alternative UX - <b>UX help needed</b>.  If automatically filling the password must be an option, then at least do not fill in the following cases.  (Note this does not secure the user from xss attacks, third party javascript, etc.)
*#* For http sites (IE 11 has this security feature)
** For http sites (IE 11 has this security feature)
*#* https sites that have mixed active content
** https sites that have mixed active content
*#* https sites that require a cert override (chrome does this)
** https sites that require a cert override (chrome does this)
*#* in iframed sites (where the parent and frame are not  same orign, or always?) (safari does this for non-same origin, chrome does this for all frames). (Open Issue - what about third party widgets that allow users to login and post comments.)
** in iframed sites (where the parent and frame are not  same orign, or always?) (safari does this for non-same origin, chrome does this for all frames). (Open Issue - what about third party widgets that allow users to login and post comments.)
*#* Invisible form fields (visibility and opacity, although this isn't going to prevent clickjacking attacks to autofill the passwords)
** Invisible form fields (visibility and opacity, although this isn't going to prevent clickjacking attacks to autofill the passwords)
*# Warn users when they are entering their passwords on HTTP sites. UX help needed for this. Some options:
* [https://bugzilla.mozilla.org/show_bug.cgi?id=748193 Bug 748193] Warn users when they are entering their passwords on HTTP sites. UX help needed for this. Some options:
*#* warning icon in the password field
** warning icon in the password field
*#* Fill-and-submit button is a different color
** Fill-and-submit button is a different color
*#* On mouseover of the fill in submit button, the user can read a tooltip that warns them that their password can be seen in the clear.
** On mouseover of the fill in submit button, the user can read a tooltip that warns them that their password can be seen in the clear.
*#* See also [https://wiki.mozilla.org/Security/Features/HighlightCleartextPasswords Highlight Cleartext Passwords] and [https://bugzilla.mozilla.org/show_bug.cgi?id=759860 Bug 759860].
** [https://bugzilla.mozilla.org/show_bug.cgi?id=1118558 Bug 1118558] Flag in the Password Manager User Interface that shows all saved logins.
*# Secure Filling (platform work only) - Passwords that are saved by the password manager should not be available to javascript on the page.  The actual password values should only be sent on submit.  This protects the password from attacks via xss, 3rd party javascript, etc.  Implementation details: when a password is filled in on a form, fill hash(uri, username, salt) instead of the actual password.  On submit, lookup the actual password value for that url and send that instead.  Username is included for cases where there are multiple usernames.
** See also [https://wiki.mozilla.org/Security/Features/HighlightCleartextPasswords Highlight Cleartext Passwords].
ii) Preventing local attacks:
* [https://bugzilla.mozilla.org/show_bug.cgi?id=1118540 Bug 1118540] Secure Filling 1.0 - Passwords that are saved by the password manager should not be available to javascript on the page.  The actual password values should only be sent on submit.  This protects the password from attacks via xss, 3rd party javascript, etc.  Implementation details: when a password is filled in on a form, fill hash(uri, username, salt) instead of the actual password.  On submit, lookup the actual password value for that url and send that instead.  Username is included for cases where there are multiple usernames.
** Explore solutions for encrypting the passwords stored locally in the password manager (for example, make use of keychain or encryption mechanisms that come with the OS).
2) Preventing local attacks:
iii) Duplicate Passwords - Protecting users from password reuse attacks
* [https://bugzilla.mozilla.org/show_bug.cgi?id=1118549 Bug 1118549] Encryption - Explore solutions for encrypting the passwords stored locally in the password manager (for example, make use of keychain or encryption mechanisms that come with the OS).
** Create UI around alerting users that they are reusing the same passwords
3) [https://bugzilla.mozilla.org/show_bug.cgi?id=1118553 Bug 1118553] Duplicate Passwords - Protecting users from password reuse attacks
iv) HSTS (platform work only): If a site is HSTS, then there is no reason to have http data for that site. Hence, if a site is marked HSTS, and the user has any data (cookies, passwords, etc) that are not https-only/secure, immediately mark that data as https-only. (Note that we'd need some way to indicate that the site has been STS for at least X weeks to prevent deleting data from a site that goes HSTS as a beta test and then goes back to non-HSTS.)
* Create UI around alerting users that they are reusing the same passwords
4) [https://bugzilla.mozilla.org/show_bug.cgi?id=1119555 Bug 1119555] HSTS: If a site is HSTS, then there is no reason to have http data for that site. Hence, if a site is marked HSTS, and the user has any data (cookies, passwords, etc) that are not https-only/secure, immediately mark that data as https-only. (Note that we'd need some way to indicate that the site has been STS for at least X weeks to prevent deleting data from a site that goes HSTS as a beta test and then goes back to non-HSTS.)


<b>If we have to keep Autofill:</b>
<b>More about Autofill:</b>
If we must allow autofilling in certain cases, then we also need the following security features:
If we must allow autofilling in certain cases, then we also need the following security features:
* If an identical password is stored for both the http version and https version of a specific website (or domain), and it is not used on the http site for X months, expire the http version after alerting the user. (This can help in cases where the website has upgraded to https, but the user's http password manager entry still exists and is open to attack).
* If an identical password is stored for both the http version and https version of a specific website (or domain), and it is not used on the http site for X months, expire the http version after alerting the user. (This can help in cases where the website has upgraded to https, but the user's http password manager entry still exists and is open to attack).


* (platform work only) When the password field name is different from the name when the password was saved, don't allow filling & submit.  Moreover, don't allow javascript to dynamically change the name of the form??  See [https://www.usenix.org/system/files/conference/usenixsecurity14/sec14-paper-silver.pdf section 5.2, Preventing self exfiltration attacks].
* When the password field name is different from the name when the password was saved, don't allow filling & submit.  Moreover, don't allow javascript to dynamically change the name of the form??  See [https://www.usenix.org/system/files/conference/usenixsecurity14/sec14-paper-silver.pdf section 5.2, Preventing self exfiltration attacks].


* (platform work only) Consider not autofilling in cases where the page has multiple login forms.
* Consider not autofilling in cases where the page has multiple login forms.




<b>Future Work:</b>
<b>Future Work:</b>


* Support [https://mikewest.github.io/credentialmanagement/spec/ Credential Management Specification] so websites can opt into better detection and protection (platform work only).
* Support [https://mikewest.github.io/credentialmanagement/spec/ Credential Management Specification] so websites can opt into better detection and protection.


* Prefer secure origins - If a password is stored in an http version of a site, see if the https version exists.  If it does, prompt the user to redirect to the https version of the site and store their password there instead. (Issue here is that we don't always know if changing the url to https will work, or if a site is set up to have a different domain or path for their secure version)
* Prefer secure origins - If a password is stored in an http version of a site, see if the https version exists.  If it does, prompt the user to redirect to the https version of the site and store their password there instead. (Issue here is that we don't always know if changing the url to https will work, or if a site is set up to have a different domain or path for their secure version)


 
* [https://bugzilla.mozilla.org/show_bug.cgi?id=1118540 Bug 1118540] Secure Filling 2.0
 
<b>Investigate more:</b>
* Only autofill forms after user interaction with the login page, not necessarily the form - https://code.google.com/p/chromium/issues/detail?id=257156.  We need more details here.
 
* Do not autofill (or even fill-and-submit for HTTP pages?) if the form action has changed since the password was stored (perhaps with a user override option). (To do this, we need to include the form action in the key, along with the url.  We'd need to check the action both on load and on submit.  What if it's a javascript action?  More investigation needed here.  We already do some flavor of this.)
* Assume we have implemented secure filling (javascript on the page can't read the password).  If the user prompts the password manager to fill in a password on an HTTP page and the form action has changed since the password was stored, don't send the password (might be tricky to implement).  Perhaps with an UI for user override. (Secure autofilling causes issues with ajax logins, a technique used mostly in china per https://www.usenix.org/conference/usenixsecurity14/technical-sessions/presentation/silver)
 
* Secure Filling 2.0 (platform work only)
** Do not give javascript access to any password fields (regardless of whether the password manager saves the password) - the actual password the user enters is only used on submit.  The problem with this is with registration pages that use javascript to test the password strength.  [https://bugzilla.mozilla.org/show_bug.cgi?id=653132 Bug 653132].  Can we detect registration pages?  If a page a registration page has only one password field (they don't ask you to confirm your password by entering it twice) do they really use javascript to test password strength?  Since they aren't asking you to confirm your password, they probably aren't too concerned with special characters.
** Do not give javascript access to any password fields (regardless of whether the password manager saves the password) - the actual password the user enters is only used on submit.  The problem with this is with registration pages that use javascript to test the password strength.  [https://bugzilla.mozilla.org/show_bug.cgi?id=653132 Bug 653132].  Can we detect registration pages?  If a page a registration page has only one password field (they don't ask you to confirm your password by entering it twice) do they really use javascript to test password strength?  Since they aren't asking you to confirm your password, they probably aren't too concerned with special characters.
|Feature ux design=* What should the UX be when we do not autofill?
|Feature ux design=* What should the UX be when we do not autofill?
Tanvi
canmove, Confirmed users
285

edits

Retrieved from "https://wiki.mozilla.org/Special:MobileDiff/1038779...1045876"

Navigation menu