Open Authentication Protocol                         T. Lodderstedt, Ed.
Internet-Draft                                             YES Europe                                                YES.com AG
Intended status: Best Current Practice                        J. Bradley
Expires: March 14, May 17, 2018                                             Yubico
                                                             A. Labunets
                                                                Facebook
                                                      September 10,
                                                       November 13, 2017

                         OAuth Security Topics
                  draft-ietf-oauth-security-topics-03
                  draft-ietf-oauth-security-topics-04

Abstract

   This draft gives a comprehensive overview on open OAuth security
   topics.  It is intended to serve as a working document for the OAuth
   working group to systematically capture and discuss these security
   topics and respective mitigations and eventually recommend best
   current practice and also OAuth extensions needed to cope with the
   respective security threats.

Status of This Memo

   This Internet-Draft is submitted in full conformance with the
   provisions of BCP 78 and BCP 79.

   Internet-Drafts are working documents of the Internet Engineering
   Task Force (IETF).  Note that other groups may also distribute
   working documents as Internet-Drafts.  The list of current Internet-
   Drafts is at https://datatracker.ietf.org/drafts/current/.

   Internet-Drafts are draft documents valid for a maximum of six months
   and may be updated, replaced, or obsoleted by other documents at any
   time.  It is inappropriate to use Internet-Drafts as reference
   material or to cite them other than as "work in progress."

   This Internet-Draft will expire on March 14, May 17, 2018.

Copyright Notice

   Copyright (c) 2017 IETF Trust and the persons identified as the
   document authors.  All rights reserved.

   This document is subject to BCP 78 and the IETF Trust's Legal
   Provisions Relating to IETF Documents
   (https://trustee.ietf.org/license-info) in effect on the date of
   publication of this document.  Please review these documents
   carefully, as they describe your rights and restrictions with respect
   to this document.  Code Components extracted from this document must
   include Simplified BSD License text as described in Section 4.e of
   the Trust Legal Provisions and are provided without warranty as
   described in the Simplified BSD License.

Table of Contents

   1.  Introduction  . . . . . . . . . . . . . . . . . . . . . . . .   3
   2.  Recommended  Best Practice Practices  . . . . . . . . . . . . . . . . . . . . . . .   4
     2.1.  Protecting redirect-based flows . . . . . . . . . . . . .   4
     2.2.  TBD .  Token Leakage Prevention  . . . . . . . . . . . . . . . .   5
   3.  Recommended Changes to OAuth  . . . . . . . . . .   5
   3.  Recommended modifications and extensions to OAuth . . . . . .   5
   4.  OAuth Credentials Leakage  Attacks and Mitigations . . . . . . . . . . . . . . . . . . .   5
     4.1.  Insufficient redirect URI validation  . . . . . . . . . .   5
       4.1.1.  Attacks on Authorization Code Grant . . . . . . . . .   6
       4.1.2.  Attacks on Implicit Grant . . . . . . . . . . . . . .   7
       4.1.3.  Proposed Countermeasures  . . . . . . . . . . . . . .   8
     4.2.  Authorization code leakage via referrer headers . . . . .  10   9
       4.2.1.  Proposed Countermeasures  . . . . . . . . . . . . . .  10
     4.3.  Attacks in the Browser  . . . . . . . . . . . . . . . . .  10
       4.3.1.  Code in browser history (TBD) . . . . . . . . . . . .  11  10
       4.3.2.  Access token in browser history (TBD) . . . . . . . .  11  10
       4.3.3.  Javascript Code stealing Access Tokens (TBD)  . . . .  11
     4.4.  Access Token Leakage at the Resource Server . . . . . . .  11
       4.4.1.  Access Token Phishing by Counterfeit Resource Server   11
         4.4.1.1.  Metadata  . .  Mix-Up  . . . . . . . . . . . . . . . . . .  12
         4.4.1.2.  Sender Constrained Access Tokens . . . . . . .  11
     4.5.  Code Injection  .  13
         4.4.1.3.  Audience Restricted Access Tokens . . . . . . . .  15
       4.4.2.  Compromised Resource Server . . . . . . . . . . . .  11
       4.5.1.  Proposed Countermeasures  .  16
       4.4.3.  TLS Terminating Reverse Proxies . . . . . . . . . . .  17
     4.5.  Mix-Up . .  13
     4.6.  XSRF (TBD)  . . . . . . . . . . . . . . . . . . . . . . .  18
     4.6.  Refresh  15
     4.7.  Access Token Leakage at the Resource Server . . . . . . . . . . . . . . . . . .  18
   5.  OAuth Credentials Injection . .  15
       4.7.1.  Access Token Phishing by Counterfeit Resource Server   15
         4.7.1.1.  Metadata  . . . . . . . . . . . . . . .  19
     5.1.  Code Injection . . . . .  16
         4.7.1.2.  Sender Constrained Access Tokens  . . . . . . . .  17
         4.7.1.3.  Audience Restricted Access Tokens . . . . . . . .  19
       5.1.1.  Proposed Countermeasures  .
       4.7.2.  Compromised Resource Server . . . . . . . . . . . . .  21
     5.2.  Access  20
     4.8.  Refresh Token Injection (TBD)  . . . . . . . . . . . . . .  22
     5.3.  XSRF Leakage (TBD) . . . . . . . . . . . . . . .  21
     4.9.  Open Redirection (TBD)  . . . . . . . .  23
   6.  Other Attacks . . . . . . . . . . .  21
     4.10. TLS Terminating Reverse Proxies . . . . . . . . . . . . .  23
   7.  22
     4.11. Other Topics  . . . . . . . . . . . . . . . . . . . . . . . .  23
   8.  22
   5.  Acknowledgements  . . . . . . . . . . . . . . . . . . . . . .  24
   9.  23
   6.  IANA Considerations . . . . . . . . . . . . . . . . . . . . .  24
   10.  23
   7.  Security Considerations . . . . . . . . . . . . . . . . . . .  24
   11.  23
   8.  References  . . . . . . . . . . . . . . . . . . . . . . . . .  24
     11.1.  23
     8.1.  Normative References  . . . . . . . . . . . . . . . . . .  24
     11.2.  23
     8.2.  Informative References  . . . . . . . . . . . . . . . . .  25  24
   Appendix A.  Document History . . . . . . . . . . . . . . . . . .  26  25
   Authors' Addresses  . . . . . . . . . . . . . . . . . . . . . . .  27  26

1.  Introduction

   It's been a while since OAuth has been published in RFC 6749
   [RFC6749] and RFC 6750 [RFC6750].  Since publication, OAuth 2.0 has
   gotten massive traction in the market and became the standard for API
   protection and, as foundation of OpenID Connect, identity providing.
   While OAuth was used in a variety of scenarios and different kinds of
   deployments, the following challenges could be observed:

   o  OAuth implementations are being attacked through known
      implementation weaknesses and anti-patterns (XSRF, referrer
      header).  Although most of these threats are discussed in RFC 6819
      [RFC6819], continued exploitation demonstrates there may be a need
      for more specific recommendations or that the existing mitigations
      are too difficult to deploy.

   o  Technology has changed, e.g. the way browsers treat fragments in
      some situations, which may change the implicit grant's underlying
      security model.

   o  OAuth is used in much more dynamic setups than originally
      anticipated, creating new challenges with respect to security.
      Those challenges go beyond the original scope of RFC 6749
      [RFC6749], RFC 6750 [RFC6749], and RFC 6819 [RFC6819].

   OAuth initially assumed a static relationship between client,
   authorization server and resource servers.  The URLs of AS and RS
   were known to the client at deployment time and built an anchor for
   the trust relationsship among those parties.  The validation whether
   the client talks to a legitimate server was based on TLS server
   authentication (see [RFC6819], Section 4.5.4).  With the increasing
   adoption of OAuth, this simple model dissolved and, in several
   scenarios, was replaced by a dynamic establishment of the
   relationship between clients on one side and the authorization and
   resource servers of a particular deployment on the other side.  This
   way the same client could be used to access services of different
   providers (in case of standard APIs, such as e-Mail or OpenID
   Connect) or serves as a frontend to a particular tenant in a multi-
   tenancy.  Extensions of OAuth, such as [RFC7591] and
   [I-D.ietf-oauth-discovery] were developed in order to support the
   usage of OAuth in dynamic scenarios.  As a challenge to the
   community, such usage scenarios open up new attack angles, which are
   discussed in this document.

   The remainder of the document is organized as follows: The next
   section gives a summary of the set of security mechanisms and
   practices, the working group shall consider to recommend to OAuth
   implementers.  This is followed by a section proposing modifications
   to OAuth intended to either simplify its usage and to strengthen its
   security.

   The remainder of the draft gives a detailed analyses of the
   weaknesses and implementation issues, which can be found in the wild
   today
   today, along with a discussion of potential counter measures.  First,
   various scenarios how OAuth credentials (namely access tokens and
   authorization codes) may be disclosed to attackers and proposes
   countermeasures are discussed.  Afterwards, the document discusses
   attacks possible with captured credential and how they may be
   prevented.  The last sections discuss additional threats.

2.  Recommended  Best Practice Practices

   This section describes the set of security mechanisms the authors
   believe should be taken into consideration by the OAuth working group
   to be recommended to OAuth implementers.

2.1.  Protecting redirect-based flows

   Authorization servers shall utilize exact matching of client redirect
   URIs against pre-registered URIs.  This measure contributes to the
   prevention of leakage of authorization codes and access tokens
   (depending on the grant type).  It also helps to detect mix up
   attacks.

   Clients shall avoid any redirects or forwards, which can be
   parameterized by URI query parameters, in order to provide a further
   layer of defence against token leakage.  If there is a need for this
   kind of redirects, clients are advised to implement appropriate
   counter measures against open redirection, e.g. as described by the
   OWASP [owasp].

   Clients shall ensure to only process redirect responses of the OAuth
   authorization server they send the respective request to and in the
   same user agent this request was initiated in.  In particular,
   clients shall implement appropriate XSRF prevention by utilizing one-
   time use XSRF tokens carried in the STATE parameter, which are
   securely bound to the user agent.  Moreover, the client shall store
   the authorization server's identity it sends an authorization request
   to in a transaction-specific manner, which is also bound to the
   particular user agent.  Furthermore, clients should use AS-specific
   redirect URIs as a means to identify the AS a particular response
   came from.  Matching this with the before mentioned information
   regarding the AS the client sent the request to helps to detect mix-
   up attacks.

   Note: [I-D.bradley-oauth-jwt-encoded-state] gives advice on how to
   implement XSRF prevention and AS matching using signed JWTs in the
   STATE parameter.

   Clients shall use PKCE [RFC7636] in order to (with the help of the
   authorization server) detect attempts to inject authorization codes
   into the authorization response.  The PKCE challenges must be
   transaction-specific and securely bound to the user agent, in which
   the transaction was started.

   Note: although PKCE so far was recommended as mechanism to protect
   native apps, this advice applies to all kinds of OAuth clients,
   including web applications.

2.2.  TBD

   Add further topics:

   o  Access  Token Leakage at resource Prevention

   Authorization servers

3.  Recommended modifications and extensions shall use TLS-based methods for sender
   constraint access tokens as described in section Section 4.7.1.2,
   such as token binding [I-D.ietf-oauth-token-binding] or Mutual TLS
   for OAuth 2.0 [I-D.ietf-oauth-mtls].  It is also recommend to use
   end-to-end TLS whenever possible.

3.  Recommended Changes to OAuth

   This section describes the set of modifications and extensions the
   authors believe should be taken into consideration by the OAuth
   working group change and extend OAuth in order to strengthen its
   security and make it simpler to implement.  It also recommends some
   changes to the OAuth set of specs.

   Remove requirement to check actual redirect URI at token endpoint -
   seems to be complicated to implement properly and could be
   compromised.  The protection goal is achieved even more effective by
   utilizing PKCE as recommended in Section 2.1.

4.  OAuth Credentials Leakage

   This section describes a couple of different ways how OAuth
   credentials, namely authorization codes  Attacks and access tokens, can be
   exposed to attackers. Mitigations

4.1.  Insufficient redirect URI validation

   Some authorization servers allow clients to register redirect URI
   patterns instead of complete redirect URIs.  In those cases, the
   authorization server, at runtime, matches the actual redirect URI
   parameter value at the authorization endpoint against this pattern.
   This approach allows clients to encode transaction state into
   additional redirect URI parameters or to register just a single
   pattern for multiple redirect URIs.  As a downside, it turned out to
   be more complex to implement and error prone to manage than exact
   redirect URI matching.  Several successful attacks have been observed
   in the wild, which utilized flaws in the pattern matching
   implementation or concrete configurations.  Such a flaw effectively
   breaks client identification or authentication (depending on grant
   and client type) and allows the attacker to obtain an authorization
   code or access token, either:

   o  by directly sending the user agent to a URI under the attackers
      control or

   o  by exposing the OAuth credentials to an attacker by utilizing an
      open redirector at the client in conjunction with the way user
      agents handle URL fragments.

4.1.1.  Attacks on Authorization Code Grant

   For a public client using the grant type code, an attack would look
   as follows:

   Let's assume the redirect URL pattern "https://*.example.com/*" had
   been registered for the client "s6BhdRkqt3".  This pattern allows
   redirect URIs from any host residing in the domain example.com.  So
   if an attacker manager to establish a host or subdomain in
   "example.com" he can impersonate the legitimate client.  Assume the
   attacker sets up the host "evil.example.com".

   (1)  The attacker needs to trick the user into opening a tampered URL
        in his browser, which launches a page under the attacker's
        control, say "https://www.evil.com".

   (2)  This URL initiates an authorization request with the client id
        of a legitimate client to the authorization endpoint.  This is
        the example authorization request (line breaks are for display
        purposes only):

   GET /authorize?response_type=code&client_id=s6BhdRkqt3&state=xyz
     &redirect_uri=https%3A%2F%2Fevil.example.com%2Fcb HTTP/1.1
   Host: server.example.com

   (1)  The authorization validates the redirect URI in order to
        identify the client.  Since the pattern allows arbitrary domains
        host names in "example.com", the authorization request is
        processed under the legitimate client's identity.  This includes
        the way the request for user consent is presented to the user.
        If auto-approval is allowed (which is not recommended for public
        clients according to RFC 6749), the attack can be performed even
        easier.

   (2)  If the user does not recognize the attack, the code is issued
        and directly sent to the attacker's client.

   (3)  Since the attacker impersonated a public client, it can directly
        exchange the code for tokens at the respective token endpoint.

   Note: This attack will not directly work for confidential clients,
   since the code exchange requires authentication with the legitimate
   client's secret.  The attacker will need to utilize the legitimate
   client to redeem the code (e.g. by mounting a code injection attack).
   This and other kinds kind of injections are is covered in Section OAuth Credentials Code Injection.

4.1.2.  Attacks on Implicit Grant

   The attack described above works for the implicit grant as well.  If
   the attacker is able to send the authorization response to a URI
   under his control, he will directly get access to the fragment
   carrying the access token.

   Additionally, implicit clients can be subject to a further kind of
   attacks.  It utilizes the fact that user agents re-attach fragments
   to the destination URL of a redirect if the location header does not
   contain a fragment (see [RFC7231], section 9.5).  The attack
   described here combines this behavior with the client as an open
   redirector in order to get access to access tokens.  This allows
   circumvention even of strict redirect URI patterns (but not strict
   URL matching!).

   Assume the pattern for client "s6BhdRkqt3" is
   "https://client.example.com/cb?*", i.e. any parameter is allowed for
   redirects to "https://client.example.com/cb".  Unfortunately, the
   client exposes an open redirector.  This endpoint supports a
   parameter "redirect_to", which takes a target URL and will send the
   browser to this URL using a HTTP 302.

   (1)  Same as above, the attacker needs to trick the user into opening
        a tampered URL in his browser, which launches a page under the
        attacker's control, say "https://www.evil.com".

   (2)  The URL initiates an authorization request, which is very
        similar to the attack on the code flow.  As differences, it
        utilizes the open redirector by encoding
        "redirect_to=https://client.evil.com" into the redirect URI and
        it uses the response type "token" (line breaks are for display
        purposes only):

   GET /authorize?response_type=token&client_id=s6BhdRkqt3&state=xyz
     &redirect_uri=https%3A%2F%2Fclient.example.com%2Fcb%26redirect_to
     %253Dhttps%253A%252F%252Fclient.evil.com%252Fcb HTTP/1.1
   Host: server.example.com

   (1)  Since the redirect URI matches the registered pattern, the
        authorization server allows the request and sends the resulting
        access token with a 302 redirect (some response parameters are
        omitted for better readability)

   HTTP/1.1 302 Found
     Location: https://client.example.com/cb?
     redirect_to%3Dhttps%3A%2F%2Fclient.evil.com%2Fcb
     #access_token=2YotnFZFEjr1zCsicMWpAA&...

   (2)  At the example.com, the request arrives at the open redirector.
        It will read the redirect parameter and will issue a HTTP 302 to
        the URL "https://evil.example.com/cb".

   HTTP/1.1 302 Found
        Location: https://client.evil.com/cb

   (3)  Since the redirector at example.com does not include a fragment
        in the Location header, the user agent will re-attach the
        original fragment
        "#access_token=2YotnFZFEjr1zCsicMWpAA&..." to the URL and will
        navigate to the following URL:

   https://client.evil.com/cb#access_token=2YotnFZFEjr1zCsicMWpAA&...

   (4)  The attacker's page at client.evil.com can access the fragment
        and obtain the access token.

4.1.3.  Proposed Countermeasures

   The complexity of implementing and managing pattern matching
   correctly obviously causes security issues.  This document therefore
   proposes to simplify the required logic and configuration by using
   exact redirect URI matching only.  This means the authorization
   server shall compare the two URIs using simple string comparison as
   defined in [RFC3986], Section 6.2.1..

   This would cause the following impacts:

   o  This change will require all OAuth clients to maintain the
      transaction state (and XSRF tokens) in the "state" parameter.
      This is a normative change to RFC 6749 since section 3.1.2.2
      allows for dynamic URI query parameters in the redirect URI.  In
      order to assess the practical impact, the working group needs to
      collect data on whether this feature is really used in deployments
      today.

   o  The working group may also consider this change as a step towards
      improved interoperability for OAuth implementations since RFC 6749
      is somewhat vague on redirect URI validation.  Notably there are
      no rules for pattern matching.  One may therefore assume all
      clients utilizing pattern matching will do so in a deployment
      specific way.  On the other hand, RFC 6749 already recommends
      exact matching if the full URL had been registered.

   o  Clients with multiple redirect URIs need to register all of them
      explicitly.
      Note: clients with just a single redirect URI would not even need
      to send a redirect URI with the authorization request.  Does it
      make sense to emphasize this option?  Would that further simplify
      use of the protocol and foster security?

   o  Exact redirect matching does not work for native apps utilizing a
      local web server due to dynamic port numbers - at least wild cards
      for port numbers are required.
      Question: Does redirect uri validation solve any problem for
      native apps?  Effective against impersonation when used in
      conjunction with claimed HTTPS redirect URIs only.
      For Windows token broker exact redirect URI matching is important
      as the redirect URI encodes the app identity.  For custom scheme
      redirects there is a question however it is probably a useful part
      of defense in depth.

   Additional recommendations:

   o  Servers on which callbacks are hosted must not expose open
      redirectors (see respective section).

   o  Clients may drop fragments via intermediary URLs with "fix
      fragments" (e.g. https://developers.facebook.com/blog/post/552/)
      to prevent the user agent from appending any unintended fragments.

   Alternatives to exact redirect URI matching:

   o  authenticate client using digital signatures (JAR?
      https://tools.ietf.org/html/draft-ietf-oauth-jwsreq-09)

4.2.  Authorization code leakage via referrer headers

   It is possible authorization codes are unintentionally disclosed to
   attackers, if a OAuth client renders a page containing links to other
   pages (ads, faq, ...) as result of a successful authorization
   request.

   If the user clicks onto one of those links and the target is under
   the control of an attacker, it can get access to the response URL in
   the referrer header.

   It is also possible that an attacker injects cross-domain content
   somehow into the page, such as <img> (f.e. if this is blog web site
   etc.): the implication is obviously the same - loading this content
   by browser results in leaking referrer with a code.

4.2.1.  Proposed Countermeasures

   There are some means to prevent leakage as described above:

   o  Use of the HTML link attribute rel="noreferrer" (Chrome
      52.0.2743.116, FF 49.0.1, Edge 38.14393.0.0, IE/Win10)

   o  Use of the "referrer" meta link attribute (possible values e.g.
      noreferrer, origin, ...) (cf. https://w3c.github.io/webappsec-
      referrer-policy/ - work in progress (seems Google, Chrome and Edge
      support it))

   o  Redirect to intermediate page (sanitize history) before sending
      user agent to other pages
      Note: double check redirect/referrer header behavior

   o  Use form post mode instead of redirect for authorization response
      (don't transport credentials via URL parameters and GET)

   Note: There shouldn't be a referer header when loading HTTP content
   from a HTTPS -loaded page (e.g. help/faq pages)

   Note: This kind of attack is not applicable to the implicit grant
   since fragments are not be included in referrer headers (cf.
   https://tools.ietf.org/html/rfc7231#section-5.5.2)

4.3.  Attacks in the Browser

4.3.1.  Code in browser history (TBD)

   When browser navigates to "client.com/redirection_endpoint?code=abcd"
   as a result of a redirect from a provider's authorization endpoint.

   Proposed countermeasures: code is one time use, has limited duration,
   is bound to client id/secret (confidential clients only)

4.3.2.  Access token in browser history (TBD)

   When a client or just a web site which already has a token
   deliberately navigates to a page like provider.com/
   get_user_profile?access_token=abcdef.. Actually RFC6750 discourages
   this practice and asks to transfer tokens via a header, but in
   practice web sites often just pass access token in query
   When browser navigates to client.com/
   redirection_endpoint#access_token=abcef as a result of a redirect
   from a provider's authorization endpoint.

   Proposal: replace implicit flow with postmessage communication

4.3.3.  Javascript Code stealing Access Tokens (TBD)

   sandboxing using service workers

4.4.  Access Token Leakage  Mix-Up

   Mix-up is another kind of attack on more dynamic OAuth scenarios (or
   at the Resource Server

4.4.1.  Access Token Phishing by Counterfeit Resource Server

   An attacker may setup his own resource server and trick least scenarios where a OAuth client into
   sending access tokens to it, which are valid for other resource
   servers.  If interacts with multiple
   authorization servers).  The goal of the client sends a valid attack is to obtain an
   authorization code or an access token by tricking the client into
   sending those credentials to this
   counterfeit resource server, the attacker in turn may use that token
   to access other services on behalf (which acts as MITM between
   client and authorization server)

   A detailed description of the resource owner.

   This attack assumes the client and potential countermeasures is not bound
   given in cf. https://tools.ietf.org/html/draft-ietf-oauth-mix-up-
   mitigation-01.

   Potential mitigations:

   o  AS returns client_id and its iss in the response.  Client compares
      this data to a certain resource
   server (and AS it believed it sent the respective URL) at development time, but user agent to.

   o  ID token carries client
   instances are configured id and issuer (requires OpenID Connect)

   o  Clients use AS-specific redirect URIs, for every authorization
      request store intended AS and compare intention with an resource server's URL at runtime.
   This kind of late binding is typical in situations, actual
      redirect URI where the client
   uses a standard API, e.g. for e-Mail, calendar, health, or banking
   and is configured by response was received (no change to OAuth
      required)

4.5.  Code Injection

   In such an user or administrator for the standard-based
   service, this particular user or company uses.

   There are several potential mitigation strategies, which will be
   discussed in attack, the following sections.

4.4.1.1.  Metadata

   An adversary attempts to inject a stolen
   authorization server could provide the code into a legitimate client with additional
   information about the location where it is safe to use its access
   tokens. on a device under his
   control.  In the simplest form, this would require case, the AS attacker would want to publish a list of
   its known resource servers, illustrated in use the following example
   using a metadata parameter "resource_servers":

   HTTP/1.1 200 OK
   Content-Type: application/json

   {
     "issuer":"https://server.example.com",
     "authorization_endpoint":"https://server.example.com/authorize",
     "resource_servers":[
       "email.example.com",
       "storage.example.com",
       "video.example.com"]
     ...
   }
   code in his own client.  But there are situations where this might
   not be possible or intended.  Example are:

   o  The AS could also return the URL(s) an access token code is good for in
   the token response, illustrated by bound to a particular confidential client and the example return parameter
   "access_token_resource_server":

HTTP/1.1 200 OK
Content-Type: application/json;charset=UTF-8
Cache-Control: no-store
Pragma: no-cache

{
  "access_token":"2YotnFZFEjr1zCsicMWpAA",
  "access_token_resource_server":"https://hostedresource.example.com/path1",
...
}

   This mitigation strategy would rely on
      attacker is unable to obtain the required client credentials to enforce
      redeem the
   security policy and code himself and/or

   o  The attacker wants to only send access tokens certain functions in this particular
      client.  As an example, the attacker potentially wants to legitimate
   destinations.  Results of OAuth related security research (see for
   example [oauth_security_ubc] and [oauth_security_cmu]) indicate
      impersonate his victim in a
   large portion of client implementations do not or fail to properly
   implement security controls, like state checks.  So relying on
   clients to detect and properly handle certain app.

   o  Another example could be that access token phishing is likely to fail as well.  Moreover given the ratio of clients to authorization and
      resource servers, it servers is considered the more viable
   approach to move as much as possible security-related logic some how limited to those
   entities.  Clearly, networks, the client has to contribute attackers is
      unable to access directly.

   How does an attack look like?

   (1)  The attacker obtains an authorization code by executing any of
        the overall
   security.  But there are alternative counter measures, as attacks described
   in the next sections, which provide a better balance between above.

   (2)  It performs an OAuth authorization process with the
   involved parties.

4.4.1.2.  Sender Constrained Access Tokens

   As legitimate
        client on his device.

   (3)  The attacker injects the name suggests, sender constraint access token scope stolen authorization code in the
   applicability
        response of an access token the authorization server to a certain sender.  This sender is
   obliged the legitimate client.

   (4)  The client sends the code to demonstrate knowledge of a certain secret as prerequisite
   for the acceptance of that authorization server's token at a resource server.

   A typical flow looks like this:

   1.
        endpoint, along with client id, client secret and actual
        redirect_uri.

   (5)  The authorization server associates data with checks the access token,
       which bind this particular token client secret, whether the
        code was issued to a certain client.  The
       binding can utilize the particular client identity, but in most cases and whether the AS
       utilizes key material (or data derived from actual
        redirect URI matches the key material)
       known redirect_uri parameter.

   (6)  If all checks succeed, the authorization server issues access
        and other tokens to the client.

   2.  This key material must be distributed somehow.  Either

   (7)  The attacker just impersonated the key
       material already exists before victim.

   Obviously, the AS creates check in step (5) will fail, if the binding or code was issued to
   another client id, e.g. a client set up by the
       AS creates ephemeral keys.  The way pre-existing key material is
       distributed varies among attacker.

   An attempt to inject a code obtained via a malware pretending to be
   the different approaches.  For example,
       X.509 Certificates can legitimate client should also be used in which case detected, if the distribution
       happens explicitly during authorization
   server stored the enrollment process.  Or complete redirect URI used in the key
       material is created authorization
   request and distributed at the TLS layer, in which
       case compares it might automatically happens during with the setup of a TLS
       connection.

   3.  The RS must implement redirect_uri parameter.

   [RFC6749], Section 4.1.3, requires the actual proof of possession check.  This AS to ...  "ensure that the
   "redirect_uri" parameter is typically done on present if the application level, it may utilize
       capabilities of "redirect_uri" parameter
   was included in the transport layer (e.g.  TLS).  Note: replay
       detection is required initial authorization request as well!

   There exists several proposals to demonstrate the proof of possession described in
   Section 4.1.1, and if included ensure that their values are
   identical."  In the scope of attack scenario described above, the OAuth working group:

   o  [I-D.ietf-oauth-token-binding]: In legitimate
   client would use the correct redirect URI it always uses for
   authorization requests.  But this approach, an access tokens
      is, via URI would not match the so-called token binding id, bound to key material
      representing a long term association between a client and a
      certain TLS host.  Negotiation of tampered
   redirect URI used by the key material attacker (otherwise, the redirect would not
   land at the attackers page).  So the authorization server would
   detect the attack and proof of
      possession in refuse to exchange the context of code.

   Note: this check could also detect attempt to inject a TLS handshake is taken care code, which
   had been obtained from another instance of by the TLS stack.  The same client on another
   device, if certain conditions are fulfilled:

   o  the redirect URI itself needs to determine the token binding id contain a nonce or another kind
      of the target resource server one-time use, secret data and pass

   o  the client has bound this data to this particular instance

   But this approach conflicts with the idea to enforce exact redirect
   URI matching at the access
      token request.  The authorization server than associates endpoint.  Moreover, it has been
   observed that providers very often ignore the
      access token with redirect_uri check
   requirement at this id.  The resource stage, maybe, because it doesn't seem to be
   security-critical from reading the spec.

   Other providers just pattern match the redirect_uri parameter against
   the registered redirect URI pattern.  This saves the authorization
   server checks on every
      invocation that from storing the token binding id of link between the active TLS connection actual redirect URI and the token binding id
   respective authorization code for every transaction.  But this kind
   of associated with check obviously does not fulfill the access token
      match.  Since all crypto-related functions are covered by intent of the TLS
      stack, this approach spec, since the
   tampered redirect URI is very client developer friendly.  As not considered.  So any attempt to inject a
      prerequisite, token binding as described in
      [I-D.ietf-tokbind-https] (including federated token bindings) must
      be supported on all ends (client, authorization server, resource
      server).

   o  [I-D.ietf-oauth-mtls]: The approach as specified in this document
      allow use of mutual TLS for both client authentication and sender
      constraint access tokens.  For
   code obtained using the purpose client_id of sender constraint
      access tokens, the a legitimate client is identified towards the resource
      server or by
   utilizing the fingerprint of its public key.  During processing of
      an access token request, the authorization server obtains the
      client's public key from the TLS stack and associates its
      fingerprint with the respective access tokens.  The resource
      server legitimate client on another device won't be detected
   in the same way obtains respective deployments.

   It is also assumed that the public key from requirements defined in [RFC6749],
   Section 4.1.3, increase client implementation complexity as clients
   need to memorize or re-construct the TLS stack
      and compares its fingerprint with correct redirect URI for the fingerprint associated with
   call to the access token.

   o  [I-D.ietf-oauth-signed-http-request] specifies an approach tokens endpoint.

   The authors therefore propose to sign
      HTTP requests.  It utilizes [I-D.ietf-oauth-pop-key-distribution]
      and represents the elements working group to drop this
   feature in favor of the signature more effective and (hopefully) simpler approaches
   to code injection prevention as described in a JSON object. the following section.

4.5.1.  Proposed Countermeasures

   The signature general proposal is built using JWS.  The mechanism has built-in
      support for signing of HTTP method, query parameters and headers.
      It also incorporates a timestamp as basis for replay detection.

   o  [I-D.sakimura-oauth-jpop]: this draft describes different ways to
      constrain access token usage, namely TLS or request signing.
      Note: Since bind every particular authorization code
   to a certain client on a certain device (or in a certain user agent)
   in the authors context of this draft contributed the TLS-related
      proposal a certain transaction.  There are multiple
   technical solutions to [I-D.ietf-oauth-mtls], achieve this document only considers
      the request signing part.  For request signing, the draft utilizes
      [I-D.ietf-oauth-pop-key-distribution] and RFC 7800 [RFC7800].  The
      signature data is represented in a JWT and JWS goal:

   Nonce   OpenID Connect's existing "nonce" parameter is used for
      signing.  Replay detection this
           purpose.  The nonce value is provided one time use and created by building the signature
      over a server-provided nonce, client-provided nonce and a nonce
      counter.

   [I-D.ietf-oauth-mtls] and [I-D.ietf-oauth-token-binding] are built on
   top of TLS
           client.  The client is supposed to bind it to the user agent
           session and this way continue sends it with the successful OAuth 2.0 philosophy
   to leverage TLS initial request to secure OAuth wherever possible.  Both mechanisms
   allow prevention of access token leakage in a fairly client developer
   friendly way.

   There are some differences between both approaches: To start with, in
   [I-D.ietf-oauth-token-binding] all key material is automatically
   managed by the TLS stack whereas [I-D.ietf-oauth-mtls] requires OpenId
           Provider (OP).  The OP associates the
   developer nonce to create and maintain the key pairs
           authorization code and respective
   certificates.  Use of self-signed certificates, attests this binding in the ID token,
           which is supported by issued as part of the draft, significantly reduce code exchange at the complexity of this task.
   Furthermore, [I-D.ietf-oauth-token-binding] allows to use different
   key pairs for different resource servers, which is a privacy benefit.
   On token
           endpoint.  If an attacker injected an authorization code in
           the other hand, [I-D.ietf-oauth-mtls] only requires widely
   deployed TLS features, which means it might be easier to adopt authorization response, the nonce value in the
   short term.

   Application level signing approaches, like
   [I-D.ietf-oauth-signed-http-request] client
           session and [I-D.sakimura-oauth-jpop]
   have been debated for a long time the nonce value in the OAuth working group without
   a clear outcome.

   As one advantage, application-level signing allows for end-to-end
   protection including non-repudiation even if ID token will not match
           and the TLS connection attack is
   terminated between client and resource server.  But deployment
   experiences have revealed challenges regarding robustness (e.g.
   reproduction detected.  assumption: attacker cannot get
           hold of the signature base string including correct URL) as
   well as user agent state management (e.g. replay detection).

   This document therefore recommends implementors to consider one of
   TLS-based approaches wherever possible.

4.4.1.3.  Audience Restricted Access Tokens

   An audience restriction essentially restricts on the resource server a
   particular access token can be used at.  The victims device, where he
           has stolen the respective authorization server
   associates code.
           pro:
           - existing feature, used in the access token with a certain resource server and every
   resource server is obliged wild
           con:
           - OAuth does not have an ID Token - would need to verify for every request, whether the
   access token send with push that request was meant to be used at
           down the
   particular resource server.  If not, stack

   Code-bound State  It has been discussed in the resource server must refuse security workshop in
           December to serve the respective request.  In use the general case, audience
   restrictions limit OAuth state value much similar in the impact of a token leakage. way
           as described above.  In the case of a
   counterfeit resource server, it may (as described see below) also
   prevent abuse of the phished access token at state value, the legitimate resource
   server.

   The audience can basically be expressed using logical names or
   physical addresses (like URLs).  In order to detect phishing, it idea
           is
   necessary to use the actual URL the client will send requests to.  In
   the phishing case, this URL will point to the counterfeit resource
   server.  If the attacker tries to use the access token at the
   legitimate resource server (which has add a different URL), the resource
   server will detect the mismatch (wrong audience) and refuse further parameter state to serve the code exchange
           request.

   In deployments where the authorization server knows the URLs of all
   resource servers, the authorization server may just refuse to issue
   access tokens for unknown resource server URLs.  The client needs to tell the authorization server, at which URL it
   will use server then compares the access token state
           value it is requesting.  It could use associated with the
   mechanism proposed [I-D.campbell-oauth-resource-indicators] or encode code and the information state value in the scope value.

   Instead of the URL,
           parameter.  If those values do not match, it is also possible to utilize the fingerprint of considered an
           attack and the resource server's X.509 certificate as audience value.  This request fails.  Note: a variant of this
           solution would also allow to detect an attempt to spoof the legit
   resource server's URL by using be send a valid TLS certificate obtained from
   a different CA.  It might also be considered a privacy benefit hash of the state (in order to
   hide
           prevent bulky requests and DoS).
           pro:
           - use existing concept
           con:
           - state needs to fulfil certain requirements (one time use,
           complexity)
           - new parameter means normative spec change

   PKCE    Basically, the resource server URL from PKCE challenge/verifier could be used in the authorization server.

   Audience restriction seems easy
           same way as Nonce or State.  In contrast to use since it does not require any
   crypto on its original
           intention, the verifier check would fail although the client side.  But since every access token
           uses its correct verifier but the code is bound to associated with a certain resource server, the client also needs
           challenge, which does not match.
           pro:
           - existing and deployed OAuth feature
           con:
           - currently used and recommended for native apps, not web
           apps

   Token Binding  Code must be bind to obtain different
   RS-specific access tokens, if it wants UA-AS and UA-Client legs -
           requires further data (extension to access several resource
   services.  [I-D.ietf-oauth-token-binding] has the same property,
   since different response) to manifest
           binding id for particular code.

           Note: token binding ids must could be associated used in conjunction with the access
   token.  [I-D.ietf-oauth-mtls] PKCE as
           an option (https://tools.ietf.org/html/draft-ietf-oauth-
           token-binding-02#section-4).
           pro:
           - highly secure
           con:
           - highly sophisticated, requires browser support, will it
           work for native apps?

   per instance client id/secret  ...

   Note on pre-warmed secrets: An attacker can circumvent the other hand allows a client
   countermeasures described above if he is able to
   use create or capture
   the access token at multiple resource servers.

   It shall be noted that audience restrictions, respective secret or generally speaking
   an indication by the client to code_challenge on a device under his
   control, which is then used in the victim's authorization server where it
   wants to use request.
   Exact redirect URI matching of authorization requests can prevent the access token, has additional benefits beyond
   attacker from using the
   scope of token leakage prevention.  It allows pre-warmed secret in the faked authorization
   server to create different access token whose format and content
   transaction on the victim's device.
   Unfortunately it does not work for all kinds of OAuth clients.  It is
   specifically minted
   effective for the respective server.  This has huge
   functional web and privacy advantages in deployments using structured JS apps and for native apps with claimed URLs.
   What about other native apps?  Treat nonce or PKCE challenge as
   replay detection tokens (needs to ensure cluster-wide one-time use)?

4.6.  XSRF (TBD)

   injection of code or access tokens.

4.4.2.  Compromised token on a victim's device (e.g. to cause
   client to access resources under the attacker's control)

   mitigation: XSRF tokens (one time use) w/ user agent binding (cf.
   https://www.owasp.org/index.php/
   CrossSite_Request_Forgery_(CSRF)_Prevention_Cheat_Sheet)

4.7.  Access Token Leakage at the Resource Server

4.7.1.  Access Token Phishing by Counterfeit Resource Server

   An attacker may compromise a setup his own resource server in order to get and trick a client into
   sending access tokens to its resources and it, which are valid for other resources of resource
   servers.  If the respective deployment.
   Such client sends a compromise may range from partial valid access token to this
   counterfeit resource server, the system, e.g.
   its logfiles, attacker in turn may use that token
   to full control access other services on behalf of the respective server.

   If the attacker was able to take over full control including shell
   access it will be able to circumvent all controls in place and access
   resources without access control.  It will also get access to access
   tokens, which are sent to the compromised system and which
   potentially are valid for access to other resource servers as well.
   Even if owner.

   This attack assumes the attacker "only" client is able to access logfiles or databases
   of the server system, it may get access not bound to valid access tokens.

   Preventing and detecting a certain resource
   server breaches by way (and the respective URL) at development time, but client
   instances are configured with an resource server's URL at runtime.
   This kind of hardening and
   monitoring server systems late binding is considered typical in situations, where the client
   uses a standard operational
   procedure API, e.g. for e-Mail, calendar, health, or banking
   and therefore out of scope of is configured by an user or administrator for the standard-based
   service, this document.  This section particular user or company uses.

   There are several potential mitigation strategies, which will focus on be
   discussed in the impact of such breaches on OAuth-related parts of following sections.

4.7.1.1.  Metadata

   An authorization server could provide the ecosystem, which is client with additional
   information about the replay of captured location where it is safe to use its access tokens on
   tokens.

   In the
   compromised resource server and other resource servers of simplest form, this would require the
   respective deployment.

   The following measures shall be taken into account by implementors in
   order AS to cope with access token replay:

   o  The publish a list of
   its known resource server must treat access tokens like any other
      credentials.  It is considered good practice to not log them and
      not to store them in plain text.

   o  Sender constraint access tokens as described servers, illustrated in Section 4.4.1.2
      will prevent the attacker from replaying following example
   using a metadata parameter "resource_servers":

   HTTP/1.1 200 OK
   Content-Type: application/json

   {
     "issuer":"https://server.example.com",
     "authorization_endpoint":"https://server.example.com/authorize",
     "resource_servers":[
       "email.example.com",
       "storage.example.com",
       "video.example.com"]
     ...
   }

   The AS could also return the URL(s) an access tokens on
      other resource servers.  Depending on token is good for in
   the severity of token response, illustrated by the
      penetration, it will also prevent replay example return parameter
   "access_token_resource_server":

HTTP/1.1 200 OK
Content-Type: application/json;charset=UTF-8
Cache-Control: no-store
Pragma: no-cache

{
  "access_token":"2YotnFZFEjr1zCsicMWpAA",
  "access_token_resource_server":"https://hostedresource.example.com/path1",
...
}

   This mitigation strategy would rely on the compromised
      system.

   o  Audience restriction as described in Section 4.4.1.3 may be used client to prevent replay of captured enforce the
   security policy and to only send access tokens on other resource
      servers.

4.4.3.  TLS Terminating Reverse Proxies

   A common deployment architecture for HTTP applications is to have the
   application server sitting behind legitimate
   destinations.  Results of OAuth related security research (see for
   example [oauth_security_ubc] and [oauth_security_cmu]) indicate a reverse proxy, which terminates
   the TLS connection
   large portion of client implementations do not or fail to properly
   implement security controls, like state checks.  So relying on
   clients to detect and dispatches the incoming requests properly handle access token phishing is likely
   to fail as well.  Moreover given the
   respective application server nodes.

   This section highlights some attack angles ratio of this deployment
   architecture, which are relevant clients to OAuth,
   authorization and give recommendations
   for security controls.

   In some situations, resource servers, it is considered the reverse proxy needs more viable
   approach to pass move as much as possible security-related
   data logic to those
   entities.  Clearly, the upstream application servers for further processing.
   Examples include the IP address of the request originator, token
   binding ids and authenticated TLS client certificates.

   If has to contribute to the reverse proxy would pass through any header sent from overall
   security.  But there are alternative counter measures, as described
   in the
   outside, an attacker could try to directly send next sections, which provide a better balance between the faked header
   values through
   involved parties.

4.7.1.2.  Sender Constrained Access Tokens

   As the proxy to name suggests, sender constraint access token scope the application server in order
   applicability of an access token to
   circumvent security controls that way.  For example, it a certain sender.  This sender is standard
   practice of reverse proxies
   obliged to accept "forwarded_for" headers and
   just add demonstrate knowledge of a certain secret as prerequisite
   for the origin acceptance of that token at a resource server.

   A typical flow looks like this:

   1.  The authorization server associates data with the inbound request (making it access token,
       which bind this particular token to a list).
   Depending on certain client.  The
       binding can utilize the logic performed client identity, but in most cases the application server, AS
       utilizes key material (or data derived from the
   attacker could simply add a whitelisted IP address key material)
       known to the header and
   render a IP whitelist useless.  A reverse proxy client.

   2.  This key material must therefore
   sanitize any inbound requests to ensure be distributed somehow.  Either the authenticity and
   integrity of all header values relevant for key
       material already exists before the security of AS creates the
   application servers.

   If an attacker would binding or the
       AS creates ephemeral keys.  The way pre-existing key material is
       distributed varies among the different approaches.  For example,
       X.509 Certificates can be able to get access to used in which case the internal network
   between proxy distribution
       happens explicitly during the enrollment process.  Or the key
       material is created and application server, it could also try to circumvent
   security controls distributed at the TLS layer, in place.  It is therefore important to ensure which
       case it might automatically happens during the
   authenticity setup of a TLS
       connection.

   3.  The RS must implement the communicating entities.  Furthermore, actual proof of possession check.  This
       is typically done on the
   communication link between reverse proxy and application server must
   therefore be protected against tapping and injection (including level, it may utilize
       capabilities of the transport layer (e.g.  TLS).  Note: replay prevention).

4.5.  Mix-Up

   Mix-up
       detection is another kind required as well!

   There exists several proposals to demonstrate the proof of attack on more dynamic OAuth scenarios (or
   at least scenarios where a OAuth client interacts with multiple
   authorization servers).  The goal possession
   in the scope of the attack is to obtain an
   authorization code or OAuth working group:

   o  [I-D.ietf-oauth-token-binding]: In this approach, an access token by tricking tokens
      is, via the client into
   sending those credentials so-called token binding id, bound to the attacker (which acts as MITM key material
      representing a long term association between a client and authorization server)

   A detailed description a
      certain TLS host.  Negotiation of the attack and potential countermeasures is
   given in cf. https://tools.ietf.org/html/draft-ietf-oauth-mix-up-
   mitigation-01.

   Potential mitigations:

   o  AS returns client_id key material and its iss proof of
      possession in the response.  Client compares
      this data context of a TLS handshake is taken care of by
      the TLS stack.  The client needs to AS it believed it sent determine the user agent to.

   o  ID token carries client binding id
      of the target resource server and issuer (requires OpenID Connect)

   o  Clients use AS-specific redirect URIs, for every pass this data to the access
      token request.  The authorization
      request store intended AS and compare intention with actual
      redirect URI where server than associates the response was received (no change to OAuth
      required)

4.6.  Refresh Token Leakage

   mitm, log files
      access token with this id.  The resource server checks on every
      invocation that the device, ...

   refresh token rotation, mutual binding id of the active TLS authentication at connection
      and the token
   endpoint

5.  OAuth Credentials Injection

   Credential injection means an attacker somehow obtained a valid OAuth
   credential (code or token) and is able to utilize this to impersonate binding id of associated with the legitimate resource owner or to cause a victim to access
   resources under the attacker's control (XSRF).

5.1.  Code Injection

   In such an attack, token
      match.  Since all crypto-related functions are covered by the adversary attempts to inject a stolen
   authorization code into a legitimate TLS
      stack, this approach is very client on developer friendly.  As a device under his
   control.  In the simplest case, the attacker would want to use the
   code
      prerequisite, token binding as described in his own client.  But there are situations where this might
   not
      [I-D.ietf-tokbind-https] (including federated token bindings) must
      be possible or intended.  Example are: supported on all ends (client, authorization server, resource
      server).

   o  [I-D.ietf-oauth-mtls]: The code is bound to a particular confidential client and the
      attacker is unable to obtain the required client credentials to
      redeem the code himself and/or

   o  The attacker wants to access certain functions approach as specified in this particular
      client.  As an example, the attacker potentially wants to
      impersonate his victim in a certain app.

   o  Another example could be that access to the authorization document
      allow use of mutual TLS for both client authentication and
      resource servers is some how limited to networks, the attackers is
      unable to sender
      constraint access directly.

   How does an attack look like?

   (1)  The attacker obtains an authorization code by executing any of tokens.  For the attacks described above (OAuth Credentials Leakage).

   (2)  It performs an OAuth authorization process with purpose of sender constraint
      access tokens, the legitimate client on his device.

   (3)  The attacker injects is identified towards the stolen authorization code in resource
      server by the
        response fingerprint of its public key.  During processing of
      an access token request, the authorization server to the legitimate client.

   (4)  The client sends obtains the code to
      client's public key from the authorization server's token
        endpoint, along with client id, client secret TLS stack and actual
        redirect_uri.

   (5) associates its
      fingerprint with the respective access tokens.  The authorization resource
      server checks in the client secret, whether same way obtains the
        code was issued to public key from the particular client TLS stack
      and whether the actual
        redirect URI matches compares its fingerprint with the redirect_uri parameter.

   (6)  If all checks succeed, fingerprint associated with
      the authorization server issues access
        and other tokens token.

   o  [I-D.ietf-oauth-signed-http-request] specifies an approach to sign
      HTTP requests.  It utilizes [I-D.ietf-oauth-pop-key-distribution]
      and represents the client.

   (7)  The attacker just impersonated the victim.

   Obviously, elements of the check signature in step (5) will fail, if a JSON object.
      The signature is built using JWS.  The mechanism has built-in
      support for signing of HTTP method, query parameters and headers.
      It also incorporates a timestamp as basis for replay detection.

   o  [I-D.sakimura-oauth-jpop]: this draft describes different ways to
      constrain access token usage, namely TLS or request signing.
      Note: Since the code was issued authors of this draft contributed the TLS-related
      proposal to
   another client id, e.g. [I-D.ietf-oauth-mtls], this document only considers
      the request signing part.  For request signing, the draft utilizes
      [I-D.ietf-oauth-pop-key-distribution] and RFC 7800 [RFC7800].  The
      signature data is represented in a client set up JWT and JWS is used for
      signing.  Replay detection is provided by building the attacker.

   An attempt to inject signature
      over a code obtained via server-provided nonce, client-provided nonce and a malware pretending to be nonce
      counter.

   [I-D.ietf-oauth-mtls] and [I-D.ietf-oauth-token-binding] are built on
   top of TLS and this way continue the legitimate successful OAuth 2.0 philosophy
   to leverage TLS to secure OAuth wherever possible.  Both mechanisms
   allow prevention of access token leakage in a fairly client should also be detected, if the authorization
   server stored the complete redirect URI used developer
   friendly way.

   There are some differences between both approaches: To start with, in
   [I-D.ietf-oauth-token-binding] all key material is automatically
   managed by the authorization
   request and compares it with the redirect_uri parameter.

   [RFC6749], Section 4.1.3, TLS stack whereas [I-D.ietf-oauth-mtls] requires the AS
   developer to ...  "ensure that create and maintain the
   "redirect_uri" parameter key pairs and respective
   certificates.  Use of self-signed certificates, which is present if supported by
   the "redirect_uri" parameter
   was included in draft, significantly reduce the initial authorization request as described complexity of this task.
   Furthermore, [I-D.ietf-oauth-token-binding] allows to use different
   key pairs for different resource servers, which is a privacy benefit.
   On the other hand, [I-D.ietf-oauth-mtls] only requires widely
   deployed TLS features, which means it might be easier to adopt in
   Section 4.1.1, the
   short term.

   Application level signing approaches, like
   [I-D.ietf-oauth-signed-http-request] and if included ensure that their values are
   identical."  In [I-D.sakimura-oauth-jpop]
   have been debated for a long time in the attack scenario described above, OAuth working group without
   a clear outcome.

   As one advantage, application-level signing allows for end-to-end
   protection including non-repudiation even if the legitimate TLS connection is
   terminated between client would use and resource server.  But deployment
   experiences have revealed challenges regarding robustness (e.g.
   reproduction of the signature base string including correct redirect URI it always uses for
   authorization requests.  But this URI would not match URL) as
   well as state management (e.g. replay detection).

   This document therefore recommends implementors to consider one of
   TLS-based approaches wherever possible.

4.7.1.3.  Audience Restricted Access Tokens

   An audience restriction essentially restricts the tampered
   redirect URI resource server a
   particular access token can be used by at.  The authorization server
   associates the attacker (otherwise, access token with a certain resource server and every
   resource server is obliged to verify for every request, whether the redirect would not
   land
   access token send with that request was meant to be used at the attackers page).  So
   particular resource server.  If not, the authorization resource server would
   detect the attack and must refuse
   to exchange serve the code.

   Note: this check could also detect attempt to inject a code, which
   had been obtained from another instance of respective request.  In the same client on another
   device, if certain conditions are fulfilled:

   o general case, audience
   restrictions limit the redirect URI itself needs to contain a nonce or another kind impact of one-time use, secret data and

   o a token leakage.  In the client has bound this data to this particular instance

   But this approach conflicts with case of a
   counterfeit resource server, it may (as described see below) also
   prevent abuse of the idea to enforce exact redirect
   URI matching phished access token at the authorization endpoint.  Moreover, legitimate resource
   server.

   The audience can basically be expressed using logical names or
   physical addresses (like URLs).  In order to detect phishing, it has been
   observed that providers very often ignore is
   necessary to use the redirect_uri check
   requirement at actual URL the client will send requests to.  In
   the phishing case, this stage, maybe, because it doesn't seem URL will point to be
   security-critical from reading the spec.

   Other providers just pattern match counterfeit resource
   server.  If the redirect_uri parameter against attacker tries to use the registered redirect URI pattern.  This saves access token at the authorization
   legitimate resource server from storing (which has a different URL), the link between resource
   server will detect the actual redirect URI mismatch (wrong audience) and refuse to serve
   the request.

   In deployments where the
   respective authorization code for every transaction.  But this kind
   of check obviously does not fulfill server knows the intent URLs of all
   resource servers, the spec, since the
   tampered redirect URI is not considered.  So any attempt authorization server may just refuse to inject a
   code obtained using the client_id of a legitimate client or by
   utilizing the legitimate issue
   access tokens for unknown resource server URLs.

   The client on another device won't be detected
   in needs to tell the respective deployments.

   It authorization server, at which URL it
   will use the access token it is also assumed that requesting.  It could use the requirements defined in [RFC6749],
   Section 4.1.3, increase client implementation complexity as clients
   need to memorize
   mechanism proposed [I-D.campbell-oauth-resource-indicators] or re-construct encode
   the correct redirect URI for information in the
   call to scope value.

   Instead of the tokens endpoint.

   The authors therefore propose URL, it is also possible to utilize the working group to drop this
   feature in favor fingerprint of more effective and (hopefully) simpler approaches
   to code injection prevention as described in
   the following section.

5.1.1.  Proposed Countermeasures

   The general proposal is resource server's X.509 certificate as audience value.  This
   variant would also allow to bind every particular authorization code detect an attempt to spoof the legit
   resource server's URL by using a certain client on valid TLS certificate obtained from
   a certain device (or in different CA.  It might also be considered a certain user agent)
   in privacy benefit to
   hide the context of a certain transaction.  There are multiple
   technical solutions resource server URL from the authorization server.

   Audience restriction seems easy to achieve this goal:

   Nonce   OpenID Connect's existing "nonce" parameter is used for this
           purpose.  The nonce value is one time use and created by since it does not require any
   crypto on the
           client.  The client side.  But since every access token is supposed to bind it bound to
   a certain resource server, the user agent
           session and sends it with the initial request client also needs to the OpenId
           Provider (OP).  The OP associates the nonce obtain different
   RS-specific access tokens, if it wants to access several resource
   services.  [I-D.ietf-oauth-token-binding] has the
           authorization code and attests this same property,
   since different token binding in ids must be associated with the ID token,
           which is issued as part of access
   token.  [I-D.ietf-oauth-mtls] on the code exchange at other hand allows a client to
   use the access token
           endpoint.  If an attacker injected at multiple resource servers.

   It shall be noted that audience restrictions, or generally speaking
   an authorization code in indication by the authorization response, client to the nonce value in authorization server where it
   wants to use the client
           session and access token, has additional benefits beyond the nonce value in
   scope of token leakage prevention.  It allows the ID authorization
   server to create different access token will not match whose format and the attack content is detected.  assumption:
   specifically minted for the respective server.  This has huge
   functional and privacy advantages in deployments using structured
   access tokens.

4.7.2.  Compromised Resource Server

   An attacker cannot may compromise a resource server in order to get
           hold access
   to its resources and other resources of the user agent state on the victims device, where he
           has stolen the respective authorization code.
           pro:
           - existing feature, used in the wild
           con:
           - OAuth does not have an ID Token - would need deployment.
   Such a compromise may range from partial access to push that
           down the stack

   Code-bound State  It has been discussed in the security workshop in
           December system, e.g.
   its logfiles, to use full control of the OAuth state value much similar in respective server.

   If the way
           as described above.  In the case of attacker was able to take over full control including shell
   access it will be able to circumvent all controls in place and access
   resources without access control.  It will also get access to access
   tokens, which are sent to the state value, compromised system and which
   potentially are valid for access to other resource servers as well.
   Even if the idea attacker "only" is able to add a further parameter state to access logfiles or databases
   of the code exchange
           request.  The authorization server then compares the state
           value system, it associated with the code may get access to valid access tokens.

   Preventing and the state value in the
           parameter.  If those values do not match, it detecting server breaches by way of hardening and
   monitoring server systems is considered an
           attack and the request fails.  Note: a variant standard operational
   procedure and therefore out of scope of this
           solution would be send a hash document.  This section
   will focus on the impact of such breaches on OAuth-related parts of
   the state (in order to
           prevent bulky requests ecosystem, which is the replay of captured access tokens on the
   compromised resource server and DoS).
           pro:
           - use existing concept
           con:
           - state needs to fulfil certain requirements (one time use,
           complexity)
           - new parameter means normative spec change

   PKCE    Basically, other resource servers of the PKCE challenge/verifier could
   respective deployment.

   The following measures shall be used taken into account by implementors in the
           same way as Nonce or State.  In contrast
   order to its original
           intention, the verifier check would fail although the client
           uses its correct verifier but the code is associated cope with a
           challenge, which does not match.
           pro:
           - existing and deployed OAuth feature
           con:
           - currently used and recommended for native apps, not web
           apps

   Token Binding  Code access token replay:

   o  The resource server must be bind treat access tokens like any other
      credentials.  It is considered good practice to UA-AS not log them and UA-Client legs -
           requires further data (extension to response)
      not to manifest
           binding id for particular code.
           Note: token binding could be used store them in conjunction with PKCE plain text.

   o  Sender constraint access tokens as
           an option (https://tools.ietf.org/html/draft-ietf-oauth-
           token-binding-02#section-4).
           pro:
           - highly secure
           con:
           - highly sophisticated, requires browser support, described in Section 4.7.1.2
      will prevent the attacker from replaying the access tokens on
      other resource servers.  Depending on the severity of the
      penetration, it
           work for native apps?

   per instance client id/secret  ...

   Note will also prevent replay on pre-warmed secrets: An attacker can circumvent the
   countermeasures described above if he compromised
      system.

   o  Audience restriction as described in Section 4.7.1.3 may be used
      to prevent replay of captured access tokens on other resource
      servers.

4.8.  Refresh Token Leakage (TBD)

   mitm, log files on the device, ...

   refresh token rotation, mutual TLS authentication at the token
   endpoint

4.9.  Open Redirection (TBD)

   Using the AS as Open Redirector - error handling AS (redirects)
   (draft-ietf-oauth-closing-redirectors-00)

   Using the Client as Open Redirector

4.10.  TLS Terminating Reverse Proxies

   A common deployment architecture for HTTP applications is able to create or capture have the respective secret or code_challenge on
   application server sitting behind a device under his
   control, reverse proxy, which is then used in the victim's authorization request.
   Exact redirect URI matching of authorization requests can prevent the
   attacker from using terminates
   the pre-warmed secret in TLS connection and dispatches the faked authorization
   transaction on incoming requests to the victim's device.
   Unfortunately it does not work for all kinds
   respective application server nodes.

   This section highlights some attack angles of OAuth clients.  It is
   effective for web and JS apps this deployment
   architecture, which are relevant to OAuth, and give recommendations
   for native apps with claimed URLs.
   What about other native apps?  Treat nonce or PKCE challenge as
   replay detection tokens (needs security controls.

   In some situations, the reverse proxy needs to ensure cluster-wide one-time use)?

5.2.  Access Token Injection (TBD)

   Note: An attacker in possession pass security-related
   data to the upstream application servers for further processing.
   Examples include the IP address of an access the request originator, token can access
   binding ids and authenticated TLS client certificates.

   If the reverse proxy would pass through any
   resources header sent from the access token gives him
   outside, an attacker could try to directly send the permission to.  This kind of
   attacks simply illustrates faked header
   values through the fact that bearer tokens utilized by
   OAuth are reusable similar proxy to passwords unless they are protected by
   further means.

   (where do we treat access token replay/use at the resource server?
   https://tools.ietf.org/html/rfc6819#section-4.6.4 has some text about application server in order to
   circumvent security controls that way.  For example, it but is standard
   practice of reverse proxies to accept "forwarded_for" headers and
   just add the origin of the inbound request (making it sufficient?)

   The attack described in this section is about injecting a stolen
   access token into a legitimate client list).
   Depending on a device under the
   adversaries control.  The logic performed in the application server, the
   attacker wants could simply add a whitelisted IP address to impersonate the header and
   render a victim IP whitelist useless.  A reverse proxy must therefore
   sanitize any inbound requests to ensure the authenticity and
   cannot use his own client, since he wants
   integrity of all header values relevant for the security of the
   application servers.

   If an attacker would be able to get access certain functions
   in this particular client.

   Proposal: token binding, hybrid flow+nonce(OIDC), other
   cryptographical binding to the internal network
   between access token proxy and user agent instance

5.3.  XSRF (TBD)

   injection of code or access token on a victim's device (e.g. application server, it could also try to cause
   client circumvent
   security controls in place.  It is therefore important to access resources under ensure the attacker's control)

   mitigation: XSRF tokens (one time use) w/ user agent binding (cf.
   https://www.owasp.org/index.php/
   CrossSite_Request_Forgery_(CSRF)_Prevention_Cheat_Sheet)

6.  Other Attacks

   Using
   authenticity of the AS as Open Redirector - error handling AS (redirects)
   (draft-ietf-oauth-closing-redirectors-00)

   Using communicating entities.  Furthermore, the Client as Open Redirector
   communication link between reverse proxy and application server must
   therefore be protected against tapping and injection (including
   replay prevention).

4.11.  Other Topics

   o  redirect via status code 307 - use 302

7.  Other Topics

   o  why to rotate refresh tokens

   o  how to support multi AS per RS

   o  differentiate native, JS and web clients

   o  do not put sensitive data in URL/GET parameters (Jim Manico)
   o  Incorporate Christian Mainka's feedback

   o  WPAD attack - https://www.blackhat.com/docs/us-16/materials/us-16-
      Kotler-Crippling-HTTPS-With-Unholy-PAC.pdf

8.

5.  Acknowledgements

   We would like to thank Jim Manico, Phil Hunt, and Brian Campbell for
   their valuable feedback.

9.

6.  IANA Considerations

   This draft includes no request to IANA.

10.

7.  Security Considerations

   All relevant security considerations have been given in the
   functional specification.

11.

8.  References

11.1.

8.1.  Normative References

   [RFC3986]  Berners-Lee, T., Fielding, R., and L. Masinter, "Uniform
              Resource Identifier (URI): Generic Syntax", STD 66,
              RFC 3986, DOI 10.17487/RFC3986, January 2005,
              <https://www.rfc-editor.org/info/rfc3986>.

   [RFC6749]  Hardt, D., Ed., "The OAuth 2.0 Authorization Framework",
              RFC 6749, DOI 10.17487/RFC6749, October 2012,
              <https://www.rfc-editor.org/info/rfc6749>.

   [RFC6750]  Jones, M. and D. Hardt, "The OAuth 2.0 Authorization
              Framework: Bearer Token Usage", RFC 6750,
              DOI 10.17487/RFC6750, October 2012,
              <https://www.rfc-editor.org/info/rfc6750>.

   [RFC6819]  Lodderstedt, T., Ed., McGloin, M., and P. Hunt, "OAuth 2.0
              Threat Model and Security Considerations", RFC 6819,
              DOI 10.17487/RFC6819, January 2013,
              <https://www.rfc-editor.org/info/rfc6819>.

   [RFC7231]  Fielding, R., Ed. and J. Reschke, Ed., "Hypertext Transfer
              Protocol (HTTP/1.1): Semantics and Content", RFC 7231,
              DOI 10.17487/RFC7231, June 2014,
              <https://www.rfc-editor.org/info/rfc7231>.

   [RFC7591]  Richer, J., Ed., Jones, M., Bradley, J., Machulak, M., and
              P. Hunt, "OAuth 2.0 Dynamic Client Registration Protocol",
              RFC 7591, DOI 10.17487/RFC7591, July 2015,
              <https://www.rfc-editor.org/info/rfc7591>.

11.2.

8.2.  Informative References

   [I-D.bradley-oauth-jwt-encoded-state]
              Bradley, J., Lodderstedt, T., and H. Zandbelt, "Encoding
              claims in the OAuth 2 state parameter using a JWT", draft-
              bradley-oauth-jwt-encoded-state-07 (work in progress),
              March 2017.

   [I-D.campbell-oauth-resource-indicators]
              Campbell, B., Bradley, J., and H. Tschofenig, "Resource
              Indicators for OAuth 2.0", draft-campbell-oauth-resource-
              indicators-02 (work in progress), November 2016.

   [I-D.ietf-oauth-discovery]
              Jones, M., Sakimura, N., and J. Bradley, "OAuth 2.0
              Authorization Server Metadata", draft-ietf-oauth-
              discovery-07 (work in progress), September 2017.

   [I-D.ietf-oauth-mtls]
              Campbell, B., Bradley, J., Sakimura, N., and T.
              Lodderstedt, "Mutual TLS Profile for OAuth 2.0", draft-
              ietf-oauth-mtls-03
              ietf-oauth-mtls-05 (work in progress), July November 2017.

   [I-D.ietf-oauth-pop-key-distribution]
              Bradley, J., Hunt, P., Jones, M., and H. Tschofenig,
              "OAuth 2.0 Proof-of-Possession: Authorization Server to
              Client Key Distribution", draft-ietf-oauth-pop-key-
              distribution-03 (work in progress), February 2017.

   [I-D.ietf-oauth-signed-http-request]
              Richer, J., Bradley, J., and H. Tschofenig, "A Method for
              Signing HTTP Requests for OAuth", draft-ietf-oauth-signed-
              http-request-03 (work in progress), August 2016.

   [I-D.ietf-oauth-token-binding]
              Jones, M., Bradley, J., Campbell, B., Bradley, J., and W. Denniss,
              "OAuth 2.0 Token Binding", draft-ietf-oauth-token-
              binding-04
              binding-05 (work in progress), July October 2017.

   [I-D.ietf-tokbind-https]
              Popov, A., Nystrom, M., Balfanz, D., Langley, A., Harper,
              N., and J. Hodges, "Token Binding over HTTP", draft-ietf-
              tokbind-https-10 (work in progress), July 2017.

   [I-D.sakimura-oauth-jpop]
              Sakimura, N., Li, K., and J. Bradley, "The OAuth 2.0
              Authorization Framework: JWT Pop Token Usage", draft-
              sakimura-oauth-jpop-04 (work in progress), March 2017.

   [oauth_security_cmu]
              Carnegie Mellon University, Carnegie Mellon University,
              Microsoft Research, Carnegie Mellon University, Carnegie
              Mellon University, and Carnegie Mellon University, "OAuth
              Demystified for Mobile Application Developers", November
              2014.

   [oauth_security_ubc]
              University of British Columbia and University of British
              Columbia, "The Devil is in the (Implementation) Details:
              An Empirical Analysis of OAuth SSO Systems", October 2012,
              <http://passwordresearch.com/papers/paper267.html>.

   [owasp]    "Open Web Application Security Project Home Page",
              <https://www.owasp.org/>.

   [RFC7636]  Sakimura, N., Ed., Bradley, J., and N. Agarwal, "Proof Key
              for Code Exchange by OAuth Public Clients", RFC 7636,
              DOI 10.17487/RFC7636, September 2015,
              <https://www.rfc-editor.org/info/rfc7636>.

   [RFC7800]  Jones, M., Bradley, J., and H. Tschofenig, "Proof-of-
              Possession Key Semantics for JSON Web Tokens (JWTs)",
              RFC 7800, DOI 10.17487/RFC7800, April 2016,
              <https://www.rfc-editor.org/info/rfc7800>.

Appendix A.  Document History

   [[ To be removed from the final specification ]]

   -04

   o  Restructured document for better readability

   o  Added best practices on Token Leakage prevention

   -03

   o  Added section on Access Token Leakage at Resource Server

   o  incorporated Brian Campbell's findings

   -02
   o  Folded Mix up and Access Token leakage through a bad AS into new
      section for dynamic OAuth threats

   o  reworked dynamic OAuth section

   -01

   o  Added references to mitigation methods for token leakage

   o  Added reference to Token Binding for Authorization Code

   o  incorporated feedback of Phil Hunt

   o  fixed numbering issue in attack descriptions in section 2

   -00 (WG document)

   o  turned the ID into a WG document and a BCP

   o  Added federated app login as topic in Other Topics

Authors' Addresses

   Torsten Lodderstedt (editor)
   YES Europe
   YES.com AG

   Email: torsten@lodderstedt.net

   John Bradley
   Yubico

   Email: ve7jtb@ve7jtb.com

   Andrey Labunets
   Facebook

   Email: isciurus@fb.com