RFC 7516
JSON Web Encryption (JWE)
5. Producing and Consuming JWEs
5.1. Message Encryption
The message encryption process is as follows. The order of the steps is not significant in cases where there are no dependencies between the inputs and outputs of the steps. 1. Determine the Key Management Mode employed by the algorithm used to determine the Content Encryption Key value. (This is the algorithm recorded in the "alg" (algorithm) Header Parameter of the resulting JWE.) 2. When Key Wrapping, Key Encryption, or Key Agreement with Key Wrapping are employed, generate a random CEK value. See RFC 4086 [RFC4086] for considerations on generating random values. The CEK MUST have a length equal to that required for the content encryption algorithm. 3. When Direct Key Agreement or Key Agreement with Key Wrapping are employed, use the key agreement algorithm to compute the value of the agreed upon key. When Direct Key Agreement is employed, let the CEK be the agreed upon key. When Key Agreement with Key Wrapping is employed, the agreed upon key will be used to wrap the CEK. 4. When Key Wrapping, Key Encryption, or Key Agreement with Key Wrapping are employed, encrypt the CEK to the recipient and let the result be the JWE Encrypted Key. 5. When Direct Key Agreement or Direct Encryption are employed, let the JWE Encrypted Key be the empty octet sequence.
6. When Direct Encryption is employed, let the CEK be the shared
symmetric key.
7. Compute the encoded key value BASE64URL(JWE Encrypted Key).
8. If the JWE JSON Serialization is being used, repeat this process
(steps 1-7) for each recipient.
9. Generate a random JWE Initialization Vector of the correct size
for the content encryption algorithm (if required for the
algorithm); otherwise, let the JWE Initialization Vector be the
empty octet sequence.
10. Compute the encoded Initialization Vector value BASE64URL(JWE
Initialization Vector).
11. If a "zip" parameter was included, compress the plaintext using
the specified compression algorithm and let M be the octet
sequence representing the compressed plaintext; otherwise, let M
be the octet sequence representing the plaintext.
12. Create the JSON object(s) containing the desired set of Header
Parameters, which together comprise the JOSE Header: one or more
of the JWE Protected Header, the JWE Shared Unprotected Header,
and the JWE Per-Recipient Unprotected Header.
13. Compute the Encoded Protected Header value BASE64URL(UTF8(JWE
Protected Header)). If the JWE Protected Header is not present
(which can only happen when using the JWE JSON Serialization and
no "protected" member is present), let this value be the empty
string.
14. Let the Additional Authenticated Data encryption parameter be
ASCII(Encoded Protected Header). However, if a JWE AAD value is
present (which can only be the case when using the JWE JSON
Serialization), instead let the Additional Authenticated Data
encryption parameter be ASCII(Encoded Protected Header || '.' ||
BASE64URL(JWE AAD)).
15. Encrypt M using the CEK, the JWE Initialization Vector, and the
Additional Authenticated Data value using the specified content
encryption algorithm to create the JWE Ciphertext value and the
JWE Authentication Tag (which is the Authentication Tag output
from the encryption operation).
16. Compute the encoded ciphertext value BASE64URL(JWE Ciphertext).
17. Compute the encoded Authentication Tag value BASE64URL(JWE
Authentication Tag).
18. If a JWE AAD value is present, compute the encoded AAD value
BASE64URL(JWE AAD).
19. Create the desired serialized output. The Compact Serialization
of this result is the string BASE64URL(UTF8(JWE Protected
Header)) || '.' || BASE64URL(JWE Encrypted Key) || '.' ||
BASE64URL(JWE Initialization Vector) || '.' || BASE64URL(JWE
Ciphertext) || '.' || BASE64URL(JWE Authentication Tag). The
JWE JSON Serialization is described in Section 7.2.
5.2. Message Decryption
The message decryption process is the reverse of the encryption
process. The order of the steps is not significant in cases where
there are no dependencies between the inputs and outputs of the
steps. If any of these steps fail, the encrypted content cannot be
validated.
When there are multiple recipients, it is an application decision
which of the recipients' encrypted content must successfully validate
for the JWE to be accepted. In some cases, encrypted content for all
recipients must successfully validate or the JWE will be considered
invalid. In other cases, only the encrypted content for a single
recipient needs to be successfully validated. However, in all cases,
the encrypted content for at least one recipient MUST successfully
validate or the JWE MUST be considered invalid.
1. Parse the JWE representation to extract the serialized values
for the components of the JWE. When using the JWE Compact
Serialization, these components are the base64url-encoded
representations of the JWE Protected Header, the JWE Encrypted
Key, the JWE Initialization Vector, the JWE Ciphertext, and the
JWE Authentication Tag, and when using the JWE JSON
Serialization, these components also include the base64url-
encoded representation of the JWE AAD and the unencoded JWE
Shared Unprotected Header and JWE Per-Recipient Unprotected
Header values. When using the JWE Compact Serialization, the
JWE Protected Header, the JWE Encrypted Key, the JWE
Initialization Vector, the JWE Ciphertext, and the JWE
Authentication Tag are represented as base64url-encoded values
in that order, with each value being separated from the next by
a single period ('.') character, resulting in exactly four
delimiting period characters being used. The JWE JSON
Serialization is described in Section 7.2.
2. Base64url decode the encoded representations of the JWE
Protected Header, the JWE Encrypted Key, the JWE Initialization
Vector, the JWE Ciphertext, the JWE Authentication Tag, and the
JWE AAD, following the restriction that no line breaks,
whitespace, or other additional characters have been used.
3. Verify that the octet sequence resulting from decoding the
encoded JWE Protected Header is a UTF-8-encoded representation
of a completely valid JSON object conforming to RFC 7159
[RFC7159]; let the JWE Protected Header be this JSON object.
4. If using the JWE Compact Serialization, let the JOSE Header be
the JWE Protected Header. Otherwise, when using the JWE JSON
Serialization, let the JOSE Header be the union of the members
of the JWE Protected Header, the JWE Shared Unprotected Header
and the corresponding JWE Per-Recipient Unprotected Header, all
of which must be completely valid JSON objects. During this
step, verify that the resulting JOSE Header does not contain
duplicate Header Parameter names. When using the JWE JSON
Serialization, this restriction includes that the same Header
Parameter name also MUST NOT occur in distinct JSON object
values that together comprise the JOSE Header.
5. Verify that the implementation understands and can process all
fields that it is required to support, whether required by this
specification, by the algorithms being used, or by the "crit"
Header Parameter value, and that the values of those parameters
are also understood and supported.
6. Determine the Key Management Mode employed by the algorithm
specified by the "alg" (algorithm) Header Parameter.
7. Verify that the JWE uses a key known to the recipient.
8. When Direct Key Agreement or Key Agreement with Key Wrapping are
employed, use the key agreement algorithm to compute the value
of the agreed upon key. When Direct Key Agreement is employed,
let the CEK be the agreed upon key. When Key Agreement with Key
Wrapping is employed, the agreed upon key will be used to
decrypt the JWE Encrypted Key.
9. When Key Wrapping, Key Encryption, or Key Agreement with Key
Wrapping are employed, decrypt the JWE Encrypted Key to produce
the CEK. The CEK MUST have a length equal to that required for
the content encryption algorithm. Note that when there are
multiple recipients, each recipient will only be able to decrypt
JWE Encrypted Key values that were encrypted to a key in that
recipient's possession. It is therefore normal to only be able
to decrypt one of the per-recipient JWE Encrypted Key values to
obtain the CEK value. Also, see Section 11.5 for security
considerations on mitigating timing attacks.
10. When Direct Key Agreement or Direct Encryption are employed,
verify that the JWE Encrypted Key value is an empty octet
sequence.
11. When Direct Encryption is employed, let the CEK be the shared
symmetric key.
12. Record whether the CEK could be successfully determined for this
recipient or not.
13. If the JWE JSON Serialization is being used, repeat this process
(steps 4-12) for each recipient contained in the representation.
14. Compute the Encoded Protected Header value BASE64URL(UTF8(JWE
Protected Header)). If the JWE Protected Header is not present
(which can only happen when using the JWE JSON Serialization and
no "protected" member is present), let this value be the empty
string.
15. Let the Additional Authenticated Data encryption parameter be
ASCII(Encoded Protected Header). However, if a JWE AAD value is
present (which can only be the case when using the JWE JSON
Serialization), instead let the Additional Authenticated Data
encryption parameter be ASCII(Encoded Protected Header || '.' ||
BASE64URL(JWE AAD)).
16. Decrypt the JWE Ciphertext using the CEK, the JWE Initialization
Vector, the Additional Authenticated Data value, and the JWE
Authentication Tag (which is the Authentication Tag input to the
calculation) using the specified content encryption algorithm,
returning the decrypted plaintext and validating the JWE
Authentication Tag in the manner specified for the algorithm,
rejecting the input without emitting any decrypted output if the
JWE Authentication Tag is incorrect.
17. If a "zip" parameter was included, uncompress the decrypted
plaintext using the specified compression algorithm.
18. If there was no recipient for which all of the decryption steps
succeeded, then the JWE MUST be considered invalid. Otherwise,
output the plaintext. In the JWE JSON Serialization case, also
return a result to the application indicating for which of the
recipients the decryption succeeded and failed.
Finally, note that it is an application decision which algorithms may be used in a given context. Even if a JWE can be successfully decrypted, unless the algorithms used in the JWE are acceptable to the application, it SHOULD consider the JWE to be invalid.5.3. String Comparison Rules
The string comparison rules for this specification are the same as those defined in Section 5.3 of [JWS].6. Key Identification
The key identification methods for this specification are the same as those defined in Section 6 of [JWS], except that the key being identified is the public key to which the JWE was encrypted.7. Serializations
JWEs use one of two serializations: the JWE Compact Serialization or the JWE JSON Serialization. Applications using this specification need to specify what serialization and serialization features are used for that application. For instance, applications might specify that only the JWE JSON Serialization is used, that only JWE JSON Serialization support for a single recipient is used, or that support for multiple recipients is used. JWE implementations only need to implement the features needed for the applications they are designed to support.7.1. JWE Compact Serialization
The JWE Compact Serialization represents encrypted content as a compact, URL-safe string. This string is: BASE64URL(UTF8(JWE Protected Header)) || '.' || BASE64URL(JWE Encrypted Key) || '.' || BASE64URL(JWE Initialization Vector) || '.' || BASE64URL(JWE Ciphertext) || '.' || BASE64URL(JWE Authentication Tag) Only one recipient is supported by the JWE Compact Serialization and it provides no syntax to represent JWE Shared Unprotected Header, JWE Per-Recipient Unprotected Header, or JWE AAD values.7.2. JWE JSON Serialization
The JWE JSON Serialization represents encrypted content as a JSON object. This representation is neither optimized for compactness nor URL safe.
Two closely related syntaxes are defined for the JWE JSON Serialization: a fully general syntax, with which content can be encrypted to more than one recipient, and a flattened syntax, which is optimized for the single-recipient case.7.2.1. General JWE JSON Serialization Syntax
The following members are defined for use in top-level JSON objects used for the fully general JWE JSON Serialization syntax: protected The "protected" member MUST be present and contain the value BASE64URL(UTF8(JWE Protected Header)) when the JWE Protected Header value is non-empty; otherwise, it MUST be absent. These Header Parameter values are integrity protected. unprotected The "unprotected" member MUST be present and contain the value JWE Shared Unprotected Header when the JWE Shared Unprotected Header value is non-empty; otherwise, it MUST be absent. This value is represented as an unencoded JSON object, rather than as a string. These Header Parameter values are not integrity protected. iv The "iv" member MUST be present and contain the value BASE64URL(JWE Initialization Vector) when the JWE Initialization Vector value is non-empty; otherwise, it MUST be absent. aad The "aad" member MUST be present and contain the value BASE64URL(JWE AAD)) when the JWE AAD value is non-empty; otherwise, it MUST be absent. A JWE AAD value can be included to supply a base64url-encoded value to be integrity protected but not encrypted. ciphertext The "ciphertext" member MUST be present and contain the value BASE64URL(JWE Ciphertext). tag The "tag" member MUST be present and contain the value BASE64URL(JWE Authentication Tag) when the JWE Authentication Tag value is non-empty; otherwise, it MUST be absent. recipients The "recipients" member value MUST be an array of JSON objects. Each object contains information specific to a single recipient. This member MUST be present with exactly one array element per
recipient, even if some or all of the array element values are the
empty JSON object "{}" (which can happen when all Header Parameter
values are shared between all recipients and when no encrypted key
is used, such as when doing Direct Encryption).
The following members are defined for use in the JSON objects that
are elements of the "recipients" array:
header
The "header" member MUST be present and contain the value JWE Per-
Recipient Unprotected Header when the JWE Per-Recipient
Unprotected Header value is non-empty; otherwise, it MUST be
absent. This value is represented as an unencoded JSON object,
rather than as a string. These Header Parameter values are not
integrity protected.
encrypted_key
The "encrypted_key" member MUST be present and contain the value
BASE64URL(JWE Encrypted Key) when the JWE Encrypted Key value is
non-empty; otherwise, it MUST be absent.
At least one of the "header", "protected", and "unprotected" members
MUST be present so that "alg" and "enc" Header Parameter values are
conveyed for each recipient computation.
Additional members can be present in both the JSON objects defined
above; if not understood by implementations encountering them, they
MUST be ignored.
Some Header Parameters, including the "alg" parameter, can be shared
among all recipient computations. Header Parameters in the JWE
Protected Header and JWE Shared Unprotected Header values are shared
among all recipients.
The Header Parameter values used when creating or validating per-
recipient ciphertext and Authentication Tag values are the union of
the three sets of Header Parameter values that may be present: (1)
the JWE Protected Header represented in the "protected" member, (2)
the JWE Shared Unprotected Header represented in the "unprotected"
member, and (3) the JWE Per-Recipient Unprotected Header represented
in the "header" member of the recipient's array element. The union
of these sets of Header Parameters comprises the JOSE Header. The
Header Parameter names in the three locations MUST be disjoint.
Each JWE Encrypted Key value is computed using the parameters of the
corresponding JOSE Header value in the same manner as for the JWE
Compact Serialization. This has the desirable property that each JWE
Encrypted Key value in the "recipients" array is identical to the
value that would have been computed for the same parameter in the JWE
Compact Serialization. Likewise, the JWE Ciphertext and JWE
Authentication Tag values match those produced for the JWE Compact
Serialization, provided that the JWE Protected Header value (which
represents the integrity-protected Header Parameter values) matches
that used in the JWE Compact Serialization.
All recipients use the same JWE Protected Header, JWE Initialization
Vector, JWE Ciphertext, and JWE Authentication Tag values, when
present, resulting in potentially significant space savings if the
message is large. Therefore, all Header Parameters that specify the
treatment of the plaintext value MUST be the same for all recipients.
This primarily means that the "enc" (encryption algorithm) Header
Parameter value in the JOSE Header for each recipient and any
parameters of that algorithm MUST be the same.
In summary, the syntax of a JWE using the general JWE JSON
Serialization is as follows:
{
"protected":"<integrity-protected shared header contents>",
"unprotected":<non-integrity-protected shared header contents>,
"recipients":[
{"header":<per-recipient unprotected header 1 contents>,
"encrypted_key":"<encrypted key 1 contents>"},
...
{"header":<per-recipient unprotected header N contents>,
"encrypted_key":"<encrypted key N contents>"}],
"aad":"<additional authenticated data contents>",
"iv":"<initialization vector contents>",
"ciphertext":"<ciphertext contents>",
"tag":"<authentication tag contents>"
}
See Appendix A.4 for an example JWE using the general JWE JSON
Serialization syntax.
7.2.2. Flattened JWE JSON Serialization Syntax
The flattened JWE JSON Serialization syntax is based upon the general
syntax, but flattens it, optimizing it for the single-recipient case.
It flattens it by removing the "recipients" member and instead
placing those members defined for use in the "recipients" array (the
"header" and "encrypted_key" members) in the top-level JSON object
(at the same level as the "ciphertext" member).
The "recipients" member MUST NOT be present when using this syntax.
Other than this syntax difference, JWE JSON Serialization objects
using the flattened syntax are processed identically to those using
the general syntax.
In summary, the syntax of a JWE using the flattened JWE JSON
Serialization is as follows:
{
"protected":"<integrity-protected header contents>",
"unprotected":<non-integrity-protected header contents>,
"header":<more non-integrity-protected header contents>,
"encrypted_key":"<encrypted key contents>",
"aad":"<additional authenticated data contents>",
"iv":"<initialization vector contents>",
"ciphertext":"<ciphertext contents>",
"tag":"<authentication tag contents>"
}
Note that when using the flattened syntax, just as when using the
general syntax, any unprotected Header Parameter values can reside in
either the "unprotected" member or the "header" member, or in both.
See Appendix A.5 for an example JWE using the flattened JWE JSON
Serialization syntax.
8. TLS Requirements
The Transport Layer Security (TLS) requirements for this
specification are the same as those defined in Section 8 of [JWS].
9. Distinguishing between JWS and JWE Objects
There are several ways of distinguishing whether an object is a JWS
or JWE. All these methods will yield the same result for all legal
input values; they may yield different results for malformed inputs.
o If the object is using the JWS Compact Serialization or the JWE
Compact Serialization, the number of base64url-encoded segments
separated by period ('.') characters differs for JWSs and JWEs.
JWSs have three segments separated by two period ('.') characters.
JWEs have five segments separated by four period ('.') characters.
o If the object is using the JWS JSON Serialization or the JWE JSON
Serialization, the members used will be different. JWSs have a
"payload" member and JWEs do not. JWEs have a "ciphertext" member
and JWSs do not.
o The JOSE Header for a JWS can be distinguished from the JOSE
Header for a JWE by examining the "alg" (algorithm) Header
Parameter value. If the value represents a digital signature or
MAC algorithm, or is the value "none", it is for a JWS; if it
represents a Key Encryption, Key Wrapping, Direct Key Agreement,
Key Agreement with Key Wrapping, or Direct Encryption algorithm,
it is for a JWE. (Extracting the "alg" value to examine is
straightforward when using the JWS Compact Serialization or the
JWE Compact Serialization and may be more difficult when using the
JWS JSON Serialization or the JWE JSON Serialization.)
o The JOSE Header for a JWS can also be distinguished from the JOSE
Header for a JWE by determining whether an "enc" (encryption
algorithm) member exists. If the "enc" member exists, it is a
JWE; otherwise, it is a JWS.
10. IANA Considerations
10.1. JSON Web Signature and Encryption Header Parameters Registration
This section registers the Header Parameter names defined in
Section 4.1 in the IANA "JSON Web Signature and Encryption Header
Parameters" registry established by [JWS].
10.1.1. Registry Contents
o Header Parameter Name: "alg"
o Header Parameter Description: Algorithm
o Header Parameter Usage Location(s): JWE
o Change Controller: IESG
o Specification Document(s): Section 4.1.1 of RFC 7516
o Header Parameter Name: "enc"
o Header Parameter Description: Encryption Algorithm
o Header Parameter Usage Location(s): JWE
o Change Controller: IESG
o Specification Document(s): Section 4.1.2 of RFC 7516
o Header Parameter Name: "zip"
o Header Parameter Description: Compression Algorithm
o Header Parameter Usage Location(s): JWE
o Change Controller: IESG
o Specification Document(s): Section 4.1.3 of RFC 7516
o Header Parameter Name: "jku" o Header Parameter Description: JWK Set URL o Header Parameter Usage Location(s): JWE o Change Controller: IESG o Specification Document(s): Section 4.1.4 of RFC 7516 o Header Parameter Name: "jwk" o Header Parameter Description: JSON Web Key o Header Parameter Usage Location(s): JWE o Change Controller: IESG o Specification Document(s): Section 4.1.5 of RFC 7516 o Header Parameter Name: "kid" o Header Parameter Description: Key ID o Header Parameter Usage Location(s): JWE o Change Controller: IESG o Specification Document(s): Section 4.1.6 of RFC 7516 o Header Parameter Name: "x5u" o Header Parameter Description: X.509 URL o Header Parameter Usage Location(s): JWE o Change Controller: IESG o Specification Document(s): Section 4.1.7 of RFC 7516 o Header Parameter Name: "x5c" o Header Parameter Description: X.509 Certificate Chain o Header Parameter Usage Location(s): JWE o Change Controller: IESG o Specification Document(s): Section 4.1.8 of RFC 7516 o Header Parameter Name: "x5t" o Header Parameter Description: X.509 Certificate SHA-1 Thumbprint o Header Parameter Usage Location(s): JWE o Change Controller: IESG o Specification Document(s): Section 4.1.9 of RFC 7516 o Header Parameter Name: "x5t#S256" o Header Parameter Description: X.509 Certificate SHA-256 Thumbprint o Header Parameter Usage Location(s): JWE o Change Controller: IESG o Specification Document(s): Section 4.1.10 of RFC 7516 o Header Parameter Name: "typ" o Header Parameter Description: Type o Header Parameter Usage Location(s): JWE o Change Controller: IESG o Specification Document(s): Section 4.1.11 of RFC 7516
o Header Parameter Name: "cty" o Header Parameter Description: Content Type o Header Parameter Usage Location(s): JWE o Change Controller: IESG o Specification Document(s): Section 4.1.12 of RFC 7516 o Header Parameter Name: "crit" o Header Parameter Description: Critical o Header Parameter Usage Location(s): JWE o Change Controller: IESG o Specification Document(s): Section 4.1.13 of RFC 751611. Security Considerations
All of the security issues that are pertinent to any cryptographic application must be addressed by JWS/JWE/JWK agents. Among these issues are protecting the user's asymmetric private and symmetric secret keys and employing countermeasures to various attacks. All the security considerations in the JWS specification also apply to this specification. Likewise, all the security considerations in XML Encryption 1.1 [W3C.REC-xmlenc-core1-20130411] also apply, other than those that are XML specific.11.1. Key Entropy and Random Values
See Section 10.1 of [JWS] for security considerations on key entropy and random values. In addition to the uses of random values listed there, note that random values are also used for Content Encryption Keys (CEKs) and Initialization Vectors (IVs) when performing encryption.11.2. Key Protection
See Section 10.2 of [JWS] for security considerations on key protection. In addition to the keys listed there that must be protected, implementations performing encryption must protect the key encryption key and the Content Encryption Key. Compromise of the key encryption key may result in the disclosure of all contents protected with that key. Similarly, compromise of the Content Encryption Key may result in disclosure of the associated encrypted content.
11.3. Using Matching Algorithm Strengths
Algorithms of matching strengths should be used together whenever possible. For instance, when AES Key Wrap is used with a given key size, using the same key size is recommended when AES GCM is also used. If the key encryption and content encryption algorithms are different, the effective security is determined by the weaker of the two algorithms. Also, see RFC 3766 [RFC3766] for information on determining strengths for public keys used for exchanging symmetric keys.11.4. Adaptive Chosen-Ciphertext Attacks
When decrypting, particular care must be taken not to allow the JWE recipient to be used as an oracle for decrypting messages. RFC 3218 [RFC3218] should be consulted for specific countermeasures to attacks on RSAES-PKCS1-v1_5. An attacker might modify the contents of the "alg" Header Parameter from "RSA-OAEP" to "RSA1_5" in order to generate a formatting error that can be detected and used to recover the CEK even if RSAES-OAEP was used to encrypt the CEK. It is therefore particularly important to report all formatting errors to the CEK, Additional Authenticated Data, or ciphertext as a single error when the encrypted content is rejected. Additionally, this type of attack can be prevented by restricting the use of a key to a limited set of algorithms -- usually one. This means, for instance, that if the key is marked as being for "RSA-OAEP" only, any attempt to decrypt a message using the "RSA1_5" algorithm with that key should fail immediately due to invalid use of the key.11.5. Timing Attacks
To mitigate the attacks described in RFC 3218 [RFC3218], the recipient MUST NOT distinguish between format, padding, and length errors of encrypted keys. It is strongly recommended, in the event of receiving an improperly formatted key, that the recipient substitute a randomly generated CEK and proceed to the next step, to mitigate timing attacks.
(next page on part 3)
