Simple Cloud Identity Management: Protocol 1.0

The SCIM protocol specifies well known endpoints and HTTP methods for managing Resources defined in the core schema; i.e., User and Group Resources correspond to /Users and /Groups respectively. Service Providers that support extended Resources SHOULD define Resource endpoints using the established convention; pluralize the Resource name defined in the extended schema by appending an 's'. Given there are cases where Resource pluralization is ambiguous; e.g., a Resource named 'person' is legitimately 'persons' and 'people' Consumers SHOULD discover Resource endpoints via the Schema Sub-Attribute 'endpoint'. Retrieves a complete or partial Resource. Create new Resource or bulk modify Resources. Modifies a Resource with a complete, Consumer specified Resource (replace). Modifies a Resource with a set of Consumer specified changes (partial update). Deletes a Resource. Resource Endpoint Operations Description User /Users GET, POST, PUT, PATCH, DELETE Retrieve/Add/Modify Users Group /Groups GET, POST, PUT, PATCH, DELETE Retrieve/Add/Modify Groups Service Provider Configuration /ServiceProviderConfigs GET Retrieve the Service Provider's Configuration Schema /Schemas GET Retrieve a Resource's Schema Bulk /Bulk POST Bulk modify Resources All requests to the Service Provider are made via HTTP operations on a URL derived from the Base URL. Responses are returned in the body of the HTTP response, formatted as JSON or XML, depending on what is requested. Response and error codes SHOULD be transmitted via the HTTP status code of the response (if possible), and SHOULD also be specified in the body of the response.

To create new Resources, clients send POST requests to the Resource endpoint; i.e., /Users or /Groups. Successful Resource creation is indicated with a 201 ("Created") response code. Upon successful creation, the response body MUST contain the newly created Resource. Since the server is free to alter and/or ignore POSTed content, returning the full representation can be useful to the client, enabling it to correlate the client and server views of the new Resource. When a Resource is created, its URI must be returned in the response Location header. Below, the client sends a POST request containing a User

The server signals a successful creation with a status code of 201. The response includes a Location header indicating the User URI, and a representation of that User in the body of the response.

Users and Group Resources are retrieved via opaque, unique URLs or via Query. Service Providers MAY choose to respond with a sub-set of Resource attributes, though MUST minimally return the Resource id and meta attributes.

To retrieve a known Resource, clients send GET requests to the Resource endpoint; e.g., /Users/{id} or /Groups/{id}. If the Resource exists the server responds with a status code of 200 and includes the result in the body of the response. The below example retrieves a single User via the /Users endpoint.

The server responds with:

SCIM defines a standard set of operations that can be used to filter, sort, and paginate response results. The operations are specified by adding query parameters to the Resource's endpoint. Service Providers MAY support additional query parameters not specified here, and Providers SHOULD ignore any query parameters they don't recognize. The below example returns the userName for all Users:

Filtering is OPTIONAL. Consumers may request a subset of Resources by specifying the 'filter' URL query parameter containing a filter expression. When specified only those Resources matching the filter expression SHALL be returned. The expression language that is used in the filter parameter supports references to attributes and literals. The literal values can be strings enclosed in double quotes, numbers, date times enclosed in double quotes, and Boolean values; i.e., true or false. String literals MUST be valid JSON strings. The attribute name and attribute operator are case insensitive. For example, the following two expressions will evaluate to the same logical value:

The filter parameter MUST contain at least one valid Boolean expression. Each expression MUST contain an attribute name followed by an attribute operator and optional value. Multiple expressions MAY be combined using the two logical operators. Furthermore expressions can be grouped together using "()". The operators supported in the expression are listed in the following table. Operator Description Behavior eq equal The attribute and operator values must be identical for a match. co contains The entire operator value must be a substring of the attribute value for a match. sw starts with The entire operator value must be a substring of the attribute value, starting at the beginning of the attribute value. This criterion is satisfied if the two strings are identical. pr present (has value) If the attribute has a non-empty value, or if it contains a non-empty node for complex attributes there is a match. gt greater than If the attribute value is greater than operator value, there is a match. The actual comparison is dependent on the attribute type. For string attribute types, this is a lexicographical comparison and for DateTime types, it is a chronological comparison. ge greater than or equal If the attribute value is greater than or equal to the operator value, there is a match. The actual comparison is dependent on the attribute type. For string attribute types, this is a lexicographical comparison and for DateTime types, it is a chronological comparison. lt less than If the attribute value is less than operator value, there is a match. The actual comparison is dependent on the attribute type. For string attribute types, this is a lexicographical comparison and for DateTime types, it is a chronological comparison. le less than or equal If the attribute value is less than or equal to the operator value, there is a match. The actual comparison is dependent on the attribute type. For string attribute types, this is a lexicographical comparison and for DateTime types, it is a chronological comparison. Operator Description Behavior and Logical And The filter is only a match if both expressions evaluate to true. or Logical or The filter is a match if either expression evaluates to true. Operator Description Behavior () Precedence grouping Boolean expressions may be grouped using parentheses to change the standard order of operations; i.e., evaluate OR logical operators before logical AND operators. Filters MUST be evaluated using standard order of operations . Attribute operators have the highest precedence, followed by the grouping operator (i.e, parentheses), followed by the logical AND operator, followed by the logical OR operator. If the specified attribute in a filter expression is a multi-valued attribute, the Resource MUST match if any of the instances of the given attribute match the specified criterion; e.g. if a User has multiple emails values, only one has to match for the entire User to match. For complex attributes, a fully qualified Sub-Attribute MUST be specified using standard attribute notation. For example, to filter by userName the parameter value is userName and to filter by first name, the parameter value is name.givenName. Providers MAY support additional filter operations if they choose. Providers MUST decline to filter results if the specified filter operation is not recognized and return a HTTP 400 error with an appropriate human readable response. For example, if a Consumer specified an unsupported operator named 'regex' the Service Provider should specify an error response description identifying the Consumer error; e.g., 'The operator 'regex' is not supported.' String type attributes are case insensitive by default unless the attribute type is defined as a caseExact string. Attribute operators 'eq', 'co', and 'sw' MUST perform caseIgnore matching for all string attributes unless the attribute is defined as caseExact. By default all string attributes are caseIgnore. Examples:

Sort is OPTIONAL. Sorting allows Consumers to specify the order in which Resources are returned by specifying a combination of sortBy and sortOrder URL parameters. The sortBy parameter specifies the attribute whose value SHALL be used to order the returned responses. If the sortBy attribute corresponds to a Singular Attribute, Resources are sorted according to that attribute's value; if it's a Multi-valued Attribute, Resources are sorted by the value of the primary attribute, if any, or else the first value in the list, if any. If the attribute is complex the attribute name must be a path to a Sub-Attribute in standard attribute notation ; e.g., sortBy=name.givenName. For all attribute types, if there is no data for the specified sortBy value they are sorted via the 'sortOrder' parameter; i.e., they are ordered last if ascending and first if descending. The order in which the sortBy parameter is applied. Allowed values are "ascending" and "descending". If a value for sortBy is provided and no sortOrder is specified, the sortOrder SHALL default to ascending. String type attributes are case insensitive by default unless the attribute type is defined as a caseExact string. sortOrder MUST sort according to the attribute type; i.e., for caseIgnore attributes, sort the result using case insensitive, Unicode alphabetic sort order, with no specific locale implied and for caseExact attribute types, sort the result using case sensitive, Unicode alphabetic sort order.

Pagination parameters can be used together to "page through" large numbers of Resources so as not to overwhelm the Consumer or Service Provider. Pagination is not session based hence Consumers SHOULD never assume repeatable results. For example, a request for a list of 10 Resources beginning with a startIndex of 1 may return different results when repeated as a Resource in the original result could be deleted or new ones could be added in-between requests. Pagination parameters and general behavior are derived from the OpenSearch Protocol . The following table describes the URL pagination parameters. Parameter Description Default startIndex The 1-based index of the first search result. 1 count Non-negative Integer. Specifies the desired maximum number of search results per page; e.g., 10. None. When specified the Service Provider MUST not return more results than specified though MAY return fewer results. If unspecified, the maximum number of results is set by the Service Provider. The following table describes the query response pagination attributes specified by the Service Provider. Element Description itemsPerPage Non-negative Integer. Specifies the number of search results returned in a query response page; e.g., 10. totalResults Non-negative Integer. Specifies the total number of results matching the Consumer query; e.g., 1000. startIndex The 1-based index of the first result in the current set of search results; e.g., 1. For example, to retrieve the first 10 Users set the startIndex to 1 and the count to 10.

Given the example above, to continue paging set the startIndex to 11 and re-fetch; i.e., /Users?startIndex=11&count=10

Resources can be modified in whole or in part via PUT or PATCH, respectively. Implementers MUST support PUT as specified in RFC2616 . Resources such as Groups may be very large hence implementers SHOULD support PATCH to enable partial resource modifications.

PUT performs a full update. Consumers must retrieve the entire Resource and PUT the desired modifications as the operation overwrites all previously stored data. Unless otherwise specified a successful PUT operation returns a 200 OK response code and the entire Resource within the response body. Example:

The service responds with the entire, updated User

PATCH is OPTIONAL. PATCH enables consumers to send only those attributes requiring modification, reducing network and processing overhead. Attributes may be deleted, replaced, merged, or added in a single request. The body of a PATCH request MUST contain a partial Resource with the desired modifications. The server MUST return either a 200 OK response code and the entire Resource (subject to the "attributes" query parameter - see Additional Retrieval Query Parameters) within the response body, or a 204 No Content response code and the appropriate response headers for a successful PATCH request. The server MUST return a 200 OK if the "attributes" parameter is specified on the request. The server MUST process a PATCH request by first removing any attributes specified in the meta.attributes Sub-Attribute (if present) and then merging the attributes in the PATCH request body into the Resource. The meta.attributes Sub-Attribute MAY contain a list of attributes to be removed from the Resource. If the PATCH request body contains an attribute that is present in the meta.attributes list, the attribute on the Resource is replaced with the value from the PATCH body. If the attribute is complex the attribute name must be a path to a Sub-Attribute in standard attribute notation; e.g., name.givenName. Attributes that exist in the PATCH request body but not in the meta.attributes Sub-Attribute will be either be updated or added to the Resource according to the following rules. Singular attributes in the PATCH request body replace the attribute on the Resource. Complex Sub-Attribute values in the PATCH request body are merged into the complex attribute on the Resource. An attribute value in the PATCH request body is added to the value collection if the value does not exist and merged if a matching value is present. Values are matched by comparing the value Sub-Attribute from the PATCH request body to the value Sub-Attribute of the Resource. Attributes that do not have a value Sub-Attribute; e.g., addresses, or do not have unique value Sub-Attributes cannot be matched and must instead be deleted then added. Specific values can be removed from a Resource by adding an "operation" Sub-Attribute with the value "delete" to the attribute in the PATCH request body. As with adding/updating attribute value collections, the value to delete is determined by comparing the value Sub-Attribute from the PATCH request body to the value Sub-Attribute of the Resource. Attributes that do not have a value Sub-Attribute or that have a non-unique value Sub-Attribute are matched by comparing all Sub-Attribute values from the PATCH request body to the Sub-Attribute values of the Resource. A delete operation is ignored if the attribute's name is in the meta.attributes list. If the requested value to delete does not match a unique value on the Resource the server MAY return a HTTP 400 error. The following example shows how to add a member to a group:

The "display" Sub-Attribute in this request is optional since the value attribute uniquely identifies the user to be added. If the user was already a member of this group, no changes should be made to the Resource and a success response should be returned. The server responds with either the entire updated Group or no response body:

The following example shows how to remove a member from a group. As with the previous example, the "display" Sub-Attribute is optional. If the user was not a member of this group, no changes should be made to the Resource and a success response should be returned. Note that server responses have been omitted for the rest of the PATCH examples.

The following example shows how to remove all members from a group:

The following example shows how to replace all of the members of a group with a different members list:

The following example shows how to add a member to and remove a member from a Group in a single request:

The following example shows how to change a User's primary email. If the User already has the email address, it is made the primary address and the current primary address (if present) is made non-primary. If the User does not already have the email address, it is added and made the primary address.

The following example shows how to change a User's address. Since address does not have a value Sub-Attribute, the existing address must be removed and the modified address added.

The following example shows how to change a User's nickname:

The following example shows how to remove a User's nickname:

The following example shows how to change a User's familyName. This only updates the familyName and formatted on the "name" complex attribute. Any other name Sub-Attributes on the Resource remain unchanged.

The following example shows how to remove a complex Sub-Attribute and an extended schema attribute from a User.

Consumers request Resource removal via DELETE. Service Providers MAY choose not to permanently delete the Resource, but MUST return a 404 error code for all operations associated with the previously deleted Id. Service Providers MUST also omit the Resource from future query results.

Server Response:

Example: Consumer attempt to retrieve the previously deleted User

Server Response:

Bulk is OPTIONAL. The bulk operation enables Consumers to send a potentially large collection of Resource operations in a single request. The body of a a bulk operation contains a set of HTTP Resource operations using one of the API supported HTTP methods; i.e., POST, PUT, PATCH or DELETE. The following Singular Attribute is defined in addition to the common attributes defined in SCIM core schema. An Integer specifying the number of errors that the Service Provider will accept before the operation is terminated and an error response is returned. OPTIONAL. The following Complex Multi-valued Attribute is defined in addition to the common attributes defined in core schema. Defines operations within a bulk job. Each operation corresponds to a single HTTP request against a Resource endpoint. REQUIRED. The HTTP method of the current operation. Possible values are POST, PUT, PATCH or DELETE. REQUIRED. The transient identifier of a newly created Resource, unique within a bulk request and created by the Consumer. The bulkId serves as a surrogate Resource id enabling Consumers to uniquely identify newly created Resources in the Response and cross reference new Resources in and across operations within a bulk request. REQUIRED when method is POST. The current Resource version. Version is REQUIRED if the Service Provider supports ETags and the method is PUT, DELETE, or PATCH. The Resource's relative path. If the method is POST the value must specify a Resource type endpoint; e.g., /Users or /Groups whereas all other method values must specify the path to a specific Resource; e.g., /Users/2819c223-7f76-453a-919d-413861904646. REQUIRED in a request. The Resource data as it would appear for a single POST, PUT or PATCH Resource operation. REQUIRED in a request when method is POST, PUT and PATCH. The Resource endpoint URL. REQUIRED in a response, except in the event of a POST failure. A complex type that contains information about the success or failure of one operation within the bulk job. REQUIRED in a response. The HTTP response code that would have been returned if a a single HTTP request would have been used. REQUIRED. A human readable error message. REQUIRED when an error occurred. If a bulk job is processed successfully the HTTP response code 200 OK MUST be returned, otherwise an appropriate HTTP error code MUST be returned. The Service Provider MUST continue performing as many changes as possible and disregard partial failures. The Consumer MAY override this behavior by specifying a value for failOnErrors attribute. The failOnErrors attribute defines the number of errors that the Service Provider should accept before failing the remaining operations returning the response. To be able to reference a newly created Resource the attribute bulkId MUST be specified when creating new Resources. The bulkId is defined by the Consumer as a surrogate identifier in a POST operation. The Service Provider MUST return the same bulkId together with the newly created Resource. The bulkId can then be used by the Consumer to map the Service Provider id with the bulkId of the created Resource. There can be more then one operation per Resource in each bulk job. The Service Consumer MUST take notice of the unordered structure of JSON and the Service Provider can process operations in any order. For example, if the Service Consumer sends two PUT operations in one request, the outcome is non-deterministic. The Service Provider response MUST include the result of all processed operations. A location attribute that includes the Resource's end point MUST be returned for all operations excluding failed POSTs. The status attribute includes information about the success or failure of one operation within the bulk job. The attribute status MUST include the code attribute that holds the HTTP response code that would have been returned if a single HTTP request would have been used. If an error occurred the status MUST also include the description attribute containing a human readable explanation of the error.

The following is an example of a status in a failed operation.

The following example shows how to add, update, and remove a user. The failOnErrors attribute is set to '1' indicating the Service Provider should return on the first error. The POST operation's bulkId value is set to 'qwerty' enabling the Consumer to match the new User with the returned Resource id '92b725cd-9465-4e7d-8c16-01f8e146b87a'.

The Service Provider returns the following response.

The following response is returned if an error occurred when attempting to create the User 'Alice'. The Service Provider stops processing the bulk operation and immediately returns a response to the Consumer. The response contains the error and any successful results prior to the error.

If the failOnErrors attribute is not specified or the Service Provider has not reached the error limit defined by the Consumer the Service Provider will continue to process all operations. The following is an example in which all operations failed.

The Consumer can, within one bulk operation, create a new User, a new Group and add the newly created User to the newly created Group. In order to add the new User to the Group the Consumer must use the surrogate id attribute, bulkId, to reference the User. The bulkId attribute value must be pre-pended with the literal "bulkId:"; e.g., if the bulkId is 'qwerty' the value is “bulkId:qwerty”. The Service Provider MUST replace the string “bulkId:qwerty” with the permanent Resource id once created. The following example creates a User with the userName 'Alice' and a Group with the displayName 'Tour Guides' with Alice as a member.

The Service Provider returns the following response.

A subsequent request for the 'Tour Guides' Group ('e9e30dba-f08f-4109-8486-d5c6a331660a') returns the following:

Extensions that include references to other Resources MUST be handled in the same way by the Service Provider. The following example uses the bulkId attribute within the enterprise extension managerId attribute.

The Service Provider MUST try to resolve circular cross references between Resources in a single bulk job but MAY stop after a failed attempt and instead return the status code 409 Conflict. The following example exhibits the potential conflict.

If the Service Provider resolved the above circular references the following is returned from a subsequent GET request.

The Service Provider MUST define the maximum number of operations and maximum payload size a Consumer may send in a single request. If either limits are exceeded the Service Provider MUST return the HTTP response code 413 Request Entity Too Large. The returned response MUST specify the limit exceeded in the body of the error response. The following example the Consumer sent a request exceeding the Service Provider's max payload size of 1 megabyte.

Consumers MUST specify the format in which the data is submitted via the HTTP header content-type and MAY specify the desired response data format via an HTTP Accept Header; e.g.,"Accept: application/json" or via URI suffix; e.g.,

Service Providers MUST support the Accept Headers "Accept: application/json" for JSON and, if supported, "Accept: application/xml" for XML. The format defaults to JSON if no format is specified. The data structure returned is equivalent in both formats; the only difference is in the encoding of the data. Singular attributes are encoded as string name-value-pairs in JSON; e.g.,

and elements in XML; e.g.,

value ]]> Multi-valued attributes in JSON are encoded as arrays; e.g.,

and repeated tags in XML; e.g.,

value1 value2 ]]> Elements with nested elements are represented as objects in JSON; e.g,

and repeated tags in XML; e.g.,

value1 value2 ]]>

Consumers MAY request a partial Resource representation on any operation that returns a Resource within the response by specifying the URL query parameter 'attributes'. When specified, each Resource returned MUST contain the minimal set of Resource attributes and, MUST contain no other attributes or Sub-Attributes than those explicitly requested. The query parameter attributes value is a comma separated list of Resource attribute names in standard, attribute notation form (e.g. userName, name, emails).

Giving the response

All operations share a common scheme for referencing simple and complex attributes. In general, attributes are identified by prefixing the attribute name with its schema URN separated by a ':' character; e.g., the core User Resource attribute 'userName' is identified as 'urn:scim:schemas:core:1.0:userName'. Consumers MAY omit core schema attribute URN prefixes though MUST fully qualify extended attributes with the associated Resource URN; e.g., the attribute 'age' defined in 'urn:hr:schemas:user' is fully encoded as 'urn:hr:schemas:user:age'. A Complex attributes' Sub-Attributes are referenced via nested, dot ('.') notation; i.e., {urn}:{Attribute name}.{Sub-Attribute name}. For example, the fully qualified path for a User's givenName is urn:scim:schemas:core:1.0:name.givenName All facets (URN, attribute and Sub-Attribute name) of the fully encoded Attribute name are case insensitive.

The SCIM Protocol uses the response status codes defined in HTTP to indicate operation success or failure. In addition to returning a HTTP response code implementers MUST return the errors in the body of the response in the client requested format containing the error response and, per the HTTP specification, human-readable explanations. Implementers SHOULD handle the identified errors as described below. Code Applicability Suggested Explanation 400 BAD REQUEST GET,POST,PUT,PATCH,DELETE Request is unparseable, syntactically incorrect, or violates schema 401 UNAUTHORIZED GET,POST,PUT,PATCH,DELETE Authorization failure 403 FORBIDDEN GET,POST,PUT,PATCH,DELETE Server does not support requested operation 404 NOT FOUND GET,PUT,PATCH,DELETE Specified resource; e.g., User, does not exist 409 CONFLICT PUT,PATCH,DELETE The specified version number does not match the resource's latest version number or a Service Provider refused to create a new, duplicate resource 412 PRECONDITION FAILED PUT,PATCH,DELETE Failed to update as Resource {id} changed on the server last retrieved 413 REQUEST ENTITY TOO LARGE POST {"maxOperations": 1000,"maxPayload": 1048576} 500 INTERNAL SERVER ERROR GET,POST,PUT,PATCH,DELETE An internal error. Implementers SHOULD provide descriptive debugging advice 501 NOT IMPLEMENTED GET,POST,PUT,PATCH,DELETE Service Provider does not support the request operation; e.g., PATCH Error example in response to a non-existent GET request.

The Base URL MAY be appended with a version identifier as a separate segment in the URL path. At this time the only valid identifier is 'v1'. If specified, the version identifier MUST appear in the URL path immediately preceding the Resource endpoint and conform to the following scheme: the character 'v' followed by the desired SCIM version number; e.g., a version 'v1' User request is specified as /v1/Users. When specified Service Providers MUST perform the operation using the desired version or reject the request. When omitted Service Providers SHOULD perform the operation using the most recent API supported by the Service Provider.

The API supports resource versioning via standard, HTTP ETags. Service providers MAY support weak ETags as the preferred mechanism for performing conditional retrievals and ensuring Consumers do not inadvertently overwrite each others changes, respectively. When supported SCIM ETags MUST be specified as an HTTP header and SHOULD be specified within the 'version' attribute contained in the Resource's 'meta' attribute. Example:

The server responds with an ETag in the response header and meta structure.

With the returned ETag, Consumers MAY choose to retrieve the Resource only if the Resource has been modified. In addition, if updating, specifying an ETag guarantees that previous updates will not be overwritten. Conditional retrieval example using If-None-Match header:

If the Resource has not changed the Service Provider simply returns an empty body with a 304 "Not Modified" response code. Similarly, consumers MAY supply an If-Match ETag header for PUT, PATCH, and DELETE operations to ensure that the requested operation succeeds only if the supplied ETag matches the latest Service Provider Resource; e.g., If-Match: W/"e180ee84f0671b1" To perform DELETE, PATCH, or PUT operations and override ETag versioning specify "If-Match: *" in lieu of an ETag header.

In recognition that some clients, servers and firewalls prevent PUT, PATCH and DELETE operations a client MAY override the POST operation by specifying the custom header "X-HTTP-Method-Override" with the desired PUT, PATCH, DELETE operation. For example: