AWS Support (support) awsJson1_1 service for fakecloud.
The full 16-operation AWS Support Smithy model: the Support Cases API
(CreateCase / DescribeCases / DescribeCommunications /
AddCommunicationToCase / ResolveCase), attachment sets
(AddAttachmentsToSet / DescribeAttachment), the Trusted Advisor API
(DescribeTrustedAdvisorChecks / DescribeTrustedAdvisorCheckResult /
DescribeTrustedAdvisorCheckSummaries /
DescribeTrustedAdvisorCheckRefreshStatuses / RefreshTrustedAdvisorCheck),
severity levels (DescribeSeverityLevels), and the case-creation
reference operations (DescribeServices / DescribeCreateCaseOptions /
DescribeSupportedLanguages).
Requests carry X-Amz-Target: AWSSupport_20130415.<Operation>; dispatch
keys off req.action. Every operation runs model-driven input validation
first (required / length / range / enum / pattern), then real,
account-partitioned, persisted CRUD.
Support cases are real: CreateCase mints an AWS-shaped case id
(case-{account}-{year}-{hex}) and a numeric displayId, opens the case
status: "opened", and seeds its communication thread with the initial
communicationBody. DescribeCases filters by case-id list / display id /
time window / resolved inclusion / language and paginates with a
round-tripping nextToken. AddCommunicationToCase appends to the thread,
DescribeCommunications pages it, and ResolveCase flips the case to
resolved, returning the initial and final status. Attachment sets are real
(AddAttachmentsToSet mints/extends an attachmentSetId with an
expiryTime; DescribeAttachment returns a stored attachment by id).
Severity levels and the Trusted Advisor check catalogue are faithful static
AWS reference data; the Trusted Advisor refresh status is a real per-check
none -> enqueued -> processing -> success state machine.
Honest gap: fakecloud does not run the real Trusted Advisor analysis engine
or attach a live AWS Support agent. DescribeTrustedAdvisorCheckResult and
DescribeTrustedAdvisorCheckSummaries return well-formed, structurally
correct result shapes reporting zero flagged resources (an all-clear
account) rather than fabricating findings, and no automated support-agent
reply is generated for a case. Cases, communications, attachment sets,
severity levels, the check catalogue, and the refresh state machine are all
real, account-partitioned, and persisted.