Task: Land the role request

Covered by Complete the ores.inbox approvals and notifications: its task Land the role request kind lands the request, the four-eyes guard and the grant, and Build the request and notification screens the member's ask screen and the administrator's queue.

• An approval request of kind iam.role_grant carries the ask, and the asker cannot decide it.
• An approval grants the role through the ordinary assignment path, exactly once.
• The member asks from My access and reads the answer there; the administrator answers from the Requests queue.

The journey planned the request as a workflow_instance and named five IAM request subjects. Both were replaced before anything was built: the request is an approval request of its own with a kind, and the queue and the decision are generic across kinds, so they belong to ores.inbox. The work moved to the Complete the ores.inbox approvals and notifications story, whose tasks Land the role request kind and Build the request and notification screens delivered it. This task is closed as covered, and the two gaps the journey names that nothing has answered are new tasks on this story.

Every number this task claims about a tree, with the command that produced it and the commit it was measured at. A number without both is not a measurement, because a clean-up record describes a tree and trunk moves under a branch: a consumer count, a gate's component count and a suite's size all change with no code changing, and the reader of a bare number cannot tell which tree it came from. Where the raw output matters, put it in an evidence/ file beside the task and name that file in the Command column.

Manual QA scenarios (scaffolded via compass add test_scenario) that verify this task. Link new ones here as they're created; the scenario doc itself links back via its "Verifies task" field.

Covered, not built here. The request, the four-eyes guard, the grant, the two notifications and both screens landed under the inbox story, and the evidence for them sits beside its tasks. Its screens task records the live walk: a member asked for three roles in turn, the queue named the role on every row, each decision granted the role, and the member read the decision in the bell.

Two of the journey's seven gaps were left unanswered, and they are now tasks of their own on this story: nothing says which roles may be asked for, and a request that nobody answers waits forever.