Skip to content

AIFACTORY-FEAT-078: Studio work area: Operations & usage

Generated from the architecture model

This page is rendered from model/**/*.yaml by tools/docs_model during mkdocs build. Edit the YAML, not this page; hand-written design lives on the linked service, platform and plane pages.

A design artefact, not a tracker

This backlog is generated from model/backlog/ and validated against the architecture model. No work item exists in any tracker, and no item carries a date, an estimate or an assignee (ADR-0029).

Epic Factory Studio (PF-14) · Increment I3 Studio · State elaborated · Route frontend-developer · Build order 118 of 185 (Factory code)

The 13 screens of the Studio work area Operations & usage on the Blazor shell and microfrontends, bound to the read models and commands the model names for them, with their state families, roles, locales and accessibility contracts. Screens come from model/ux/screens.yaml and the prototype alignment.

Delivers —
Stubs —
Realizes ux work areas: Operations & usage
Requirements —
Depends on —
Sources model/ux/screens.yaml, model/ux/prototype-alignment.yaml

AIFACTORY-US-1401: Operations: I read the timeline of a run from its workflow graph and its attempt timelines while the run projection of the dashboard is not yet delivered

behavior · persona operator As an operator, I want to read, for one run, the state of every step, the phases, checkpoints, calls and decisions of each attempt and its outcome, so that I understand where a run stands or why it failed without opening a log, and I am told which part of the timeline is not yet available.

A requester, a reviewer, an operator or an observer opens the screen run-timeline at /operations/runs/{runId}. StudioBff composes it under the scope of the session from GetWorkflowRunGraph of WorkflowEngineService, the graph of the run with the state of every step run, its timers and its compensation entries, and from GetAttemptTimeline of AgentRuntimeService for each attempt the user expands: phases, checkpoints, calls by hash and receipt reference and never by content, decisions and the outcome. Both services are reached through AgentRuntimeWorker, the internal gateway of the Agent Mesh platform. The third read model of the screen, GetRunTimeline, is the run projection of DashboardService, which is delivered in I4; in I3 the page shows that source as unknown in the freshness family, named, and builds the timeline from the graph and the attempts only, never from a fake projection. The screen has no command. StudioBff consumes StepRunStateChanged of WorkflowEngineService and AttemptPhaseChanged of AgentRuntimeService, refreshes the workflow run and attempt views and signals the open page, which updates the state of the step and the phase of the attempt without a reload. A run of another tenant is answered 404 like a run that does not exist, a missing scope 403, and the page says which. When AgentRuntimeWorker does not answer, the composition shows the Agent Mesh source degraded with its last freshness and keeps the parts it already holds marked stale. A failed attempt shows its failure class and its last receipt reference, which is the adverse branch of the flow recover-run of model/ux/flows.yaml that I3 can show.

Acceptance criteria

AC1 Given an operator on run-timeline, when StudioBff/GetScreenModel composes it, then the steps come from WorkflowEngineService/GetWorkflowRunGraph and each expanded attempt from AgentRuntimeService/GetAttemptTimeline, calls are shown by hash and receipt reference and never by content, and the run projection source is shown unknown and named (FUX-04).
AC2 Given an open timeline, when WorkflowEngineService/StepRunStateChanged or AgentRuntimeService/AttemptPhaseChanged is consumed by StudioBff, then the page shows the new step state or phase without a reload, and a failed attempt shows its failure class and its last receipt reference as the flow recover-run of model/ux/flows.yaml describes (FUX-07).
AC3 Given a run of another tenant, a missing scope or an unreachable AgentRuntimeWorker, when the timeline is requested, then the answer is 404 or 403 explained in words, or the Agent Mesh source is shown degraded with its last freshness and the parts already held are marked stale (ADR-0015, model/services/factory-studio/StudioBff.yaml).
Traces: services StudioBff, WorkflowEngineService, AgentRuntimeService
Traces: apis StudioBff/GetScreenModel, WorkflowEngineService/GetWorkflowRunGraph, AgentRuntimeService/GetAttemptTimeline
Traces: events WorkflowEngineService/StepRunStateChanged, AgentRuntimeService/AttemptPhaseChanged
Traces: adrs ADR-0015, ADR-0024
Traces: requirements FUX-04, FUX-07
Relates: screens run-timeline
Depends on feature:studio-bff, feature:component-blazor-shell, feature:workflow-engine-service, feature:agent-runtime-service, feature:agent-runtime-worker
Sources model/ux/screens.yaml, model/ux/flows.yaml, model/services/agent-mesh/WorkflowEngineService.yaml, model/services/agent-mesh/AgentRuntimeService.yaml

Tasks

Task Kind Expected output Repository
AIFACTORY-TASK-5651 Run timeline screen in the Operations microfrontend ui The Operations microfrontend renders run-timeline with the run and freshness families, the steps of the graph with their states, timers and compensation entries, expandable attempts with phases, checkpoints, calls by hash and receipt reference, decisions and outcome, and the run projection source marked unknown with its name. ConnectSoft.AI.SoftwareFactory.Studio
AIFACTORY-TASK-5652 Screen model of the run timeline in StudioBff contract StudioBff composes run-timeline through AgentRuntimeWorker from GetWorkflowRunGraph and GetAttemptTimeline with the freshness of each source, marks the run projection source unknown in I3, and refreshes the workflow run and attempt views on StepRunStateChanged and AttemptPhaseChanged. ConnectSoft.AI.SoftwareFactory.StudioBff
AIFACTORY-TASK-5653 Verify the run timeline with live changes and a failed attempt verification Run dotnet test -c Release --no-build and npx playwright test --project=chromium with a StudioBff stub serving a graph, two attempts, a failed attempt with its failure class, a pushed step change, a 404 of another tenant and an unreachable gateway; the tests prove AC1 to AC3 of AIFACTORY-US-1401 (WorkflowEngineService/GetWorkflowRunGraph, AgentRuntimeService/GetAttemptTimeline). ConnectSoft.AI.SoftwareFactory.Studio
AIFACTORY-TASK-5702 Verify the StudioBff side of the run timeline verification Run dotnet test -c Release --filter "Category=Acceptance&Feature=studio-operations-run-timeline-from-attempts-and-graph" in StudioBff against stubbed platform gateways, where a filter that matches no test fails the run; the tests prove on the StudioBff side AC1 (composition through AgentRuntimeWorker from GetWorkflowRunGraph and GetAttemptTimeline, calls by hash only, the run projection source marked unknown), AC2 (refresh on StepRunStateChanged and AttemptPhaseChanged) and AC3 (404 and 403 returned unchanged, the Agent Mesh source degraded with its last freshness while AgentRuntimeWorker does not answer) of AIFACTORY-US-1401 (StudioBff/GetScreenModel). ConnectSoft.AI.SoftwareFactory.StudioBff

AIFACTORY-US-1402: Operations: I read project and run usage by status with uncertain amounts kept apart, and set a project budget whose threshold may only lower the platform one

behavior · persona workspace-admin As a workspace admin, I want to see what the project and each run consumed, by meter and by status, and to set the project budget, so that I steer spending on attributed numbers and never mistake an uncertain amount for a settled one.

A workspace admin, an operator or an observer opens the screen usage-cost at /operations/usage. StudioBff composes it under the scope of the session from CostUsageService, reached through ControlPlaneApi: GetProjectUsage for a period, with usage by meter and by status (estimated, reserved, observed, reconciled, uncertain), and GetRunUsage for the usage records and the reservation of one run. The usage family of model/ux/states.yaml governs the display: every amount shows its status and its currency code, and an uncertain amount is shown apart and never folded into a total, which is the adverse branch of the flow submit-assessment of model/ux/flows.yaml. A workspace admin sets the project budget and its threshold with SetProjectBudget, sent through ForwardCommand with an idempotency key; SetProjectBudget of model/services/control-plane/CostUsageService.yaml takes a threshold not above the platform one, so a higher threshold is answered 409 and the page says that a project may only lower it; the flow govern-knowledge-and-usage words the same refusal against the threshold of the tenant. A project of another tenant is answered 404. An operator and an observer see the control disabled with the reason. The two commercial read models of the screen, GetSubscription and ListInvoices of FactoryCommercialService, are delivered in I7; in I3 the page shows that source as unknown in the freshness family, named, and shows no subscription or invoice value. StudioBff consumes BudgetUpdated and ReservationSettled and refreshes the budget and reservation views, so a settled reservation moves from reserved to observed on the open page without a reload. When ControlPlaneApi does not answer, the source is shown degraded with its last freshness and the budget change is refused with 503, never queued.

Acceptance criteria

AC1 Given a workspace admin on usage-cost, when StudioBff/GetScreenModel composes it, then usage comes from CostUsageService/GetProjectUsage and CostUsageService/GetRunUsage by meter and status with the currency code, and an uncertain amount is shown apart and never added to a total (FUX-08, NFR-12).
AC2 Given a workspace admin, when they set a budget, then StudioBff/ForwardCommand forwards CostUsageService/SetProjectBudget with an idempotency key, a threshold above the platform one is answered 409 and explained as a project that may only lower it, as CostUsageService/SetProjectBudget in model/services/control-plane/CostUsageService.yaml states, a project of another tenant is answered 404, and an operator or observer sees the control disabled with the reason (ADR-0015).
AC3 Given the commercial source of the screen, when it opens in I3, then the subscription and invoice part is shown unknown and named with no invented value, and when CostUsageService/ReservationSettled or CostUsageService/BudgetUpdated is consumed by StudioBff the open page refreshes without a reload (model/ux/states.yaml).
Traces: services StudioBff, CostUsageService
Traces: apis StudioBff/GetScreenModel, StudioBff/ForwardCommand, CostUsageService/GetProjectUsage, CostUsageService/GetRunUsage, CostUsageService/SetProjectBudget
Traces: events CostUsageService/BudgetUpdated, CostUsageService/ReservationSettled
Traces: adrs ADR-0015
Traces: requirements FUX-08
Relates: screens usage-cost
Relates: nfrs NFR-12
Depends on feature:studio-bff, feature:component-blazor-shell, feature:cost-usage-service-complete, feature:control-plane-api
Sources model/ux/screens.yaml, model/ux/states.yaml, model/ux/flows.yaml, model/services/control-plane/CostUsageService.yaml

Tasks

Task Kind Expected output Repository
AIFACTORY-TASK-5654 Usage and cost screen in the Operations microfrontend ui The Operations microfrontend renders usage-cost with the usage family, amounts by meter and status with their currency code, uncertain amounts in their own row never added to a total, a budget form with the platform threshold shown beside it, the refusal in words, and the commercial source marked unknown with its name. ConnectSoft.AI.SoftwareFactory.Studio
AIFACTORY-TASK-5655 Screen model and budget route of the usage screen in StudioBff contract StudioBff composes usage-cost through ControlPlaneApi from GetProjectUsage and GetRunUsage with the freshness of the source, marks the commercial source unknown in I3, forwards SetProjectBudget with an idempotency key and the scope of the session, refuses it with 503 while ControlPlaneApi does not answer, and refreshes the views on BudgetUpdated and ReservationSettled. ConnectSoft.AI.SoftwareFactory.StudioBff
AIFACTORY-TASK-5656 Verify the usage screen with uncertain amounts and a refused threshold verification Run dotnet test -c Release --no-build and npx playwright test --project=chromium with a StudioBff stub serving usage in all five statuses, a 409 on a threshold above the platform one, a 404 of another tenant and a settled reservation pushed live; the tests prove AC1 to AC3 of AIFACTORY-US-1402 (CostUsageService/SetProjectBudget). ConnectSoft.AI.SoftwareFactory.Studio
AIFACTORY-TASK-5703 Verify the StudioBff side of the usage screen verification Run dotnet test -c Release --filter "Category=Acceptance&Feature=studio-operations-usage-and-cost" in StudioBff against stubbed platform gateways, where a filter that matches no test fails the run; the tests prove on the StudioBff side AC1 (composition through ControlPlaneApi from GetProjectUsage and GetRunUsage with every status and the uncertain amounts apart), AC2 (SetProjectBudget forwarded once with its idempotency key, its 409 and 404 returned unchanged, 503 while ControlPlaneApi does not answer and nothing queued) and AC3 (the commercial source marked unknown, refresh on ReservationSettled and BudgetUpdated) of AIFACTORY-US-1402 (StudioBff/GetScreenModel, StudioBff/ForwardCommand). ConnectSoft.AI.SoftwareFactory.StudioBff

AIFACTORY-US-1403: Operations: I see worker pools, effective quotas and capacity leases, and change a pool or tighten a quota with the role and class ExecutionCapacityService requires

behavior · persona operator As an operator, I want to see the worker pools, the quotas in force and who holds which lease, and to have a pool or a quota changed by the role allowed to do it, so that runs that wait for capacity are explained by a named limit and capacity changes follow their class.

An operator, a workspace admin, a platform admin or an observer opens the screen capacity at /operations/capacity. StudioBff composes it under the scope of the session from ExecutionCapacityService, reached through RuntimeCloudHost: ListWorkerPools with kind, boundary, region, size and utilisation, GetQuotas with the effective quotas of the tenant and its projects and the edition source of each, and ListCapacityLeases with the active and recent leases by holder, pool and state. A platform admin creates or changes the bounds and autoscaling of a pool with SetWorkerPool, a production-class action that is applied through provisioning, so the page sends it through ForwardCommand with an idempotency key and the approval binding its class requires and shows the pool as changing until the state is observed; a workspace admin tightens a project or tenant quota with SetQuota and a platform admin sets the edition defaults with it. A value the service rejects is answered 422 and explained per field. An operator and an observer see the commands disabled with the reason. The read model ListComputePools and the command DrainComputePool belong to ComputePoolService, delivered in I8; in I3 the page shows that source as unknown in the freshness family, named, and offers no drain. StudioBff consumes WorkerPoolStateChanged, QuotaChanged and QuotaExceeded and refreshes the pool and quota views; a quota that is exceeded is shown with its limit named, which is the adverse branch of the flow manage-editions-and-billing of model/ux/flows.yaml that runs wait with the limit named. When RuntimeCloudHost does not answer, the source is shown degraded with its last freshness and the commands are refused with 503, never queued.

Acceptance criteria

AC1 Given an operator on capacity, when StudioBff/GetScreenModel composes it, then pools, quotas and leases come from ExecutionCapacityService/ListWorkerPools, ExecutionCapacityService/GetQuotas and ExecutionCapacityService/ListCapacityLeases with the edition source of each quota, and the compute pool source is shown unknown and named with no drain offered (FUX-08).
AC2 Given a platform admin who changes a pool or a workspace admin who tightens a quota, when StudioBff/ForwardCommand forwards ExecutionCapacityService/SetWorkerPool with the approval binding of its production class or ExecutionCapacityService/SetQuota, then a 422 is explained per field, the pool shows as changing until ExecutionCapacityService/WorkerPoolStateChanged is consumed, and an operator or observer sees the commands disabled with the reason (NFR-08).
AC3 Given a quota reached, when ExecutionCapacityService/QuotaExceeded or ExecutionCapacityService/QuotaChanged is consumed by StudioBff, then the open page shows the limit named without a reload, as the flow manage-editions-and-billing of model/ux/flows.yaml describes, and when RuntimeCloudHost does not answer the source is degraded and commands answer 503 (model/services/factory-studio/StudioBff.yaml).
Traces: services StudioBff, ExecutionCapacityService
Traces: apis StudioBff/GetScreenModel, StudioBff/ForwardCommand, ExecutionCapacityService/ListWorkerPools, ExecutionCapacityService/GetQuotas, ExecutionCapacityService/ListCapacityLeases, ExecutionCapacityService/SetWorkerPool, ExecutionCapacityService/SetQuota
Traces: events ExecutionCapacityService/WorkerPoolStateChanged, ExecutionCapacityService/QuotaChanged, ExecutionCapacityService/QuotaExceeded
Traces: requirements FUX-08
Relates: screens capacity
Relates: nfrs NFR-08
Depends on feature:studio-bff, feature:component-blazor-shell, feature:execution-capacity-service-complete, feature:runtime-cloud-host
Sources model/ux/screens.yaml, model/ux/flows.yaml, model/services/runtime-cloud/ExecutionCapacityService.yaml

Tasks

Task Kind Expected output Repository
AIFACTORY-TASK-5657 Capacity and quotas screen in the Operations microfrontend ui The Operations microfrontend renders capacity with the freshness family, pools with utilisation, quotas with their edition source, leases by holder and state, the pool editor for a platform admin with the class of the action beside it, the quota editor that only tightens for a workspace admin, field errors of a 422, and the compute pool source marked unknown. ConnectSoft.AI.SoftwareFactory.Studio
AIFACTORY-TASK-5658 Screen model and command routes of the capacity screen in StudioBff contract StudioBff composes capacity through RuntimeCloudHost from ListWorkerPools, GetQuotas and ListCapacityLeases with the freshness of the source, marks the compute pool source unknown in I3, forwards SetWorkerPool with its approval binding and SetQuota with an idempotency key, and refreshes the views on WorkerPoolStateChanged, QuotaChanged and QuotaExceeded. ConnectSoft.AI.SoftwareFactory.StudioBff
AIFACTORY-TASK-5659 Verify the capacity screen by role with a quota exceeded verification Run dotnet test -c Release --no-build and npx playwright test --project=chromium with a StudioBff stub serving pools, quotas and leases to each of the four roles, a 422 on a pool change, a pushed QuotaExceeded and an unreachable RuntimeCloudHost; the tests prove AC1 to AC3 of AIFACTORY-US-1403 (ExecutionCapacityService/SetWorkerPool, ExecutionCapacityService/SetQuota). ConnectSoft.AI.SoftwareFactory.Studio
AIFACTORY-TASK-5704 Verify the StudioBff side of the capacity screen verification Run dotnet test -c Release --filter "Category=Acceptance&Feature=studio-operations-capacity-and-quotas" in StudioBff against stubbed platform gateways, where a filter that matches no test fails the run; the tests prove on the StudioBff side AC1 (composition through RuntimeCloudHost from the pool, quota and lease reads, the compute pool source marked unknown), AC2 (SetWorkerPool forwarded with the approval binding of its production class and SetQuota with its idempotency key, a 422 returned unchanged, refresh on WorkerPoolStateChanged) and AC3 (refresh on QuotaExceeded and QuotaChanged, 503 while RuntimeCloudHost does not answer and nothing queued) of AIFACTORY-US-1403 (StudioBff/GetScreenModel, StudioBff/ForwardCommand). ConnectSoft.AI.SoftwareFactory.StudioBff

AIFACTORY-US-132: The Operations & usage work area of the Studio is bound to the model

behavior · persona operator As an operator, I want the Operations & usage work area of the Studio (13 screens) bound to the read models, commands, state families and roles the model names for it, so that I can follow runs, usage and capacity from one place and see plainly which operations views arrive later, with every control reflecting what I may do and how fresh what I see is.

This is the integration PBI of the Operations and usage area. The Operations microfrontend is generated from the microfrontend library template, composed into the Studio shell through its registry and holds the thirteen screens of model/ux/screens.yaml with their routes, read models, commands, state families and roles; it calls StudioBff only. A screen is offered to the roles its list names; for any other role it is absent from the navigation and GetScreenModel refuses it with 403. Three screens are delivered in I3 by AIFACTORY-US-1401 to AIFACTORY-US-1403. Four read the observability services of I4: run-list and overview read DashboardService, which in I4 serves only the run projections and the project and tenant overviews (D131), status-objectives reads SloService and alerts reads AlertRuleService. Six read the fleet services of I7: fleet-board, product, rollouts, incidents, on-call and offboardings. In I3 each of these ten screens is registered, offered to its roles and shows its sources as unknown in the freshness family, named, never hidden and never filled from a fake; none of its commands is offered, so the rebuild of a run projection, the acknowledgement of an alert and the fleet commands wait for those increments. Every screen subscribes through Realtime.Subscribe of StudioRealtimeGateway; when the channel drops the page marks its data stale and when it returns the page reloads the screen model. A command needing the approval binding of its class carries it to the owning service, whose gate decides. Every screen renders in English, Russian and Hebrew right to left and is built to the accessibility contract of docs/planes/experience/studio/localization-accessibility.md; the area evidence that follows this PBI checks it on the finished area.

Acceptance criteria

AC1 Given the shell with the Operations microfrontend, when each of the 13 screens of model/ux/screens.yaml opens for a role its list names, then it renders from StudioBff/GetScreenModel only, and for a role its list does not name the screen is not offered and StudioBff/GetScreenModel answers 403 (ADR-0007, ADR-0015).
AC2 Given run-list, overview, status-objectives, alerts, fleet-board, product, rollouts, incidents, on-call and offboardings, when they open in I3, then each shows its sources as unknown in the freshness family of model/ux/states.yaml with the source named, shows no invented value and offers none of its commands (FUX-04).
AC3 Given an open screen and its subscription through StudioRealtimeGateway/Realtime.Subscribe, when the channel drops, then the page marks the data stale and never presents it as current, and when the channel returns the page reloads StudioBff/GetScreenModel and shows the current state without a browser reload (FUX-08).
AC4 Given a command of a delivered screen, when a role issues it, then it is forwarded once through StudioBff/ForwardCommand with an idempotency key, the scope of the session and the approval binding its class requires, a role without the right sees the control disabled with the reason, and the screens render in en, ru and he right to left under docs/planes/experience/studio/localization-accessibility.md.
Traces: services StudioBff, StudioRealtimeGateway, WorkflowEngineService, AgentRuntimeService, CostUsageService, ExecutionCapacityService
Traces: apis StudioBff/GetStudioSession, StudioBff/GetScreenModel, StudioBff/ForwardCommand, StudioRealtimeGateway/Realtime.Subscribe
Traces: adrs ADR-0007, ADR-0015
Traces: requirements FUX-04, FUX-08
Relates: screens run-list, overview, status-objectives, alerts, fleet-board, product, rollouts, incidents, on-call, offboardings
Depends on AIFACTORY-US-1401, AIFACTORY-US-1402, AIFACTORY-US-1403, feature:studio-bff, feature:studio-realtime-gateway, feature:component-blazor-shell
Sources model/ux/screens.yaml, model/ux/states.yaml, model/ux/prototype-alignment.yaml, docs/planes/experience/studio/screens/operations-usage.md

Tasks

Task Kind Expected output Repository
AIFACTORY-TASK-442 Screens of the work area Operations & usage: Runs, Run timeline, Operations overview, Status and objectives, Alerts, Usage and cost, Capacity and quotas, Fleet board and 5 more ui The Operations microfrontend is composed into the shell through the registry with one navigation group and the thirteen routes of model/ux/screens.yaml; each screen renders from StudioBff only, and the ten screens of later sources render their sources as unknown with the source named and no command. ConnectSoft.AI.SoftwareFactory.Studio
AIFACTORY-TASK-443 Commands of the work area Operations & usage implementation The commands of the three delivered screens are forwarded through StudioBff with an idempotency key, the scope of the session and the approval binding their class requires, and the commands of the ten screens of later sources (projection rebuild, objectives, alerts, fleet, rollouts, incidents, on-call, off-boarding, compute pools) are not offered in I3. ConnectSoft.AI.SoftwareFactory.Studio
AIFACTORY-TASK-444 States, roles, locales and accessibility of the work area Operations & usage implementation Each screen shows the state families the model names for it, hides or disables what the role may not do with the reason, marks stale data on a dropped channel and reloads on its return, renders in every supported locale including right-to-left, and passes the accessibility checks of the UX model. ConnectSoft.AI.SoftwareFactory.Studio
AIFACTORY-TASK-746 Verify studio-operations-usage-work-area by running the declared command verification Run dotnet build -c Release, dotnet test -c Release --no-build and npx playwright test --project=chromium; they pass and prove AC1 to AC4 of AIFACTORY-US-132 (StudioBff/GetScreenModel, StudioRealtimeGateway/Realtime.Subscribe). ConnectSoft.AI.SoftwareFactory.Studio

AIFACTORY-US-1404: Operations and usage area evidence: every screen passes the accessibility and phone-width checks and an end-to-end walk of the area through StudioBff is recorded

evidence · persona observer As an observer, I want proof that every screen of the Operations and usage area is usable by keyboard, in every locale and at phone width, and that the screens of the fourth and later increments say they are not yet delivered, so that the area is accepted on a recorded check and no operations figure is read from a fake.

This PBI is the proof of the Operations and usage area and comes after its integration PBI AIFACTORY-US-132, which registers the thirteen screens. The accessibility and phone-width check opens each of the thirteen screens of the area for a role its list in model/ux/screens.yaml names and asserts the rules of docs/planes/experience/studio/localization-accessibility.md: WCAG 2.2 AA by the automated checks, keyboard operation, live updates announced politely and not animated under reduced motion, amounts with their currency code, refusals explained in words, in English, Russian and Hebrew right to left, light and dark, at 1440 by 900 and at 390 by 844 with no horizontal scroll. The end-to-end walk signs in as an operator and opens the timeline of the fixture run of the first outcome with its graph and attempts, then the capacity screen with its pools, quotas and leases; as a workspace admin it reads the usage of the project with its uncertain amounts apart and has a threshold above the platform one refused; and as an observer it sees the same screens without commands. The screens whose sources are delivered in I4 (runs, overview, status and objectives, alerts) or in I7 (fleet board, operated product, rollouts, incidents, on-call, off-boardings) show their sources as unknown and are listed by name. The record holds per screen and role the expected and observed result and the network log, which shows no address other than StudioBff and StudioRealtimeGateway.

Acceptance criteria

AC1 Given each of the 13 screens of the area in model/ux/screens.yaml, when the accessibility check runs in the three locales, both schemes and the two viewports of docs/planes/experience/studio/localization-accessibility.md, then every screen passes the automated WCAG 2.2 AA checks and keyboard operation and has no horizontal scroll at 390 by 844 (NFR-15, FUX-12).
AC2 Given the walk as an operator, a workspace admin and an observer, when it runs on the acceptance estate, then the timeline, usage and capacity screens render from StudioBff/GetScreenModel, the refused threshold goes through StudioBff/ForwardCommand and is explained, and the ten screens of later sources are listed by name with their sources unknown (ADR-0007, FUX-04).
AC3 Given the record and the network log of the walk, when they are read, then the browser called no address other than StudioBff and StudioRealtimeGateway and no screen showed a value its source does not deliver in I3 (FUX-12).
Traces: services StudioBff
Traces: apis StudioBff/GetScreenModel, StudioBff/ForwardCommand
Traces: adrs ADR-0007
Traces: requirements FUX-04, FUX-12
Relates: screens run-list, run-timeline, overview, status-objectives, alerts, usage-cost, capacity, fleet-board, product, rollouts, incidents, on-call, offboardings
Relates: nfrs NFR-15
Depends on AIFACTORY-US-1401, AIFACTORY-US-1402, AIFACTORY-US-1403, AIFACTORY-US-132, feature:studio-bff, feature:component-blazor-shell
Sources model/ux/screens.yaml, docs/planes/experience/studio/localization-accessibility.md, docs/planes/experience/studio/screens/operations-usage.md

Tasks

Task Kind Expected output Repository
AIFACTORY-TASK-5660 End-to-end walk of the Operations and usage area fixture A browser-driven walk in the Studio repository that signs in as an operator, a workspace admin and an observer, opens the timeline of the fixture run, the usage and capacity screens and the ten screens of later sources, attempts the refused threshold, and records per screen and role the rendered state, the source markers, the offered commands and the addresses the browser called. ConnectSoft.AI.SoftwareFactory.Studio
AIFACTORY-TASK-5661 Accessibility and phone-width check of the Operations screens verification Run dotnet test -c Release --filter Category=Accessibility, npx playwright test --project=chromium at 1440 by 900 and npx playwright test --project=phone-390x844, the Playwright project of the Studio at 390 by 844, on the thirteen screens in en, ru and he, light and dark; the results prove AC1 of AIFACTORY-US-1404 against docs/planes/experience/studio/localization-accessibility.md. ConnectSoft.AI.SoftwareFactory.Studio
AIFACTORY-TASK-5662 Verify the walk record of the Operations area verification Run the walk of AIFACTORY-TASK-5660 on the acceptance estate; its record and network log prove AC2 and AC3 of AIFACTORY-US-1404 (StudioBff/GetScreenModel, StudioBff/ForwardCommand) and are kept with the evidence of increment I3. ConnectSoft.AI.SoftwareFactory.Studio