---
id: FAS-PFM-LLM-04
title: Rhythm HQ AI Usage Oversight acceptance
feature: FEAT-PFM-LLM-04
product_area: platform
module: llm
revision: 1
status: proposed
product_intent: proposed
source_sha:
  services-platform: ad3f6a6582843008ed0f5a17b131878d15829af1
  react-console: baf6dae382d6f41dce976c7f848667ccc37e076f
  rhythm-v2-docs: 9b95399181c8ec6670ce3d12d700e59643668692
capabilities:
  - CAP-PFM-LLM-010
  - CAP-PFM-LLM-011
  - CAP-PFM-LLM-012
scenarios:
  - id: FAS-PFM-LLM-04-01
    title: Support follows usage to a refusal explanation
    kind: behaviour
    exercises:
      - CAP-PFM-LLM-010
      - CAP-PFM-LLM-011
      - CAP-PFM-LLM-012
    relies_on:
      - ATS-PFM-LLM-010-01
      - ATS-PFM-LLM-011-02
      - ATS-PFM-LLM-012-07
    preconditions: A staff reader can access an organization with a refused request.
    steps: 'Navigate from reporting to its account, request directory, detail and explanation.'
    expected: >-
      The selected organization and request stay consistent, with retained decision figures separated from
      current affordability.
    negative: false
    data_needs: >-
      An isolated workspace with stated usage, callers, clock instants and a separate workspace for scope
      checks.
    bound_to: []
    provenance: proposed
    source:
      - >-
        rhythm-v2-docs:specs/platform/llm/planning/capabilities/CAP-PFM-LLM-010-see-ai-usage-and-spend-across-every-organization.md
      - >-
        rhythm-v2-docs:specs/platform/llm/planning/capabilities/CAP-PFM-LLM-011-inspect-one-organization-s-ai-usage-and-requests.md
      - >-
        rhythm-v2-docs:specs/platform/llm/planning/capabilities/CAP-PFM-LLM-012-explain-why-an-ai-request-was-refused-or-failed.md
  - id: FAS-PFM-LLM-04-02
    title: Added capacity changes present explanation
    kind: interaction
    exercises:
      - CAP-PFM-LLM-010
      - CAP-PFM-LLM-011
      - CAP-PFM-LLM-012
    relies_on:
      - ATS-PFM-LLM-010-03
      - ATS-PFM-LLM-012-04
    preconditions: A stored refusal precedes a grant that restores live capacity.
    steps: Read HQ live status and explain the same request before and after the grant.
    expected: >-
      Present status and affordability improve while the original refusal arithmetic remains unchanged and
      explanation takes no hold.
    negative: false
    data_needs: >-
      An isolated workspace with stated usage, callers, clock instants and a separate workspace for scope
      checks.
    bound_to: []
    provenance: proposed
    source:
      - >-
        rhythm-v2-docs:specs/platform/llm/planning/capabilities/CAP-PFM-LLM-010-see-ai-usage-and-spend-across-every-organization.md
      - >-
        rhythm-v2-docs:specs/platform/llm/planning/capabilities/CAP-PFM-LLM-011-inspect-one-organization-s-ai-usage-and-requests.md
      - >-
        rhythm-v2-docs:specs/platform/llm/planning/capabilities/CAP-PFM-LLM-012-explain-why-an-ai-request-was-refused-or-failed.md
  - id: FAS-PFM-LLM-04-03
    title: Late settlement updates history and task inspection
    kind: lifecycle
    exercises:
      - CAP-PFM-LLM-010
      - CAP-PFM-LLM-011
      - CAP-PFM-LLM-012
    relies_on:
      - ATS-PFM-LLM-010-06
      - ATS-PFM-LLM-011-03
    preconditions: A request closes after its start day and its enclosing task has already closed.
    steps: Rebuild the affected day and inspect request and task histories.
    expected: >-
      History uses request start day and retained settlement figures; healthy requests remain ineligible for
      refusal explanation.
    negative: false
    data_needs: >-
      An isolated workspace with stated usage, callers, clock instants and a separate workspace for scope
      checks.
    bound_to: []
    provenance: proposed
    source:
      - >-
        rhythm-v2-docs:specs/platform/llm/planning/capabilities/CAP-PFM-LLM-010-see-ai-usage-and-spend-across-every-organization.md
      - >-
        rhythm-v2-docs:specs/platform/llm/planning/capabilities/CAP-PFM-LLM-011-inspect-one-organization-s-ai-usage-and-requests.md
      - >-
        rhythm-v2-docs:specs/platform/llm/planning/capabilities/CAP-PFM-LLM-012-explain-why-an-ai-request-was-refused-or-failed.md
  - id: FAS-PFM-LLM-04-04
    title: Customer keys cannot open staff investigations
    kind: permission
    exercises:
      - CAP-PFM-LLM-010
      - CAP-PFM-LLM-011
      - CAP-PFM-LLM-012
    relies_on:
      - ATS-PFM-LLM-010-01
      - ATS-PFM-LLM-011-01
      - ATS-PFM-LLM-012-01
    preconditions: A customer administrator holds broad customer authority.
    steps: 'Attempt the system report, own organization’s HQ account and its explanation.'
    expected: All staff-only surfaces refuse the customer; no diagnostic content is exposed.
    negative: true
    data_needs: >-
      An isolated workspace with stated usage, callers, clock instants and a separate workspace for scope
      checks.
    bound_to:
      - TST-PFM-LLM-009
      - TST-PFM-LLM-006
    provenance: test
    source:
      - >-
        services-platform:modules/llm/tests/e2e/llm-usage.e2e.spec.js#refuses a customer’s administrator,
        whatever they hold at their customer’s scope
      - >-
        services-platform:modules/llm/tests/e2e/llm-usage.e2e.spec.js#refuses a customer’s administrator a
        tenant’s usage even in their own tenant
      - >-
        services-platform:modules/llm/tests/e2e/llm-adjustments-and-organization.e2e.spec.js#is refused 403
        for a customer’s administrator, because the route is system
  - id: FAS-PFM-LLM-04-05
    title: An unavailable section remains distinguishable
    kind: regression
    exercises:
      - CAP-PFM-LLM-010
      - CAP-PFM-LLM-011
      - CAP-PFM-LLM-012
    relies_on:
      - ATS-PFM-LLM-011-03
      - ATS-PFM-LLM-012-06
    preconditions: One account subsection is unreadable while other reads remain available.
    steps: Read the overview and request a missing explanation.
    expected: >-
      Available sections remain usable, the failed section reports unavailable and the missing request remains
      not found.
    negative: true
    data_needs: >-
      An isolated workspace with stated usage, callers, clock instants and a separate workspace for scope
      checks.
    bound_to: []
    provenance: proposed
    source:
      - >-
        rhythm-v2-docs:specs/platform/llm/planning/capabilities/CAP-PFM-LLM-010-see-ai-usage-and-spend-across-every-organization.md
      - >-
        rhythm-v2-docs:specs/platform/llm/planning/capabilities/CAP-PFM-LLM-011-inspect-one-organization-s-ai-usage-and-requests.md
      - >-
        rhythm-v2-docs:specs/platform/llm/planning/capabilities/CAP-PFM-LLM-012-explain-why-an-ai-request-was-refused-or-failed.md
evidence:
  - artifact: CAP-PFM-LLM-010
    provenance: documented
    source: >-
      specs/platform/llm/planning/capabilities/CAP-PFM-LLM-010-see-ai-usage-and-spend-across-every-organization.md
  - artifact: CAP-PFM-LLM-011
    provenance: documented
    source: >-
      specs/platform/llm/planning/capabilities/CAP-PFM-LLM-011-inspect-one-organization-s-ai-usage-and-requests.md
  - artifact: CAP-PFM-LLM-012
    provenance: documented
    source: >-
      specs/platform/llm/planning/capabilities/CAP-PFM-LLM-012-explain-why-an-ai-request-was-refused-or-failed.md
  - provenance: documented
    source: 'rhythm-v2-docs:specs/CONVENTIONS.md#5.3 Feature Acceptance Suite'
execution_gap: FND-PFM-LLM-081
---

# Rhythm HQ AI Usage Oversight acceptance

## What the Feature Must Do as a Whole

A staff reader can access an organization with a refused request. Navigate from reporting to its account, request directory, detail and explanation. The suite compares the feature outcome with its capability prerequisites and keeps partial proof visible.

## Interactions Between Capabilities

A stored refusal precedes a grant that restores live capacity. Read HQ live status and explain the same request before and after the grant. The interaction scenario checks what changes across that shared boundary. Lifecycle, permission and regression scenarios preserve the wider feature obligations.

## Regression Record

| Revision | Change | Scenario | Why |
|---|---|---|---|
| 1 | initial | all | Establish cumulative feature acceptance from the pinned capability and design baseline. |
