SpecificationsAI Requests, Usage and the Jazz AI Teammate
20 acceptance test sets for MOD-PFM-LLM, one per capability, holding 234 scenarios — 166 of them already bound to an existing test. A scenario names the rules and journey assertions it proves; it never restates them.
| Acceptance test set | Proves capability | Scenarios | Bound | Negative | Levels | Status |
|---|---|---|---|---|---|---|
ATS-PFM-LLM-001 | Run AI requests with consistent attribution and accounting | 22 | 10 | 10 | 15 api · 5 unit · 2 browser | Proposed |
ATS-PFM-LLM-002 | Bound AI conversations and handle recoverable failures | 14 | 10 | 6 | 11 unit · 3 api | Proposed |
ATS-PFM-LLM-003 | Keep one account for a multi-request AI task | 11 | 9 | 3 | 11 api | Proposed |
ATS-PFM-LLM-004 | Retain redacted AI evidence and remove it safely | 7 | 5 | 2 | 6 api · 1 unit | Proposed |
| Acceptance test set | Proves capability | Scenarios | Bound | Negative | Levels | Status |
|---|---|---|---|---|---|---|
ATS-PFM-LLM-005 | Publish AI operation definitions and historical prices | 17 | 13 | 6 | 5 unit · 12 api | Proposed |
ATS-PFM-LLM-006 | Check AI behavior with scripted models and deliberate live runs | 6 | 5 | 4 | 6 unit | Proposed |
| Acceptance test set | Proves capability | Scenarios | Bound | Negative | Levels | Status |
|---|---|---|---|---|---|---|
ATS-PFM-LLM-007 | Admit AI work against the organization’s monthly allowance | 21 | 14 | 3 | 20 api · 1 unit | Proposed |
ATS-PFM-LLM-008 | Notify administrators at allowance thresholds | 14 | 9 | 3 | 13 api · 1 unit | Proposed |
ATS-PFM-LLM-009 | Recover interrupted requests and identify accounting drift | 6 | 5 | 0 | 6 api | Proposed |
| Acceptance test set | Proves capability | Scenarios | Bound | Negative | Levels | Status |
|---|---|---|---|---|---|---|
ATS-PFM-LLM-010 | Monitor AI usage across organizations | 12 | 11 | 1 | 8 api · 2 unit · 2 browser | Proposed |
ATS-PFM-LLM-011 | Inspect an organization’s AI account and request history | 18 | 14 | 6 | 12 api · 4 browser · 2 unit | Proposed |
ATS-PFM-LLM-012 | Explain an AI refusal or unsuccessful conversation | 9 | 7 | 3 | 7 api · 1 browser · 1 unit | Proposed |
| Acceptance test set | Proves capability | Scenarios | Bound | Negative | Levels | Status |
|---|---|---|---|---|---|---|
ATS-PFM-LLM-013 | Set an organization’s AI access and recurring limits | 18 | 16 | 9 | 15 api · 3 unit | Proposed |
ATS-PFM-LLM-014 | Provide temporary AI capacity and revoke it | 15 | 14 | 8 | 14 api · 1 unit | Proposed |
ATS-PFM-LLM-015 | Correct an organization’s AI charges | 8 | 7 | 6 | 7 api · 1 unit | Proposed |
ATS-PFM-LLM-016 | Inspect captured AI content with a live access grant | 6 | 6 | 3 | 4 api · 2 unit | Proposed |
| Acceptance test set | Proves capability | Scenarios | Bound | Negative | Levels | Status |
|---|---|---|---|---|---|---|
ATS-PFM-LLM-017 | Show administrators their organization’s AI allowance | 13 | 9 | 1 | 10 api · 1 browser · 2 unit | Proposed |
ATS-PFM-LLM-018 | Give people understandable AI refusal messages | 3 | 2 | 1 | 2 unit · 1 browser | Proposed |
| Acceptance test set | Proves capability | Scenarios | Bound | Negative | Levels | Status |
|---|---|---|---|---|---|---|
ATS-PFM-LLM-019 | Use Jazz AI conversations and project folders | 8 | 0 | 3 | 6 browser · 2 api | Proposed |
ATS-PFM-LLM-020 | Refine queries and directory filters with Jazz AI | 6 | 0 | 2 | 5 browser · 1 api | Proposed |