---
generated_by: tools/trace-sync.mjs
module: MOD-PFM-LLM
generated_at: 2026-10-23
---

# AI Requests, Usage and the Jazz AI Teammate: Traceability

Generated from the test side of the corpus (CONVENTIONS §15). Do not edit: run `node tools/trace-sync.mjs platform/llm`.

## Rules and Their Proof

| Capability | Rule | Must prove | Proving scenarios | Executable suites (latest result) |
|---|---|---|---|---|
| CAP-PFM-LLM-001 | `BR-CAP-PFM-LLM-001-01` An undeclared operation or a routing without a published model price is rejected before creating a request or reserving credits. | yes | ATS-PFM-LLM-001-01, ATS-PFM-LLM-001-14 |  |
| CAP-PFM-LLM-001 | `BR-CAP-PFM-LLM-001-02` Platform creates the request record before reserving credits or contacting the model, giving later recovery a record to inspect. | yes | ATS-PFM-LLM-001-02, ATS-PFM-LLM-001-13 |  |
| CAP-PFM-LLM-001 | `BR-CAP-PFM-LLM-001-03` Settlement uses reported usage when available, charges the reservation estimate when an attempt succeeds without any reported usage, and releases the hold when no usage is spent; interruptions can leave unresolved accounting. | yes | ATS-PFM-LLM-001-03, ATS-PFM-LLM-001-12, ATS-PFM-LLM-001-15 | TST-PFM-LLM-018 (not-run), TST-PFM-LLM-020 (not-run) |
| CAP-PFM-LLM-001 | `BR-CAP-PFM-LLM-001-04` Normal state changes compare the current state with the expected state, while settlement can close any nonterminal request; settled and refused requests reject a second settlement. | yes | ATS-PFM-LLM-001-04, ATS-PFM-LLM-001-21 | TST-PFM-LLM-061 (not-run) |
| CAP-PFM-LLM-001 | `BR-CAP-PFM-LLM-001-05` Request attribution uses the caller context and published application and feature routing; a service’s asserted beneficiary is retained without independent verification. | yes | ATS-PFM-LLM-001-05 |  |
| CAP-PFM-LLM-001 | `BR-CAP-PFM-LLM-001-06` Customer-relevant refusals use fixed codes and statuses: allowance and concurrency refusals are rate limits, disabled operations are conflicts, provider timeout, unavailability and refusal indicate temporary service failure, and depleted Rhythm provider credit is a server error. | yes | ATS-PFM-LLM-001-06, ATS-PFM-LLM-001-16, ATS-PFM-LLM-001-17, ATS-PFM-LLM-001-18, ATS-PFM-LLM-001-20 | TST-PFM-LLM-020 (not-run), TST-PFM-LLM-075 (not-run) |
| CAP-PFM-LLM-001 | `BR-CAP-PFM-LLM-001-07` Conversation budget endings return as outcomes for the consuming feature to interpret rather than throwing a metering error. | yes | ATS-PFM-LLM-001-07, ATS-PFM-LLM-001-11 | TST-PFM-LLM-020 (not-run) |
| CAP-PFM-LLM-001 | `BR-CAP-PFM-LLM-001-08` Provider failure messages are replaced with Rhythm wording that identifies the operation and request, allowing staff to locate the usage record. | yes | ATS-PFM-LLM-001-08 |  |
| CAP-PFM-LLM-001 | `BR-CAP-PFM-LLM-001-09` Platform’s provider-boundary guard rejects extra SDK importers, direct provider-host calls and prohibited provider SDK imports during verification. | yes | ATS-PFM-LLM-001-09, ATS-PFM-LLM-001-19 | TST-PFM-LLM-086 (not-run) |
| CAP-PFM-LLM-001 | `BR-CAP-PFM-LLM-001-10` Absent model credentials produce a configuration refusal that is distinct from allowance and provider refusals. | yes | ATS-PFM-LLM-001-10 | TST-PFM-LLM-063 (not-run) |
| CAP-PFM-LLM-001 | `BR-CAP-PFM-LLM-001-11` The consuming feature receives and validates the model’s final answer; the runtime does not apply that answer as domain state. | no |  |  |
| CAP-PFM-LLM-001 | `BR-CAP-PFM-LLM-001-12` Depleted Rhythm provider credit has an internal classification but the shared response mapper returns a generic server error without that code. | no |  |  |
| CAP-PFM-LLM-001 | `BR-CAP-PFM-LLM-001-13` Allowance refusals and provider refusals contribute to separate operational measures. | no | ATS-PFM-LLM-001-22 | TST-PFM-LLM-076 (not-run) |
| CAP-PFM-LLM-002 | `BR-CAP-PFM-LLM-002-01` A conversation without an accepted answer ends when it has used the permitted turns, after sending that many model requests. | yes | ATS-PFM-LLM-001-11, ATS-PFM-LLM-002-01, ATS-PFM-LLM-002-12 | TST-PFM-LLM-020 (not-run), TST-PFM-LLM-068 (not-run) |
| CAP-PFM-LLM-002 | `BR-CAP-PFM-LLM-002-02` Without an accepted final answer, an oversized tool batch ends the conversation before any handler in that batch runs. | yes | ATS-PFM-LLM-002-02 | TST-PFM-LLM-068 (not-run) |
| CAP-PFM-LLM-002 | `BR-CAP-PFM-LLM-002-03` After checking final answers and terminal responses, a tool-requesting turn stops further work when accumulated rounded turn costs are greater than the hold; spending exactly the hold continues and the crossing turn remains billable. | yes | ATS-PFM-LLM-002-03 | TST-PFM-LLM-068 (not-run) |
| CAP-PFM-LLM-002 | `BR-CAP-PFM-LLM-002-04` Before starting another turn, the loop refuses to send when no total time remains and caps the provider attempt timeout at the remaining conversation time. | yes | ATS-PFM-LLM-002-04 | TST-PFM-LLM-068 (not-run) |
| CAP-PFM-LLM-002 | `BR-CAP-PFM-LLM-002-05` Timeouts, ordinary rate limits, overload, server faults, broken connections and aborts can retry with increasing waits and a default of three attempts; malformed or unauthorized provider requests do not retry. | yes | ATS-PFM-LLM-002-05, ATS-PFM-LLM-002-14 | TST-PFM-LLM-065 (not-run), TST-PFM-LLM-067 (not-run) |
| CAP-PFM-LLM-002 | `BR-CAP-PFM-LLM-002-06` A provider rate limit that identifies a depleted Rhythm balance is classified as credit exhaustion and does not retry. | yes | ATS-PFM-LLM-002-06 | TST-PFM-LLM-067 (not-run) |
| CAP-PFM-LLM-002 | `BR-CAP-PFM-LLM-002-07` Attempts record their turn, retry, usage and timing and receive individual prices; after 64 retained attempts, additional attempts increase the count and mark the record incomplete without adding rows. | yes | ATS-PFM-LLM-002-07, ATS-PFM-LLM-002-13 | TST-PFM-LLM-018 (not-run), TST-PFM-LLM-020 (not-run) |
| CAP-PFM-LLM-002 | `BR-CAP-PFM-LLM-002-08` For a nonterminal turn within budget, the model receives one combined tool-results message and every answered call consumes tool budget. | no |  |  |
| CAP-PFM-LLM-002 | `BR-CAP-PFM-LLM-002-09` A thrown tool-handler error becomes an error result for the model, so handler messages must be suitable to disclose to that model. | no |  |  |
| CAP-PFM-LLM-002 | `BR-CAP-PFM-LLM-002-10` Rejected final answers return the feature’s feedback as tool errors and consume tool budget; a failing answer judge also becomes a rejection and further turns use the existing budgets. | yes | ATS-PFM-LLM-002-08 | TST-PFM-LLM-068 (not-run) |
| CAP-PFM-LLM-002 | `BR-CAP-PFM-LLM-002-11` A provider refusal ends without retry, while an output-token cutoff produces an incomplete outcome rather than an accepted answer. | no |  |  |
| CAP-PFM-LLM-002 | `BR-CAP-PFM-LLM-002-12` Stable instructions, tool definitions and explicitly stable content can receive cache markers, with excess markers dropped so the provider receives at most four. | yes | ATS-PFM-LLM-002-09 | TST-PFM-LLM-064 (not-run) |
| CAP-PFM-LLM-002 | `BR-CAP-PFM-LLM-002-13` Failures do not trigger a substitute model or provider, and consuming features cannot replace published routing or limits. | no |  |  |
| CAP-PFM-LLM-002 | `BR-CAP-PFM-LLM-002-14` An accepted final answer completes before the later tool and credit checks and does not increase the answered-tool count. | yes | ATS-PFM-LLM-002-10 |  |
| CAP-PFM-LLM-002 | `BR-CAP-PFM-LLM-002-15` A requested tool without a supplied handler produces a recoverable error result, allowing later turns within the remaining budgets. | yes | ATS-PFM-LLM-002-11 |  |
| CAP-PFM-LLM-003 | `BR-CAP-PFM-LLM-003-01` The execution starts running before its first request and closes completed or failed with its completion time and elapsed duration. | yes | ATS-PFM-LLM-003-01, ATS-PFM-LLM-003-08 | TST-PFM-LLM-018 (not-run), TST-PFM-LLM-045 (not-run) |
| CAP-PFM-LLM-003 | `BR-CAP-PFM-LLM-003-02` Only the exact context supplied by the execution wrapper associates a request with it; outside or derived contexts carry no execution association. | yes | ATS-PFM-LLM-003-02 | TST-PFM-LLM-018 (not-run) |
| CAP-PFM-LLM-003 | `BR-CAP-PFM-LLM-003-03` Opening a nested execution is rejected and makes the enclosing execution fail. | yes | ATS-PFM-LLM-003-03 | TST-PFM-LLM-018 (not-run) |
| CAP-PFM-LLM-003 | `BR-CAP-PFM-LLM-003-04` An unknown execution key is rejected before creating a record, and duplicate keys, unknown applications or empty features invalidate publication. | yes | ATS-PFM-LLM-003-04 | TST-PFM-LLM-018 (not-run), TST-PFM-LLM-057 (not-run) |
| CAP-PFM-LLM-003 | `BR-CAP-PFM-LLM-003-05` Request settlement atomically adds one request and its credits and cents to the execution; estimated settlement adds no cents, released settlement adds no cost, refusals add no settlement count, and late settlements still contribute. | yes | ATS-PFM-LLM-001-12, ATS-PFM-LLM-003-05, ATS-PFM-LLM-003-09, ATS-PFM-LLM-003-11 | TST-PFM-LLM-018 (not-run), TST-PFM-LLM-032 (not-run) |
| CAP-PFM-LLM-003 | `BR-CAP-PFM-LLM-003-06` Reusing an organization’s execution idempotency key returns the earlier running or closed execution as a replay without invoking the task again. | yes | ATS-PFM-LLM-003-06, ATS-PFM-LLM-003-10 | TST-PFM-LLM-012 (not-run), TST-PFM-LLM-018 (not-run) |
| CAP-PFM-LLM-003 | `BR-CAP-PFM-LLM-003-07` The execution retains its declared identity, application, feature, caller attribution and any subject record. | no |  |  |
| CAP-PFM-LLM-003 | `BR-CAP-PFM-LLM-003-08` A settlement log identifies its enclosing execution when one exists and leaves that association absent otherwise. | no |  |  |
| CAP-PFM-LLM-003 | `BR-CAP-PFM-LLM-003-09` A task failure remains the caller’s error even if closing its execution also fails; that close failure is logged. After successful task work, a close failure is returned to the caller. | yes | ATS-PFM-LLM-003-07 |  |
| CAP-PFM-LLM-004 | `BR-CAP-PFM-LLM-004-01` The runtime attempts redacted request capture before dispatch and response capture after a loop result under a ninety-day retention policy; allowance refusals have no bodies and capture may be skipped or fail. | yes | ATS-PFM-LLM-001-13, ATS-PFM-LLM-004-01, ATS-PFM-LLM-004-06 | TST-PFM-LLM-020 (not-run) |
| CAP-PFM-LLM-004 | `BR-CAP-PFM-LLM-004-02` Redaction replaces secret-named field values and recognizable credentials wherever they appear while retaining ordinary prompt text. | yes | ATS-PFM-LLM-004-02 | TST-PFM-LLM-020 (not-run), TST-PFM-LLM-072 (not-run) |
| CAP-PFM-LLM-004 | `BR-CAP-PFM-LLM-004-03` Evidence bodies are stored separately from request and attempt records, whose contents consist of references, counts and operational details. | yes | ATS-PFM-LLM-004-03 | TST-PFM-LLM-034 (not-run) |
| CAP-PFM-LLM-004 | `BR-CAP-PFM-LLM-004-04` Capture status distinguishes pending, captured, skipped, failed and purged, and database constraints require a storage reference for a captured body. | no | ATS-PFM-LLM-004-07 | TST-PFM-LLM-014 (not-run) |
| CAP-PFM-LLM-004 | `BR-CAP-PFM-LLM-004-05` Organization deletion removes its AI records in dependency order and all versions of its evidence without changing neighboring organizations’ data. | yes | ATS-PFM-LLM-004-04 | TST-PFM-LLM-027 (not-run) |
| CAP-PFM-LLM-004 | `BR-CAP-PFM-LLM-004-06` Deletion removes evidence after the last dependent AI record, preserving references until those records are gone. | yes | ATS-PFM-LLM-004-04 | TST-PFM-LLM-027 (not-run) |
| CAP-PFM-LLM-004 | `BR-CAP-PFM-LLM-004-07` AI provision creates no allowance or request records for a new organization; the first request creates them when needed. | no |  |  |
| CAP-PFM-LLM-004 | `BR-CAP-PFM-LLM-004-08` Runtime functions may write evidence and use model credentials, the HQ content-reading function may read evidence without model credentials, and reconciliation holds neither authority. | no |  |  |
| CAP-PFM-LLM-004 | `BR-CAP-PFM-LLM-004-09` Capture returns skipped when no evidence store is configured and failed when redaction, serialization or writing fails, allowing request processing to continue. A memory evidence store cannot be constructed in deployed configuration. | yes | ATS-PFM-LLM-004-05 |  |
| CAP-PFM-LLM-005 | `BR-CAP-PFM-LLM-005-01` The reservation is the rounded per-turn estimate multiplied by permitted turns and limited to the operation’s declared credit ceiling. | yes | ATS-PFM-LLM-005-01, ATS-PFM-LLM-005-10 | TST-PFM-LLM-058 (not-run) |
| CAP-PFM-LLM-005 | `BR-CAP-PFM-LLM-005-02` Publication rejects the whole catalog for unpriced routing, inconsistent timeouts, duplicate operations or rates, unknown applications, empty features, zero estimates or invalid prices, writing no versions. | yes | ATS-PFM-LLM-005-02, ATS-PFM-LLM-005-12 | TST-PFM-LLM-023 (not-run), TST-PFM-LLM-060 (not-run) |
| CAP-PFM-LLM-005 | `BR-CAP-PFM-LLM-005-03` An unchanged declaration adds no version, a changed declaration appends one, and reverting appends another version with the earlier content digest. | yes | ATS-PFM-LLM-002-12, ATS-PFM-LLM-005-03, ATS-PFM-LLM-005-14, ATS-PFM-LLM-005-16 | TST-PFM-LLM-023 (not-run), TST-PFM-LLM-042 (not-run), TST-PFM-LLM-056 (not-run) |
| CAP-PFM-LLM-005 | `BR-CAP-PFM-LLM-005-04` Published routing and rate versions reject updates and deletions by both application and publishing roles. | yes | ATS-PFM-LLM-005-04, ATS-PFM-LLM-005-15 | TST-PFM-LLM-011 (not-run), TST-PFM-LLM-043 (not-run) |
| CAP-PFM-LLM-005 | `BR-CAP-PFM-LLM-005-05` A request retains its original rate version when prices change; only subsequent requests resolve the newer version. | yes | ATS-PFM-LLM-001-14, ATS-PFM-LLM-005-05, ATS-PFM-LLM-005-11, ATS-PFM-LLM-005-13 | TST-PFM-LLM-040 (not-run), TST-PFM-LLM-041 (not-run) |
| CAP-PFM-LLM-005 | `BR-CAP-PFM-LLM-005-06` Usage costs are summed at per-million rates before one division, with credits rounded upward and cents rounded to nearest. | yes | ATS-PFM-LLM-005-06, ATS-PFM-LLM-005-09 | TST-PFM-LLM-058 (not-run) |
| CAP-PFM-LLM-005 | `BR-CAP-PFM-LLM-005-07` One credit represents one hundredth of a cent, and model input, output, cache-read and cache-write units have declared credit and cent prices per million. | no |  |  |
| CAP-PFM-LLM-005 | `BR-CAP-PFM-LLM-005-08` Unpriced usage dimensions contribute no charge and remain visible under their provider names for later pricing decisions. | yes | ATS-PFM-LLM-005-07, ATS-PFM-LLM-005-17 | TST-PFM-LLM-055 (not-run), TST-PFM-LLM-058 (not-run), TST-PFM-LLM-059 (not-run) |
| CAP-PFM-LLM-005 | `BR-CAP-PFM-LLM-005-09` Engineers choose and deploy model prices; Rhythm does not reconcile those declarations with the provider’s actual bill. | no |  |  |
| CAP-PFM-LLM-005 | `BR-CAP-PFM-LLM-005-10` The HQ catalog and rates show each key’s current version and up to nineteen preceding versions. | no | ATS-PFM-LLM-005-08 | TST-PFM-LLM-009 (not-run) |
| CAP-PFM-LLM-006 | `BR-CAP-PFM-LLM-006-01` Integration and end-to-end test scripts enable stub mode before test collection, preventing those tiers from making real model requests. | yes | ATS-PFM-LLM-006-01, ATS-PFM-LLM-006-05 | TST-PFM-LLM-066 (not-run), TST-PFM-LLM-087 (not-run) |
| CAP-PFM-LLM-006 | `BR-CAP-PFM-LLM-006-02` The live adapter refuses construction in stub mode or without credentials instead of deferring the failure to a provider request. | yes | ATS-PFM-LLM-005-09, ATS-PFM-LLM-006-02 | TST-PFM-LLM-064 (not-run) |
| CAP-PFM-LLM-006 | `BR-CAP-PFM-LLM-006-03` Scripted requests use the declared model’s published prices, so ledger tests exercise the production cost calculation. | no | ATS-PFM-LLM-001-15 |  |
| CAP-PFM-LLM-006 | `BR-CAP-PFM-LLM-006-04` The scripted provider rejects an operation with no script and any requested turn beyond the prepared sequence. | no | ATS-PFM-LLM-006-06 | TST-PFM-LLM-069 (not-run) |
| CAP-PFM-LLM-006 | `BR-CAP-PFM-LLM-006-05` Live smoke and evaluation commands require deliberate enablement, a credential source and a test database without stub mode; missing prerequisites produce a did-not-run result without reading or printing a key. | yes | ATS-PFM-LLM-006-03 |  |
| CAP-PFM-LLM-006 | `BR-CAP-PFM-LLM-006-06` Evaluation compares the records returned by the generated query with expected records, and its exit status describes harness execution rather than model correctness. | no |  |  |
| CAP-PFM-LLM-006 | `BR-CAP-PFM-LLM-006-07` Each deployed environment obtains its model key from the same designated secret name within its own account, and warm functions refresh that secret after the credential cache expires. | yes | ATS-PFM-LLM-006-04 | TST-PFM-LLM-063 (not-run) |
| CAP-PFM-LLM-007 | `BR-CAP-PFM-LLM-007-01` The organization receives one allowance period per local calendar month, created on demand with its configured allowance or the default 100,000 credits. | yes | ATS-PFM-LLM-007-01, ATS-PFM-LLM-007-21 | TST-PFM-LLM-030 (not-run), TST-PFM-LLM-062 (not-run) |
| CAP-PFM-LLM-007 | `BR-CAP-PFM-LLM-007-02` Admission atomically holds the estimate only if charged credits, existing holds and the new hold fit within active allowance capacity, capped at ten million credits; racing requests cannot over-reserve that capacity. | yes | ATS-PFM-LLM-005-10, ATS-PFM-LLM-007-02, ATS-PFM-LLM-007-15, ATS-PFM-LLM-007-20 | TST-PFM-LLM-019 (not-run), TST-PFM-LLM-025 (not-run), TST-PFM-LLM-030 (not-run), TST-PFM-LLM-031 (not-run) |
| CAP-PFM-LLM-007 | `BR-CAP-PFM-LLM-007-03` An unaffordable request creates an allowance-refusal record with its needed and remaining credits, but sends nothing to the model and reserves and charges nothing. | yes | ATS-PFM-LLM-001-16, ATS-PFM-LLM-007-03 | TST-PFM-LLM-020 (not-run) |
| CAP-PFM-LLM-007 | `BR-CAP-PFM-LLM-007-04` The concurrency default is eight requests unless organization limits specify another cap; an excess request gets a two-second retry hint, settlement or release returns its slot, and no daily cap applies. | yes | ATS-PFM-LLM-007-04 | TST-PFM-LLM-019 (not-run) |
| CAP-PFM-LLM-007 | `BR-CAP-PFM-LLM-007-05` When allowance and concurrency would both reject admission, the refusal reports the allowance. | yes | ATS-PFM-LLM-007-04 | TST-PFM-LLM-019 (not-run) |
| CAP-PFM-LLM-007 | `BR-CAP-PFM-LLM-007-06` Capacity grants add together and apply from their start inclusive until their end exclusive or revocation, evaluated directly at request time without an activation or expiry job. | yes | ATS-PFM-LLM-007-05, ATS-PFM-LLM-007-17 | TST-PFM-LLM-019 (not-run) |
| CAP-PFM-LLM-007 | `BR-CAP-PFM-LLM-007-07` Already admitted work is charged its measured usage even if the period exceeds its allowance. | yes | ATS-PFM-LLM-007-06 | TST-PFM-LLM-005 (not-run) |
| CAP-PFM-LLM-007 | `BR-CAP-PFM-LLM-007-08` Lower limits preserve running requests and their holds; later admission fails while charged and held credits exceed the reduced allowance, and settlement cannot drive holds below zero. | yes | ATS-PFM-LLM-007-07, ATS-PFM-LLM-007-16 | TST-PFM-LLM-019 (not-run) |
| CAP-PFM-LLM-007 | `BR-CAP-PFM-LLM-007-09` Charged usage includes request settlements and signed corrections, while available credits subtract both charged usage and outstanding holds from the allowance. | yes | ATS-PFM-LLM-007-08, ATS-PFM-LLM-007-11, ATS-PFM-LLM-007-12, ATS-PFM-LLM-007-13, ATS-PFM-LLM-007-14, ATS-PFM-LLM-007-18, ATS-PFM-LLM-007-19 | TST-PFM-LLM-046 (not-run) |
| CAP-PFM-LLM-007 | `BR-CAP-PFM-LLM-007-10` A refusal retains its decision-time allowance, committed and needed credits, active capacity, limits version, concurrency and switches once, without prompt or person content. | yes | ATS-PFM-LLM-007-09 | TST-PFM-LLM-018 (not-run), TST-PFM-LLM-078 (not-run) |
| CAP-PFM-LLM-007 | `BR-CAP-PFM-LLM-007-11` The allowance covers the organization as a whole and has no departmental or cost-centre allocations. | no |  |  |
| CAP-PFM-LLM-008 | `BR-CAP-PFM-LLM-008-01` Charged usage reaching 80% or 100% records one crossing for that threshold and month despite concurrent settlements; held credits do not create crossings. | yes | ATS-PFM-LLM-007-11, ATS-PFM-LLM-008-01 | TST-PFM-LLM-030 (not-run) |
| CAP-PFM-LLM-008 | `BR-CAP-PFM-LLM-008-02` Threshold evaluation uses whole-number arithmetic over charged credits and allowance including currently active additional capacity. | yes | ATS-PFM-LLM-008-02 |  |
| CAP-PFM-LLM-008 | `BR-CAP-PFM-LLM-008-03` Each active holder of the organization’s built-in Administrator role receives one notice, excluding suspended holders, other staff and administrators of other organizations. | yes | ATS-PFM-LLM-008-03 | TST-PFM-LLM-005 (not-run), TST-PFM-LLM-028 (not-run) |
| CAP-PFM-LLM-008 | `BR-CAP-PFM-LLM-008-04` The warning provides used, allowed and remaining credits, the local reset date, the 100% consequence and support contact; the pause notice names the reset date and shares the customer pause wording. | yes | ATS-PFM-LLM-008-04, ATS-PFM-LLM-008-13, ATS-PFM-LLM-008-14 | TST-PFM-LLM-005 (not-run), TST-PFM-LLM-028 (not-run), TST-PFM-LLM-080 (not-run) |
| CAP-PFM-LLM-008 | `BR-CAP-PFM-LLM-008-05` An unsent crossing can be claimed only below three attempts and outside a ten-minute live claim, giving racing senders one delivery owner. | yes | ATS-PFM-LLM-008-05 | TST-PFM-LLM-028 (not-run) |
| CAP-PFM-LLM-008 | `BR-CAP-PFM-LLM-008-06` Settlement sends only after its own transaction commits; an enclosing open transaction leaves the notice pending for reconciliation. | yes | ATS-PFM-LLM-008-06 | TST-PFM-LLM-028 (not-run) |
| CAP-PFM-LLM-008 | `BR-CAP-PFM-LLM-008-07` Zero successful deliveries leave the crossing unstamped for hourly retry up to three attempts; partial success stamps the successful-recipient count, and three failures leave an unresolved crossing. | yes | ATS-PFM-LLM-008-07, ATS-PFM-LLM-008-10, ATS-PFM-LLM-008-11 | TST-PFM-LLM-028 (not-run) |
| CAP-PFM-LLM-008 | `BR-CAP-PFM-LLM-008-08` A crossing without administrator recipients is stamped with zero recipients and logged as a warning without further retries. | yes | ATS-PFM-LLM-008-08 | TST-PFM-LLM-028 (not-run) |
| CAP-PFM-LLM-008 | `BR-CAP-PFM-LLM-008-09` Nonproduction email delivery follows the test-mailbox and allow-list policy, and environment-withheld messages count as handled. | no |  |  |
| CAP-PFM-LLM-008 | `BR-CAP-PFM-LLM-008-10` Corrections and allowance reductions do not trigger threshold notices even when they move usage beyond a threshold. | no | ATS-PFM-LLM-008-12 |  |
| CAP-PFM-LLM-008 | `BR-CAP-PFM-LLM-008-11` Notice failures contribute a metric whose alarm triggers above three failures in fifteen minutes. | no |  |  |
| CAP-PFM-LLM-008 | `BR-CAP-PFM-LLM-008-12` A notice is delivered before its completion stamp is saved. If processing stops between those actions, a later attempt can deliver the same crossing again. | yes | ATS-PFM-LLM-008-09 |  |
| CAP-PFM-LLM-009 | `BR-CAP-PFM-LLM-009-01` Hourly reconciliation visits operational organizations and settles requests older than twenty-four hours at their stored estimate without a money figure, releasing their recorded holds. | yes | ATS-PFM-LLM-007-12, ATS-PFM-LLM-008-10, ATS-PFM-LLM-009-01, ATS-PFM-LLM-009-06 | TST-PFM-LLM-028 (not-run), TST-PFM-LLM-046 (not-run) |
| CAP-PFM-LLM-009 | `BR-CAP-PFM-LLM-009-02` Requests inside the recovery window keep their state and reservations. | yes | ATS-PFM-LLM-009-02 | TST-PFM-LLM-046 (not-run) |
| CAP-PFM-LLM-009 | `BR-CAP-PFM-LLM-009-03` Repeating successful recovery does not add another charge or another past-window alert for the same request. | yes | ATS-PFM-LLM-009-03 | TST-PFM-LLM-046 (not-run) |
| CAP-PFM-LLM-009 | `BR-CAP-PFM-LLM-009-04` Each organization’s recovery pass handles a bounded batch and defers remaining requests to another run. | no |  |  |
| CAP-PFM-LLM-009 | `BR-CAP-PFM-LLM-009-05` Stale running executions become abandoned with the detection time, no duration and their existing settlement totals. | yes | ATS-PFM-LLM-003-08, ATS-PFM-LLM-009-04 | TST-PFM-LLM-045 (not-run) |
| CAP-PFM-LLM-009 | `BR-CAP-PFM-LLM-009-06` The drift check compares recently closed executions from the last seven days with settled requests, logs both figures and identities on disagreement, and changes no totals. | yes | ATS-PFM-LLM-009-05 | TST-PFM-LLM-045 (not-run) |
| CAP-PFM-LLM-009 | `BR-CAP-PFM-LLM-009-07` A request beyond the recovery window contributes a once-only alert measure, and missing reconciliation runs have a separate alarm. | no |  |  |
| CAP-PFM-LLM-010 | `BR-CAP-PFM-LLM-010-01` Only staff with the HQ AI usage read key can read system-wide usage; customer-scoped authority and unrelated HQ keys cannot authorize it. | yes | ATS-PFM-LLM-010-01, ATS-PFM-LLM-010-09 | TST-PFM-LLM-009 (not-run), TST-PFM-LLM-054 (not-run) |
| CAP-PFM-LLM-010 | `BR-CAP-PFM-LLM-010-02` Reporting accepts month to date or thirty or ninety UTC days including today and compares with the immediately preceding equal-length period; unknown windows are rejected. | yes | ATS-PFM-LLM-010-02 | TST-PFM-LLM-009 (not-run), TST-PFM-LLM-049 (not-run), TST-PFM-LLM-082 (not-run) |
| CAP-PFM-LLM-010 | `BR-CAP-PFM-LLM-010-03` Live status is disabled for an inactive organization or disabled AI, otherwise stopped when committed usage reaches allowance, warned from 80%, and within below that; later capacity can improve status. | yes | ATS-PFM-LLM-007-13, ATS-PFM-LLM-010-03 | TST-PFM-LLM-081 (not-run) |
| CAP-PFM-LLM-010 | `BR-CAP-PFM-LLM-010-04` Utilisation divides charged plus held credits by base plus active capacity, rounds to one decimal, may exceed 100% and has no percentage for a zero allowance. | yes | ATS-PFM-LLM-010-04 | TST-PFM-LLM-081 (not-run) |
| CAP-PFM-LLM-010 | `BR-CAP-PFM-LLM-010-05` Trends can group by organization, application, model, provider or operation, retain ten largest series and aggregate the remainder exactly as Other, omitting empty dates. | yes | ATS-PFM-LLM-010-05 | TST-PFM-LLM-049 (not-run) |
| CAP-PFM-LLM-010 | `BR-CAP-PFM-LLM-010-06` Organization rows combine spend, requests, refusals, failures, utilisation, allowance, capacity and status, with cost, utilisation or refusal sorting and stable pagination. | no |  |  |
| CAP-PFM-LLM-010 | `BR-CAP-PFM-LLM-010-07` Breakdowns provide group cost, activity, average latency and failure rate; cross-organization execution leaders identify expensive closed tasks and organizations whose reads failed. | no | ATS-PFM-LLM-010-12 | TST-PFM-LLM-048 (not-run) |
| CAP-PFM-LLM-010 | `BR-CAP-PFM-LLM-010-08` Hourly rebuilding replaces each affected organization’s UTC-day totals, assigns requests to their start day, excludes running work, and retains refused, failed, released and settled classifications with their applicable charges. | yes | ATS-PFM-LLM-005-11, ATS-PFM-LLM-009-06, ATS-PFM-LLM-010-06, ATS-PFM-LLM-010-10 | TST-PFM-LLM-017 (not-run), TST-PFM-LLM-047 (not-run) |
| CAP-PFM-LLM-010 | `BR-CAP-PFM-LLM-010-09` Rollup progress trails the run start by five minutes and stops at the earliest unfinished change on error or exhausted time without moving backward; delay beyond two hours alarms. | yes | ATS-PFM-LLM-010-07, ATS-PFM-LLM-010-11 | TST-PFM-LLM-044 (not-run), TST-PFM-LLM-047 (not-run) |
| CAP-PFM-LLM-010 | `BR-CAP-PFM-LLM-010-10` HQ presents list-price dollars with the equivalent credits alongside. | yes | ATS-PFM-LLM-010-08 | TST-PFM-LLM-003 (not-run) |
| CAP-PFM-LLM-010 | `BR-CAP-PFM-LLM-010-11` An unavailable report section shows its failure state independently of successfully read sections. | no |  |  |
| CAP-PFM-LLM-010 | `BR-CAP-PFM-LLM-010-12` HQ query recipes cover organizational, daily and model spend and organizational refusals; the HQ reporting reader can query daily rollups without gaining access to organization request records. | no |  |  |
| CAP-PFM-LLM-011 | `BR-CAP-PFM-LLM-011-01` The HQ usage read key authorizes organization usage inspection for Rhythm staff, while a customer administrator cannot use that surface even for their own organization. | yes | ATS-PFM-LLM-011-01 | TST-PFM-LLM-009 (not-run) |
| CAP-PFM-LLM-011 | `BR-CAP-PFM-LLM-011-02` Reads first resolve the named organization and scope its records accordingly, returning not found for nonexistent organizations or foreign requests and executions. | yes | ATS-PFM-LLM-010-09, ATS-PFM-LLM-011-02, ATS-PFM-LLM-011-15 | TST-PFM-LLM-009 (not-run), TST-PFM-LLM-033 (not-run), TST-PFM-LLM-050 (not-run) |
| CAP-PFM-LLM-011 | `BR-CAP-PFM-LLM-011-03` The overview independently reports the current allowance position, configured or default limits, active capacity, six recent periods, operation and model totals and twenty recent problem requests. | yes | ATS-PFM-LLM-007-14, ATS-PFM-LLM-011-03, ATS-PFM-LLM-011-10, ATS-PFM-LLM-011-14, ATS-PFM-LLM-011-18 | TST-PFM-LLM-001 (not-run), TST-PFM-LLM-004 (not-run), TST-PFM-LLM-050 (not-run) |
| CAP-PFM-LLM-011 | `BR-CAP-PFM-LLM-011-04` Historical periods evaluate capacity at their last instant and the current period evaluates it now; ended and revoked grants do not count as currently active. | yes | ATS-PFM-LLM-011-04 | TST-PFM-LLM-050 (not-run) |
| CAP-PFM-LLM-011 | `BR-CAP-PFM-LLM-011-05` The request directory is newest-first and supports period or date windows and operation, model, status, execution and caller filters, with at most one hundred rows per page and rejection of invalid filters or cursors. | yes | ATS-PFM-LLM-011-05, ATS-PFM-LLM-011-16, ATS-PFM-LLM-011-17 | TST-PFM-LLM-009 (not-run), TST-PFM-LLM-035 (not-run), TST-PFM-LLM-052 (not-run) |
| CAP-PFM-LLM-011 | `BR-CAP-PFM-LLM-011-06` Usage lists and details withhold captured bodies and storage references, exposing only their capture status. | yes | ATS-PFM-LLM-011-06, ATS-PFM-LLM-011-09, ATS-PFM-LLM-011-13 | TST-PFM-LLM-002 (not-run), TST-PFM-LLM-009 (not-run) |
| CAP-PFM-LLM-011 | `BR-CAP-PFM-LLM-011-07` Request detail identifies the caller, feature, routing, charge, ordered attempts, pricing versions, execution and any retained refusal figures. | no |  |  |
| CAP-PFM-LLM-011 | `BR-CAP-PFM-LLM-011-08` Organization execution lists include running tasks, and each detail pairs the task totals with its associated requests. | no | ATS-PFM-LLM-003-09 |  |
| CAP-PFM-LLM-011 | `BR-CAP-PFM-LLM-011-09` Grant lists order by newest start and derive scheduled, active, expired or revoked state at read time, optionally showing active grants only. | yes | ATS-PFM-LLM-011-07, ATS-PFM-LLM-011-11 | TST-PFM-LLM-009 (not-run), TST-PFM-LLM-029 (not-run) |
| CAP-PFM-LLM-011 | `BR-CAP-PFM-LLM-011-10` Limit reads distinguish defaults from configuration and show ceilings, correction reads preserve signed values, and crossing reads expose delivery time, attempt count and recipient count. | no | ATS-PFM-LLM-008-11 |  |
| CAP-PFM-LLM-011 | `BR-CAP-PFM-LLM-011-11` The AI audit trail filters this organization’s limit, grant, revocation, correction and content-read events and orders them newest-first. | no | ATS-PFM-LLM-011-12 |  |
| CAP-PFM-LLM-011 | `BR-CAP-PFM-LLM-011-12` Support readers receive no usage action menu or write controls, while staff with management authority receive the action menu. | yes | ATS-PFM-LLM-011-08 | TST-PFM-LLM-002 (not-run) |
| CAP-PFM-LLM-012 | `BR-CAP-PFM-LLM-012-01` Only Rhythm staff with the HQ AI usage read key can obtain explanations through the HQ route. | yes | ATS-PFM-LLM-012-01 | TST-PFM-LLM-006 (not-run) |
| CAP-PFM-LLM-012 | `BR-CAP-PFM-LLM-012-02` Refusal explanations preserve the recorded policy and decision-time figures and describe allowance arithmetic, concurrency or the disabled switch without recomputing history. | yes | ATS-PFM-LLM-011-09, ATS-PFM-LLM-012-02 | TST-PFM-LLM-025 (not-run) |
| CAP-PFM-LLM-012 | `BR-CAP-PFM-LLM-012-03` Referenced capacity grants are resolved with their windows, and an unreadable reference remains visible as not found. | yes | ATS-PFM-LLM-012-03 | TST-PFM-LLM-025 (not-run) |
| CAP-PFM-LLM-012 | `BR-CAP-PFM-LLM-012-04` The current-position block describes allowance, committed and remaining credits and grants since the refusal, then evaluates the same estimate under present admission rules without reserving credits. | yes | ATS-PFM-LLM-007-15, ATS-PFM-LLM-012-04, ATS-PFM-LLM-012-08 | TST-PFM-LLM-025 (not-run) |
| CAP-PFM-LLM-012 | `BR-CAP-PFM-LLM-012-05` An older refusal without a snapshot uses the retained request fields and its estimate for the current admission check. | no |  |  |
| CAP-PFM-LLM-012 | `BR-CAP-PFM-LLM-012-06` Failure explanations present classification, ending and ordered attempts with status, error, latency and retry verdict, including a truncation warning when appropriate. | yes | ATS-PFM-LLM-002-13, ATS-PFM-LLM-012-05, ATS-PFM-LLM-012-09 | TST-PFM-LLM-025 (not-run), TST-PFM-LLM-074 (not-run) |
| CAP-PFM-LLM-012 | `BR-CAP-PFM-LLM-012-07` A limit-ending explanation identifies the exhausted limit and the value recorded in the operation’s routing version. | no |  |  |
| CAP-PFM-LLM-012 | `BR-CAP-PFM-LLM-012-08` A healthy settled request is rejected as having nothing to explain, while missing and foreign requests return not found. | yes | ATS-PFM-LLM-012-06 | TST-PFM-LLM-006 (not-run), TST-PFM-LLM-025 (not-run) |
| CAP-PFM-LLM-012 | `BR-CAP-PFM-LLM-012-09` Explanation reads do not open captured content or disclose its storage reference. | no |  |  |
| CAP-PFM-LLM-013 | `BR-CAP-PFM-LLM-013-01` The sensitive HQ AI usage manage key is required to preview or change limits; read-only staff and customer administrators are refused. | yes | ATS-PFM-LLM-013-01 | TST-PFM-LLM-007 (not-run) |
| CAP-PFM-LLM-013 | `BR-CAP-PFM-LLM-013-02` A replacement supplies the AI switch, base allowance or default, disabled operations, concurrency cap or default and note, with omitted optional cap and note reset. | yes | ATS-PFM-LLM-013-02, ATS-PFM-LLM-013-15 | TST-PFM-LLM-026 (not-run), TST-PFM-LLM-037 (not-run) |
| CAP-PFM-LLM-013 | `BR-CAP-PFM-LLM-013-03` Limit validation accepts a default allowance or a whole credit amount from zero to ten million, a default or positive whole concurrency cap, and declared disabled operations stored once in sorted order. Notes are trimmed and limited to one thousand characters; reasons are trimmed to one through five hundred characters. Invalid input leaves state unchanged. | yes | ATS-PFM-LLM-005-12, ATS-PFM-LLM-013-03, ATS-PFM-LLM-013-12, ATS-PFM-LLM-013-17 | TST-PFM-LLM-016 (not-run), TST-PFM-LLM-079 (not-run) |
| CAP-PFM-LLM-013 | `BR-CAP-PFM-LLM-013-04` A changed save requires the inspected limits version, using zero for defaults, and missing or stale versions are rejected. | yes | ATS-PFM-LLM-013-04, ATS-PFM-LLM-013-14, ATS-PFM-LLM-013-16 | TST-PFM-LLM-007 (not-run), TST-PFM-LLM-036 (not-run), TST-PFM-LLM-053 (not-run) |
| CAP-PFM-LLM-013 | `BR-CAP-PFM-LLM-013-05` Values identical to stored limits return unchanged without a write, new version or audit entry, even with an older supplied version. | yes | ATS-PFM-LLM-013-05 | TST-PFM-LLM-007 (not-run), TST-PFM-LLM-026 (not-run) |
| CAP-PFM-LLM-013 | `BR-CAP-PFM-LLM-013-06` A limits save atomically updates the current month’s base allowance and audit event without rewriting charges, holds, grants or past periods; later periods use the new base when created. | yes | ATS-PFM-LLM-007-16, ATS-PFM-LLM-011-10, ATS-PFM-LLM-013-06 | TST-PFM-LLM-007 (not-run), TST-PFM-LLM-019 (not-run), TST-PFM-LLM-026 (not-run) |
| CAP-PFM-LLM-013 | `BR-CAP-PFM-LLM-013-07` The audit event includes before and after values, reason and actor and marks allowance increases, AI re-enablement or operation re-enablement as widening. | yes | ATS-PFM-LLM-013-07 | TST-PFM-LLM-026 (not-run) |
| CAP-PFM-LLM-013 | `BR-CAP-PFM-LLM-013-08` Preview reports proposed allowance, committed and remaining credits, limiting reason and effective-ceiling pressure without writing limits or audit events. | yes | ATS-PFM-LLM-013-08 | TST-PFM-LLM-007 (not-run), TST-PFM-LLM-026 (not-run) |
| CAP-PFM-LLM-013 | `BR-CAP-PFM-LLM-013-09` Disabled AI rejects every operation; an active operation exemption can bypass an operation switch but never the global AI switch, and refusal records the disabled-feature reason without a hold. | yes | ATS-PFM-LLM-001-17, ATS-PFM-LLM-013-09, ATS-PFM-LLM-013-10, ATS-PFM-LLM-013-11, ATS-PFM-LLM-013-13, ATS-PFM-LLM-013-18 | TST-PFM-LLM-021 (not-run), TST-PFM-LLM-026 (not-run), TST-PFM-LLM-077 (not-run) |
| CAP-PFM-LLM-013 | `BR-CAP-PFM-LLM-013-10` The reason belongs in the audit event and is distinct from the explanatory note retained with limits. | no |  |  |
| CAP-PFM-LLM-013 | `BR-CAP-PFM-LLM-013-11` Only usage managers see the console’s Edit Limits control, which remains disabled with an unavailable label. | no |  |  |
| CAP-PFM-LLM-014 | `BR-CAP-PFM-LLM-014-01` Grant, preview and revocation require the sensitive HQ usage manage key, excluding read-only staff and customer administrators; customer administrators also cannot list HQ grants. | yes | ATS-PFM-LLM-014-01 | TST-PFM-LLM-008 (not-run) |
| CAP-PFM-LLM-014 | `BR-CAP-PFM-LLM-014-02` Additional-credit grants accept one to one million whole credits, while an operation exemption names one declared operation and carries no credits; invalid kinds, fields or values are rejected. | yes | ATS-PFM-LLM-013-10, ATS-PFM-LLM-014-02, ATS-PFM-LLM-014-15 | TST-PFM-LLM-008 (not-run), TST-PFM-LLM-026 (not-run), TST-PFM-LLM-029 (not-run), TST-PFM-LLM-071 (not-run) |
| CAP-PFM-LLM-014 | `BR-CAP-PFM-LLM-014-03` The default grant begins now and ends at the next local month boundary; explicit windows must end after their start, end in the future and span no more than ninety-two days. | yes | ATS-PFM-LLM-014-03, ATS-PFM-LLM-014-11, ATS-PFM-LLM-014-12, ATS-PFM-LLM-014-14 | TST-PFM-LLM-010 (not-run), TST-PFM-LLM-015 (not-run), TST-PFM-LLM-029 (not-run), TST-PFM-LLM-039 (not-run) |
| CAP-PFM-LLM-014 | `BR-CAP-PFM-LLM-014-04` Grants require a trimmed reason of one to five hundred characters and accept an optional ticket reference of at most two hundred. | yes | ATS-PFM-LLM-014-04 | TST-PFM-LLM-029 (not-run) |
| CAP-PFM-LLM-014 | `BR-CAP-PFM-LLM-014-05` At every instant in a new grant window, base allowance and overlapping unrevoked capacity must fit under the one-thousand-dollar ceiling; sequential grants are separate and per-organization decisions are serialized. | yes | ATS-PFM-LLM-014-05 | TST-PFM-LLM-029 (not-run) |
| CAP-PFM-LLM-014 | `BR-CAP-PFM-LLM-014-06` The grant and widening audit event commit together with reason, ticket and window, while invalid grants write neither. | yes | ATS-PFM-LLM-014-06 | TST-PFM-LLM-029 (not-run) |
| CAP-PFM-LLM-014 | `BR-CAP-PFM-LLM-014-07` Revocation retains the original grant values and records actor, time and reason with a non-widening audit event; ended or revoked grants reject another revocation, and foreign grants are not found. | yes | ATS-PFM-LLM-007-17, ATS-PFM-LLM-011-11, ATS-PFM-LLM-012-08, ATS-PFM-LLM-014-07, ATS-PFM-LLM-014-13 | TST-PFM-LLM-019 (not-run), TST-PFM-LLM-029 (not-run), TST-PFM-LLM-038 (not-run) |
| CAP-PFM-LLM-014 | `BR-CAP-PFM-LLM-014-08` When requested, a grant that restores a stopped organization emails its administrators after saving; a grant that restores nobody sends no message. | yes | ATS-PFM-LLM-014-08 | TST-PFM-LLM-029 (not-run) |
| CAP-PFM-LLM-014 | `BR-CAP-PFM-LLM-014-09` Preview uses grant validation and projects current and granted capacity, remaining credits, restoration and the peak allowance without creating records. | yes | ATS-PFM-LLM-014-09 | TST-PFM-LLM-029 (not-run) |
| CAP-PFM-LLM-014 | `BR-CAP-PFM-LLM-014-10` Submitting the same grant twice creates and audits two grants, while a repeated revocation is refused. | no |  |  |
| CAP-PFM-LLM-014 | `BR-CAP-PFM-LLM-014-11` A failed capacity notice is logged and counted without retrying or undoing the grant. | no |  |  |
| CAP-PFM-LLM-014 | `BR-CAP-PFM-LLM-014-12` Usage managers see a disabled Grant Override control in the console with an unavailable label. | no |  |  |
| CAP-PFM-LLM-015 | `BR-CAP-PFM-LLM-015-01` Only staff with the sensitive usage manage key can correct charges; support readers and customer administrators are refused without changes. | yes | ATS-PFM-LLM-015-01 | TST-PFM-LLM-006 (not-run) |
| CAP-PFM-LLM-015 | `BR-CAP-PFM-LLM-015-02` Corrections require a month, a nonzero whole credit change within ten million either way and a one-to-five-hundred-character reason, accepting optional whole cents with matching sign and a request reference but no undeclared fields. | yes | ATS-PFM-LLM-015-02, ATS-PFM-LLM-015-08 | TST-PFM-LLM-006 (not-run), TST-PFM-LLM-022 (not-run), TST-PFM-LLM-070 (not-run) |
| CAP-PFM-LLM-015 | `BR-CAP-PFM-LLM-015-03` A correction targets an existing allowance month or creates the current month; other absent periods are rejected. | yes | ATS-PFM-LLM-015-03 | TST-PFM-LLM-022 (not-run) |
| CAP-PFM-LLM-015 | `BR-CAP-PFM-LLM-015-04` Refunds cannot exceed an existing period’s charge, but a current-month refund before any period exists bypasses that check and leaves a recorded negative correction with a zero balance. | yes | ATS-PFM-LLM-015-04 | TST-PFM-LLM-022 (not-run) |
| CAP-PFM-LLM-015 | `BR-CAP-PFM-LLM-015-05` A referenced request must belong to the organization and correction month, with foreign requests treated as absent. | yes | ATS-PFM-LLM-015-05 | TST-PFM-LLM-022 (not-run) |
| CAP-PFM-LLM-015 | `BR-CAP-PFM-LLM-015-06` The correction, period credit and money changes and audit event commit together; refunds are widening, and an audit failure rolls the correction back. | yes | ATS-PFM-LLM-007-18, ATS-PFM-LLM-008-12, ATS-PFM-LLM-011-12, ATS-PFM-LLM-015-06 | TST-PFM-LLM-022 (not-run) |
| CAP-PFM-LLM-015 | `BR-CAP-PFM-LLM-015-07` Repeated submission creates another correction and audit event rather than replaying the first. | no |  |  |
| CAP-PFM-LLM-015 | `BR-CAP-PFM-LLM-015-08` Correction history and the AI audit trail expose the recorded changes beside limits and capacity grants. | no |  |  |
| CAP-PFM-LLM-015 | `BR-CAP-PFM-LLM-015-09` The console’s charge-correction menu item is disabled and visible only with usage management authority. | no |  |  |
| CAP-PFM-LLM-015 | `BR-CAP-PFM-LLM-015-10` Applying a correction floors both period credit and cent totals at zero while retaining the signed correction amounts. | yes | ATS-PFM-LLM-015-07 |  |
| CAP-PFM-LLM-016 | `BR-CAP-PFM-LLM-016-01` Content reads require the sensitive HQ content key plus an active approved or emergency organization access grant inside its window; standing grants do not qualify. | yes | ATS-PFM-LLM-011-13, ATS-PFM-LLM-016-01, ATS-PFM-LLM-016-06 | TST-PFM-LLM-024 (not-run), TST-PFM-LLM-073 (not-run) |
| CAP-PFM-LLM-016 | `BR-CAP-PFM-LLM-016-02` A read after the access window expires is refused and records the missing eligible access. | yes | ATS-PFM-LLM-016-01 | TST-PFM-LLM-024 (not-run) |
| CAP-PFM-LLM-016 | `BR-CAP-PFM-LLM-016-03` The content service audits missing-grant refusals and successful body or status answers before returning content, but route rejections, missing records and storage failures do not get that entry. | yes | ATS-PFM-LLM-016-02 | TST-PFM-LLM-024 (not-run) |
| CAP-PFM-LLM-016 | `BR-CAP-PFM-LLM-016-04` The read returns redacted request or response content or a skipped, failed or purged status, including purged when the object has disappeared, without its storage location. | yes | ATS-PFM-LLM-004-06, ATS-PFM-LLM-016-03 | TST-PFM-LLM-024 (not-run) |
| CAP-PFM-LLM-016 | `BR-CAP-PFM-LLM-016-05` With an eligible grant, missing or foreign requests return not found without auditing; nonexistent organizations return not found before checking access grants. | yes | ATS-PFM-LLM-016-04 | TST-PFM-LLM-024 (not-run) |
| CAP-PFM-LLM-016 | `BR-CAP-PFM-LLM-016-06` HQ reader policies exclude the sensitive content key, and the console grants it only to the administrator among HQ roles. | yes | ATS-PFM-LLM-016-05 | TST-PFM-LLM-054 (not-run) |
| CAP-PFM-LLM-016 | `BR-CAP-PFM-LLM-016-07` Support readers see capture status without read buttons, while administrators see disabled read buttons. | no |  |  |
| CAP-PFM-LLM-017 | `BR-CAP-PFM-LLM-017-01` The customer usage routes admit only console users in the organization with its usage key, excluding unkeyed, anonymous, foreign-customer and HQ-only callers. | yes | ATS-PFM-LLM-017-01 | TST-PFM-LLM-006 (not-run) |
| CAP-PFM-LLM-017 | `BR-CAP-PFM-LLM-017-02` The summary contains the current allowance, charged-plus-held usage, remaining credits, percentage, within or warned or limited status, local reset date and zone, AI switch, explanatory messages, six-month history and usage by feature. | yes | ATS-PFM-LLM-007-19, ATS-PFM-LLM-017-02, ATS-PFM-LLM-017-11 | TST-PFM-LLM-051 (not-run) |
| CAP-PFM-LLM-017 | `BR-CAP-PFM-LLM-017-03` Customer usage summaries expose credits and percentages while withholding money, providers, models, prices, capacity grants and internal limits. | yes | ATS-PFM-LLM-017-03, ATS-PFM-LLM-017-12 | TST-PFM-LLM-006 (not-run), TST-PFM-LLM-083 (not-run) |
| CAP-PFM-LLM-017 | `BR-CAP-PFM-LLM-017-04` The feature list supplies labels, descriptions and availability; disabled AI or all disabled operations takes explanatory precedence over exhausted allowance. | yes | ATS-PFM-LLM-013-11, ATS-PFM-LLM-017-04, ATS-PFM-LLM-017-13 | TST-PFM-LLM-006 (not-run), TST-PFM-LLM-051 (not-run), TST-PFM-LLM-084 (not-run) |
| CAP-PFM-LLM-017 | `BR-CAP-PFM-LLM-017-05` Active exemptions are ignored by the customer feature list, so it can label an operation restricted even though admission permits its request. | yes | ATS-PFM-LLM-017-05 |  |
| CAP-PFM-LLM-017 | `BR-CAP-PFM-LLM-017-06` Declared operations group under their catalog feature and undeclared historical operations group under Other AI features. | yes | ATS-PFM-LLM-017-06 | TST-PFM-LLM-051 (not-run) |
| CAP-PFM-LLM-017 | `BR-CAP-PFM-LLM-017-07` Customer usage reads stay within the organization’s own periods. | yes | ATS-PFM-LLM-017-07 | TST-PFM-LLM-051 (not-run) |
| CAP-PFM-LLM-017 | `BR-CAP-PFM-LLM-017-08` The usage key also supports local request and task query recipes by period, feature, person and problem outcome; reporting excludes content, storage, refusal snapshots, raw usage and pricing versions and uses UTC reporting dates. | yes | ATS-PFM-LLM-017-08 | TST-PFM-LLM-013 (not-run) |
| CAP-PFM-LLM-017 | `BR-CAP-PFM-LLM-017-09` The product should provide the approved dashboard with credits, reset date, feature states, six-month history, pause explanation and request-history navigation without internal money, model or control detail. | yes | ATS-PFM-LLM-017-09 |  |
| CAP-PFM-LLM-018 | `BR-CAP-PFM-LLM-018-01` Allowance-refusal wording explains that the period’s AI allowance is used and that AI resumes next month or sooner through support-provided capacity. | yes | ATS-PFM-LLM-001-18, ATS-PFM-LLM-017-11, ATS-PFM-LLM-018-01 | TST-PFM-LLM-085 (not-run) |
| CAP-PFM-LLM-018 | `BR-CAP-PFM-LLM-018-02` Concurrency-refusal wording asks the person to wait a few seconds because too many organizational AI requests are running. | yes | ATS-PFM-LLM-018-01 | TST-PFM-LLM-085 (not-run) |
| CAP-PFM-LLM-018 | `BR-CAP-PFM-LLM-018-03` Disabled-feature wording says the organization’s AI feature is off and directs the person to Rhythm support for enablement. | yes | ATS-PFM-LLM-018-01 | TST-PFM-LLM-085 (not-run) |
| CAP-PFM-LLM-018 | `BR-CAP-PFM-LLM-018-04` The helper gives provider unavailability and the declared circuit-open refusal the same temporary-service message, although a circuit breaker is not implemented. | yes | ATS-PFM-LLM-018-01 | TST-PFM-LLM-085 (not-run) |
| CAP-PFM-LLM-018 | `BR-CAP-PFM-LLM-018-05` Operator faults such as missing configuration or depleted provider credit receive no specialized customer sentence. | no |  |  |
| CAP-PFM-LLM-018 | `BR-CAP-PFM-LLM-018-06` Customer sentences use organization language and avoid internal money, model, provider and override terminology. | yes | ATS-PFM-LLM-018-02 | TST-PFM-LLM-085 (not-run) |
| CAP-PFM-LLM-018 | `BR-CAP-PFM-LLM-018-07` The pause email shares the customer pause sentence, substituting the organization’s name for the direct customer address. | no | ATS-PFM-LLM-008-13 |  |
| CAP-PFM-LLM-018 | `BR-CAP-PFM-LLM-018-08` Describe With AI displays allowance-refusal wording using a reset day supplied by the refusal or a next-month fallback. | yes | ATS-PFM-LLM-018-03 |  |
| CAP-PFM-LLM-018 | `BR-CAP-PFM-LLM-018-09` Credit figures use thousands separators and date formatting uses day and month with a year where the message supplies one. | no |  |  |
| CAP-PFM-LLM-018 | `BR-CAP-PFM-LLM-018-10` HQ presents refusal reasons as readable labels or sentences instead of raw classifications. | no |  |  |
| CAP-PFM-LLM-019 | `BR-CAP-PFM-LLM-019-01` Signed-in staff can open Jazz AI from its page or header panel without an AI permission check or an organization AI-switch check. | yes | ATS-PFM-LLM-019-01 |  |
| CAP-PFM-LLM-019 | `BR-CAP-PFM-LLM-019-02` General Jazz AI replies use a console-side model call without Platform request accounting, allowance holds, evidence capture or organization AI-switch enforcement. | yes | ATS-PFM-LLM-001-19, ATS-PFM-LLM-019-02 |  |
| CAP-PFM-LLM-019 | `BR-CAP-PFM-LLM-019-03` General replies use thirty previous messages plus the new question and an instruction containing the configured organization and project context, requesting concise answers without invented record facts and directing lists to Queries. | no |  |  |
| CAP-PFM-LLM-019 | `BR-CAP-PFM-LLM-019-04` Reply failure presents a retry message and restores the submitted question in the composer. | no |  |  |
| CAP-PFM-LLM-019 | `BR-CAP-PFM-LLM-019-05` Direct questions allow one to four thousand trimmed characters, appended turns up to eight thousand, project names one to eighty and descriptions up to four hundred; titles allow up to one hundred and twenty and a new title may be empty. | yes | ATS-PFM-LLM-019-03 |  |
| CAP-PFM-LLM-019 | `BR-CAP-PFM-LLM-019-06` A new conversation uses the default title until its first user message supplies a whitespace-normalized sixty-character title, which staff can later rename. | no |  |  |
| CAP-PFM-LLM-019 | `BR-CAP-PFM-LLM-019-07` Conversation and project access filters a fixed sample owner in shared server memory, creating neither actual user nor workspace isolation and losing updates on restart. | yes | ATS-PFM-LLM-019-04 |  |
| CAP-PFM-LLM-019 | `BR-CAP-PFM-LLM-019-08` Projects support creation, naming, description and deletion; deleting a project retains its conversations as unfiled, and project lists sort by name with unarchived conversation counts. | yes | ATS-PFM-LLM-019-05 |  |
| CAP-PFM-LLM-019 | `BR-CAP-PFM-LLM-019-09` Archive retains a restorable conversation outside active lists and archiving the open conversation starts another conversation. | yes | ATS-PFM-LLM-019-06, ATS-PFM-LLM-019-08 |  |
| CAP-PFM-LLM-019 | `BR-CAP-PFM-LLM-019-10` Conversation and project deletion require confirmation and permanently remove the selected object. | yes | ATS-PFM-LLM-019-07 |  |
| CAP-PFM-LLM-019 | `BR-CAP-PFM-LLM-019-11` Conversation search matches titles or message text without case sensitivity within the selected unarchived list and orders results by recent activity. | no |  |  |
| CAP-PFM-LLM-019 | `BR-CAP-PFM-LLM-019-12` The conversation client writes through console agent routes rather than forwarding conversation writes to Platform’s query client. | no |  |  |
| CAP-PFM-LLM-020 | `BR-CAP-PFM-LLM-020-01` Opening assistance reuses the conversation for a saved query identity or draft scope, including directory scopes, and restores that conversation if archived. | yes | ATS-PFM-LLM-020-01 |  |
| CAP-PFM-LLM-020 | `BR-CAP-PFM-LLM-020-02` Query refinement preserves record type, context and existing column order, appends new columns and applies criteria on screen without saving the query. | yes | ATS-PFM-LLM-001-20, ATS-PFM-LLM-020-02 |  |
| CAP-PFM-LLM-020 | `BR-CAP-PFM-LLM-020-03` Applied changes are identified in conversation history and can be undone latest-first with a reversal note. | yes | ATS-PFM-LLM-019-08, ATS-PFM-LLM-020-03 |  |
| CAP-PFM-LLM-020 | `BR-CAP-PFM-LLM-020-04` Live query interpretation uses Platform’s metered Describe With AI path, while demonstration mode interprets locally. | no |  |  |
| CAP-PFM-LLM-020 | `BR-CAP-PFM-LLM-020-05` Directory interpretation uses local rules and workspace name lookups; an unsupported description leaves filters unchanged and returns an explanation. | yes | ATS-PFM-LLM-020-04 |  |
| CAP-PFM-LLM-020 | `BR-CAP-PFM-LLM-020-06` Scoped conversations accept continued assistance beside their originating query or directory, while general message submission elsewhere returns an instruction to reopen that work. | yes | ATS-PFM-LLM-020-05 |  |
| CAP-PFM-LLM-020 | `BR-CAP-PFM-LLM-020-07` Description-based query creation can ask for the query kind or missing context before opening refinement with the earlier exchange. | no |  |  |
| CAP-PFM-LLM-020 | `BR-CAP-PFM-LLM-020-08` Jazz AI updates record-list filters using the same editing behavior as the list’s own Ask control. | no |  |  |

## Where Each Capability Lives

| Capability | Repository | Role | Paths |
|---|---|---|---|
| (no capability carries repositories[] yet) | | | |

## Features and Journeys

| Artifact | Proof |
|---|---|
| FEAT-PFM-LLM-01 Metered AI Requests | FAS-PFM-LLM-01 |
| FEAT-PFM-LLM-02 AI Catalog and Pricing | FAS-PFM-LLM-02 |
| FEAT-PFM-LLM-03 Monthly AI Allowance and Notices | FAS-PFM-LLM-03 |
| FEAT-PFM-LLM-04 Rhythm HQ AI Usage Oversight | FAS-PFM-LLM-04 |
| FEAT-PFM-LLM-05 Rhythm HQ AI Controls | FAS-PFM-LLM-05 |
| FEAT-PFM-LLM-06 Organization AI Usage | FAS-PFM-LLM-06 |
| FEAT-PFM-LLM-07 Jazz AI Conversations | FAS-PFM-LLM-07 |
| JNY-PFM-LLM-001 A staff member reaches the monthly AI allowance | no product acceptance scenario |
| JNY-PFM-LLM-002 Support investigates an AI refusal | no product acceptance scenario |
| JNY-PFM-LLM-003 A Rhythm administrator grants and withdraws temporary AI capacity | no product acceptance scenario |
| JNY-PFM-LLM-004 An administrator investigates an allowance warning | no product acceptance scenario |
| JNY-PFM-LLM-005 A staff member refines a query and resumes its Jazz AI conversation | no product acceptance scenario |

## Gaps

- `CAP-PFM-LLM-019` has no executable suite
- `CAP-PFM-LLM-020` has no executable suite