@blamejs/core 0.12.26 → 0.12.28

package/CHANGELOG.md

@@ -8,6 +8,10 @@ upgrading across more than a few patches at a time.
 
 ## v0.12.x
 
+- v0.12.28 (2026-05-24) — **`b.ai.capability` — model-capability registry + cheapest-satisfying-model router.** `b.ai.capability.create({ models })` turns a fleet of AI model descriptors into a routing decision: given a set of requirements (context window, input/output modalities, tool use, structured output, reasoning tier, citation support, prompt-caching size), it picks the cheapest model that satisfies all of them. NIST AI RMF (AI 100-1) MAP 2.x requires documenting each model's capabilities and limitations; the Model Cards convention (Mitchell et al., 2019) formalizes that descriptor — this primitive makes the descriptor actionable. Routing to the cheapest sufficient model is a front-line defense against over-provisioning spend and composes directly with `b.ai.quota`'s `cost-usd` dimension (the chosen descriptor's rate feeds the budget charge); refusing to route a request to a model that cannot satisfy it (missing modality, too-small context window, no tool use) catches a capability mismatch before the inference call burns tokens on a guaranteed-bad result. Cost ranking uses a supplied `costBasis` (`{ inputTokens, outputTokens }`) for real per-call spend, else the sum of the per-1k rates; ties break by model id so the choice is deterministic across calls and nodes. **Added:** *`b.ai.capability.create({ models })` — capability registry + router* — Returns `{ describe, list, register, satisfies, route }`. A descriptor carries `maxContextTokens`, `maxOutputTokens`, `modalitiesIn` / `modalitiesOut` (arrays), `toolUse`, `structuredOutput`, `fineTunable`, `reasoningTier` (`none` / `basic` / `standard` / `advanced`, ordered), `citationSupport`, `promptCachingMaxTokens`, and the cost rates `costPer1kInputTokens` / `costPer1kOutputTokens`. Descriptors are validated + frozen at registration so a typo (negative cost, unknown reasoning tier, non-array modality list) surfaces at config time rather than as a silent mis-route. `describe(modelId)` returns the frozen descriptor; `register(modelId, descriptor)` adds or replaces one at runtime. · *`route({ requirements, fallback?, costBasis? })` — cheapest-satisfying selection* — Collects every model whose descriptor satisfies all requirements, then returns the cheapest (`{ modelId, descriptor, estimatedCost, reason }`). Requirements: `minContextTokens`, `minOutputTokens`, `modalitiesIn` / `modalitiesOut` (model must support every listed modality), `toolUse`, `structuredOutput`, `fineTunable`, `minReasoningTier` (tier ordering — `standard` is met by `standard` or `advanced`), `citationSupport`, `minPromptCachingTokens`. When no model matches, `fallback` (a registered model id) is returned with `reason: "fallback"`, or the call refuses with `aiCapability/no-candidate` if no fallback was supplied. Routing decisions emit `ai/capability-routed` / `ai/capability-fallback` / `ai/capability-no-candidate` through the drop-silent audit chain. · *`satisfies(modelId, requirements)` — precise capability-mismatch reasons* — Returns `{ ok, failures }` where each failure names the `requirement`, the `need`, and what the model `have`s — so a caller surfaces a precise reason (e.g. `minReasoningTier need advanced have basic`) instead of a bare boolean. Use it to explain a routing miss or to gate a request against a specific model before calling it.
+
+- v0.12.27 (2026-05-24) — **`b.ai.quota` — per-tenant, per-model AI usage budgets with atomic consume-and-check.** `b.ai.quota.create(opts)` builds an enforcer that caps AI inference usage per `(tenant, model, dimension, period)` and defends OWASP LLM Top 10 2025 LLM10 (Unbounded Consumption) — the class that includes denial-of-wallet, where an attacker drives a high volume of pay-per-use inferences until the bill itself is the attack. Meter by `tokens`, `requests`, `cost-usd`, or `compute-hours` over a calendar-aligned UTC window (`second` through `month`). `consume(tenant, model, amount)` is a single atomic check-and-charge: under the default `hard` enforcement it reserves the amount only if it fits under the ceiling, otherwise it refuses without charging — the limit test and the charge are one indivisible operation, so there is no charge-then-refund window for a concurrent call to observe. The in-memory counter is per-process; multi-node deployments supply an `opts.store` adapter whose `reserve` (an atomic conditional test-and-charge — a Redis Lua script, a SQL `UPDATE ... WHERE used + :amt <= :limit RETURNING used`) and `add` are atomic on the shared backend to enforce one aggregate ceiling across the cluster without false denials under contention. Limit resolution is most-specific-first: `perTenantModel` over `perTenant` over `perModel` over the default `limit`; tenant and model identifiers are percent-encoded into the counter key so a hostile tenant name cannot collide with another tenant's budget. **Added:** *`b.ai.quota.create(opts)` — per-tenant AI usage-budget enforcer* — Returns `{ consume, check, snapshot, reset }` scoped to one `dimension` (`tokens` / `requests` / `cost-usd` / `compute-hours`) and one `period` (`second` / `minute` / `hour` / `day` / `week` (Monday-aligned) / `month` (1st-of-month), all UTC-aligned). `consume(tenant, model, amount, opts?)` returns `{ used, limit, remaining, allowed, exceeded, windowStart, resetsAt, ... }`. `check(tenant, model)` is the read-only snapshot. Spin up one enforcer per dimension you meter — a monthly `cost-usd` budget and a per-minute `tokens` burst cap coexist as two `create()` calls sharing one store. Defends OWASP LLM10:2025 Unbounded Consumption / denial-of-wallet; maps to NIST AI RMF (AI 100-1) MANAGE 2.x and EU AI Act Art. 15 (robustness / resource-exhaustion resilience). · *`hard` / `soft` / `warn` enforcement* — `hard` (default) refuses the over-budget call and throws `aiQuota/exceeded` without charging — the rejected reservation is refunded so the counter is untouched. `soft` admits the charge but reports `allowed: false` so the caller decides whether to honor it. `warn` admits and allows (advisory), flagging `exceeded: true`. A per-call `consume(..., { enforcement })` override lets one endpoint soften the mode for a trusted internal caller without a second enforcer. Every over-budget event emits `ai/quota-exceeded` through the drop-silent audit chain (`ai/quota-applied` on success), tagged with the active cluster node id for attribution. · *Cross-node aggregate budgets via `opts.store`* — The default counter is in-memory (per-process). Supply `opts.store` exposing atomic `reserve` / `add` / `get` / `reset` (a Redis Lua script, a shared SQL row) and the ceiling is enforced on the cluster-wide aggregate. `hard` mode goes through `reserve`, an atomic conditional test-and-charge that adds the amount only if it fits — so a concurrent over-budget call cannot transiently inflate the counter and falsely deny a smaller call that should fit. Per-tenant and per-model limit overrides (`perTenant` / `perModel` / `perTenantModel`) are validated at config time so a malformed cap surfaces at boot, not as a silent fall-through to the default.
+
 - v0.12.26 (2026-05-24) — **`b.compliance` posture cascades — `eu-ai-act` + `ca-ab-853` + `cac-genai-label` POSTURE_DEFAULTS + backup encryption refusal.** Three new posture cascades wired into `b.compliance.POSTURE_DEFAULTS` + `KNOWN_POSTURES` + `REGIME_MAP` so operators globally pinning the EU AI Act / California AB-853 / China CAC GenAI postures get the right floors automatically: backupEncryptionRequired:true, auditChainSignedRequired:true, tlsMinVersion:TLSv1.3, requireVacuumAfterErase:true. `b.backup.bundleAdapterStorage` extends the encryption-required posture list to include the three new postures so `cryptoStrategy: "none"` is refused upfront under any of them (parity with HIPAA + PCI-DSS, which the operator surface has carried since v0.12.10). The canonical `eu-ai-act` posture is the production name; the legacy `ai-act` short name stays in KNOWN_POSTURES for back-compat with operators who pinned it pre-v0.12.26. **Added:** *`eu-ai-act` posture cascade — Regulation (EU) 2024/1689* — POSTURE_DEFAULTS entry: backupEncryptionRequired:true (Art. 12 logging + Art. 15 robustness/cybersecurity demand encryption-at-rest for high-risk system training logs), auditChainSignedRequired:true (Art. 12 + Art. 13 audit-chain integrity), tlsMinVersion:TLSv1.3, requireVacuumAfterErase:true (Art. 50(4) synthetic-content provenance — residual EXIF / metadata pointing at the generating model must be cleared on erase). REGIME_MAP entry under jurisdiction:"EU" domain:"ai-governance". KNOWN_POSTURES carries both `eu-ai-act` (canonical) and `ai-act` (legacy short name). · *`ca-ab-853` posture cascade — California AB-853 effective 2026* — Same encryption + audit floor as eu-ai-act; jurisdiction:"US-CA". Model-generated content watermarking + disclosure regime. Operators serving California traffic pin this posture for the AB-853 §22949.91 obligations the v0.12.12 deepfake primitive's crossWalk references. · *`cac-genai-label` posture cascade — China CAC GenAI Service Measures* — Synthetic-content labelling per Art. 12 + algorithm filing per Art. 4. Same backup encryption + signed audit chain floor. Operators serving Chinese traffic pin this posture so the bundleAdapterStorage refuses plaintext bundles and the disclosure primitive's `jurisdiction: "cn"` cross-walk produces the right legal-reference array. · *`bundleAdapterStorage` BACKUP_ENCRYPTION_REQUIRED_POSTURES extended* — `hipaa` + `pci-dss` (the v0.12.10 baseline) joined by the three AI postures. `cryptoStrategy: "none"` refused upfront under any of `eu-ai-act` / `ca-ab-853` / `cac-genai-label` with `backup/posture-requires-encryption`. Operators wiring backup storage in a regulated AI deployment now get the same posture-driven gate that the storage primitive has always applied to health + payment data.
 
 - v0.12.25 (2026-05-24) — **`b.ai.disclosure.applyAll(scenario)` — bundle Art. 50(1) / 50(3) / 50(4) disclosures for mixed-modality AI systems.** Composes the three v0.12.12 disclosure primitives (chatbot / deepfake / emotion) into a single bundled emit. Operators running mixed-modality AI systems (e.g. a chatbot that also generates images, or an emotion-recognition system embedded in a chat flow) declare which Art. 50 obligations apply via `scenario.kinds` and the primitive fans out to the per-obligation emit calls in one pass. Shared opts (jurisdiction, language, audit, correlationId) propagate to every per-kind emission so the cross-walk + audit-chain entries stay correlated across the bundle. **Added:** *`b.ai.disclosure.applyAll(scenario)` — multi-obligation bundled emit* — `scenario.kinds: ["chatbot", "deepfake", "emotion"]` (subset) selects which Art. 50 obligations to satisfy. Per-kind required fields (session for chatbot, content + contentType for deepfake) refused upfront when missing. Returns `{ disclosures: { chatbot?, deepfake?, emotion? } }` with each entry being the corresponding primitive's emission payload. Shared opts propagate: `scenario.jurisdiction` / `scenario.language` / `scenario.audit` / `scenario.correlationId` reach every per-kind call so a US-CA deployment serving chat + image gets both the AB-853 cross-walk AND the Art. 50(1) audit event under the same correlationId.

package/README.md

@@ -162,6 +162,8 @@ The framework bundles the surface a typical Node app reaches for. Every primitiv
 - **Prompt-injection classification** — OWASP LLM01:2025 / NIST COSAIS RFI (`b.ai.input.classify`)
 - **Agent identity** — A2A signed agent-card primitive (Linux Foundation Agentic AI Foundation v1.x, ML-DSA-87) (`b.a2a`)
 - **Content provenance** — C2PA 2.1 + California SB-942 / AB-853 manifest builder for AI-generated media (provider, model id + version, timestamp, content ID, signed) (`b.contentCredentials`)
+- **AI usage quotas** — per-tenant / per-model budgets metered by tokens / requests / cost-usd / compute-hours over calendar-aligned windows, with an atomic conditional reserve (no charge-then-refund race) + hard/soft/warn enforcement and an optional cross-node store; defends OWASP LLM10:2025 unbounded consumption / denial-of-wallet (`b.ai.quota`)
+- **AI capability routing** — model-capability registry (context window / modalities / tool use / reasoning tier / cost rates) + a router that picks the cheapest model satisfying a request's requirements, refusing capability mismatches before the inference call (NIST AI RMF MAP + Model Cards); composes with `b.ai.quota` cost budgets (`b.ai.capability`)
 ### Compliance regimes
 
 - **Posture coordinator** — `b.compliance` cascades operator-declared regime into retention / audit / db / cryptoField via POSTURE_DEFAULTS:

package/index.js

@@ -443,6 +443,8 @@ module.exports = {
     aiContentDetect: require("./lib/ai-content-detect"),
     modelManifest:   require("./lib/ai-model-manifest"),
     disclosure:      require("./lib/ai-disclosure"),
+    quota:           require("./lib/ai-quota"),
+    capability:      require("./lib/ai-capability"),
   },
   promisePool:      require("./lib/promise-pool"),
   sdNotify:         require("./lib/sd-notify"),

package/lib/ai-capability.js

@@ -0,0 +1,482 @@
+"use strict";
+/**
+ * @module b.ai.capability
+ * @nav    AI
+ * @title  AI capability routing
+ *
+ * @intro
+ *   A capability registry + capability-aware router for AI model
+ *   fleets. NIST AI RMF (AI 100-1) MAP 2.x requires documenting each
+ *   model's capabilities and limitations; the Model Cards convention
+ *   (Mitchell et al., 2019) formalizes that descriptor. This module
+ *   turns those descriptors into a routing decision: given a set of
+ *   requirements (context window, modalities, tool use, reasoning
+ *   tier, …), pick the <em>cheapest</em> model in the fleet that
+ *   satisfies all of them, or fall back deterministically.
+ *
+ *   <code>b.ai.capability.create({ models })</code> builds a registry
+ *   from operator-supplied descriptors and returns:
+ *
+ *   - <code>describe(modelId)</code> — the frozen descriptor.
+ *   - <code>list()</code> — every registered model id.
+ *   - <code>register(modelId, descriptor)</code> — add / replace one.
+ *   - <code>satisfies(modelId, requirements)</code> —
+ *     <code>{ ok, failures }</code> where each failure names the
+ *     requirement, the need, and what the model has.
+ *   - <code>route({ requirements, fallback?, costBasis? })</code> —
+ *     the cheapest satisfying model, or the fallback, or a refusal.
+ *
+ *   A descriptor carries: <code>maxContextTokens</code>,
+ *   <code>maxOutputTokens</code>, <code>modalitiesIn</code> /
+ *   <code>modalitiesOut</code> (arrays — e.g. <code>"text"</code>,
+ *   <code>"image"</code>, <code>"audio"</code>, <code>"video"</code>),
+ *   <code>toolUse</code>, <code>structuredOutput</code>,
+ *   <code>fineTunable</code>, <code>reasoningTier</code>
+ *   (<code>"none" | "basic" | "standard" | "advanced"</code>,
+ *   ordered), <code>citationSupport</code>,
+ *   <code>promptCachingMaxTokens</code>, and the cost rates
+ *   <code>costPer1kInputTokens</code> / <code>costPer1kOutputTokens</code>.
+ *
+ *   <strong>Routing picks the cheapest match.</strong> When a
+ *   <code>costBasis</code> (<code>{ inputTokens, outputTokens }</code>)
+ *   is supplied the router estimates the per-call cost and ranks by
+ *   it; otherwise it ranks by the sum of the per-1k rates. Ties break
+ *   by model id so the choice is deterministic. Routing to the
+ *   cheapest sufficient model is the front-line defense against
+ *   over-provisioning spend — it composes with
+ *   <code>b.ai.quota</code>'s <code>cost-usd</code> dimension, where
+ *   the chosen descriptor's rate feeds the budget charge.
+ *
+ *   Refusing to route a request to a model that cannot satisfy it
+ *   (missing modality, too-small context window, no tool use) catches
+ *   a capability mismatch before the inference call burns tokens on a
+ *   guaranteed-bad result.
+ *
+ * @card
+ *   Capability registry + cheapest-satisfying-model router for AI
+ *   model fleets (context / modalities / tool use / reasoning tier /
+ *   cost). Composes with b.ai.quota cost budgets.
+ */
+
+var lazyRequire = require("./lazy-require");
+var validateOpts = require("./validate-opts");
+var { defineClass } = require("./framework-error");
+
+var AiCapabilityError = defineClass("AiCapabilityError", { alwaysPermanent: true });
+
+var audit = lazyRequire(function () { return require("./audit"); });
+
+// Ordered reasoning tiers — a requirement of `minReasoningTier:
+// "standard"` is satisfied by "standard" or "advanced", not "basic".
+var REASONING_TIERS = ["none", "basic", "standard", "advanced"];
+
+// Cost rates are quoted per 1000 tokens (industry convention; the
+// descriptor fields are costPer1kInputTokens / costPer1kOutputTokens).
+// Dividing a token count by this rate unit converts a per-1k rate into
+// the per-token multiplier — a rate denominator, not a byte size.
+var COST_RATE_TOKEN_UNIT = 1000;   // allow:raw-byte-literal — per-1k-token cost-rate denominator, not a byte count
+
+var DESCRIPTOR_KEYS = [
+  "maxContextTokens", "maxOutputTokens", "modalitiesIn", "modalitiesOut",
+  "toolUse", "structuredOutput", "fineTunable", "reasoningTier",
+  "citationSupport", "promptCachingMaxTokens",
+  "costPer1kInputTokens", "costPer1kOutputTokens", "provider", "version",
+];
+
+var REQUIREMENT_KEYS = [
+  "minContextTokens", "minOutputTokens", "modalitiesIn", "modalitiesOut",
+  "toolUse", "structuredOutput", "fineTunable", "minReasoningTier",
+  "citationSupport", "minPromptCachingTokens",
+];
+
+function _isPositiveInt(n) {
+  return typeof n === "number" && isFinite(n) && n > 0 && Math.floor(n) === n;
+}
+function _isNonNegFinite(n) {
+  return typeof n === "number" && isFinite(n) && n >= 0;
+}
+function _isStringArray(a) {
+  if (!Array.isArray(a)) return false;
+  for (var i = 0; i < a.length; i++) {
+    if (typeof a[i] !== "string" || a[i].length === 0) return false;
+  }
+  return true;
+}
+
+// Normalize + validate one descriptor at registration time so a typo
+// (negative cost, unknown reasoning tier, non-array modality list)
+// surfaces at config time rather than as a silent mis-route.
+function _normalizeDescriptor(modelId, d) {
+  if (!d || typeof d !== "object" || Array.isArray(d)) {
+    throw new AiCapabilityError("aiCapability/bad-descriptor",
+      "ai.capability: descriptor for '" + modelId + "' must be a plain object");
+  }
+  validateOpts(d, DESCRIPTOR_KEYS, "ai.capability descriptor['" + modelId + "']");
+
+  if (!_isPositiveInt(d.maxContextTokens)) {
+    throw new AiCapabilityError("aiCapability/bad-descriptor",
+      "ai.capability: '" + modelId + "'.maxContextTokens must be a positive integer");
+  }
+  var maxOut = (d.maxOutputTokens == null) ? d.maxContextTokens : d.maxOutputTokens;
+  if (!_isPositiveInt(maxOut)) {
+    throw new AiCapabilityError("aiCapability/bad-descriptor",
+      "ai.capability: '" + modelId + "'.maxOutputTokens must be a positive integer");
+  }
+
+  var modIn = (d.modalitiesIn == null) ? ["text"] : d.modalitiesIn;
+  var modOut = (d.modalitiesOut == null) ? ["text"] : d.modalitiesOut;
+  if (!_isStringArray(modIn) || !_isStringArray(modOut)) {
+    throw new AiCapabilityError("aiCapability/bad-descriptor",
+      "ai.capability: '" + modelId + "'.modalitiesIn / modalitiesOut must be arrays of non-empty strings");
+  }
+
+  var tier = (d.reasoningTier == null) ? "standard" : d.reasoningTier;
+  if (REASONING_TIERS.indexOf(tier) === -1) {
+    throw new AiCapabilityError("aiCapability/bad-descriptor",
+      "ai.capability: '" + modelId + "'.reasoningTier must be one of " + REASONING_TIERS.join(" / "));
+  }
+
+  var cachingMax = (d.promptCachingMaxTokens == null) ? 0 : d.promptCachingMaxTokens;
+  var costIn = (d.costPer1kInputTokens == null) ? 0 : d.costPer1kInputTokens;
+  var costOut = (d.costPer1kOutputTokens == null) ? 0 : d.costPer1kOutputTokens;
+  if (!_isNonNegFinite(cachingMax) || !_isNonNegFinite(costIn) || !_isNonNegFinite(costOut)) {
+    throw new AiCapabilityError("aiCapability/bad-descriptor",
+      "ai.capability: '" + modelId + "'.promptCachingMaxTokens / costPer1kInputTokens / " +
+      "costPer1kOutputTokens must be non-negative finite numbers");
+  }
+
+  return Object.freeze({
+    modelId:                modelId,
+    maxContextTokens:       d.maxContextTokens,
+    maxOutputTokens:        maxOut,
+    modalitiesIn:           Object.freeze(modIn.slice()),
+    modalitiesOut:          Object.freeze(modOut.slice()),
+    toolUse:                d.toolUse === true,
+    structuredOutput:       d.structuredOutput === true,
+    fineTunable:            d.fineTunable === true,
+    reasoningTier:          tier,
+    citationSupport:        d.citationSupport === true,
+    promptCachingMaxTokens: cachingMax,
+    costPer1kInputTokens:   costIn,
+    costPer1kOutputTokens:  costOut,
+    provider:               (typeof d.provider === "string") ? d.provider : null,
+    version:                (typeof d.version === "string") ? d.version : null,
+  });
+}
+
+/**
+ * @primitive b.ai.capability.create
+ * @signature b.ai.capability.create(opts)
+ * @since     0.12.28
+ * @status    stable
+ * @compliance soc2
+ * @related   b.ai.quota.create, b.ai.modelManifest.build
+ *
+ * Build a capability registry + router from operator-supplied model
+ * descriptors. Returns <code>{ describe, list, register, satisfies,
+ * route }</code>. Pair it with <code>b.ai.quota</code>:
+ * <code>route()</code> picks the cheapest model that meets the
+ * request, and the chosen descriptor's cost rate feeds the
+ * <code>cost-usd</code> budget charge.
+ *
+ * @opts
+ *   {
+ *     models: {                       // required, ≥ 1 entry
+ *       [modelId: string]: {
+ *         maxContextTokens:        number,    // required, positive int
+ *         maxOutputTokens?:        number,    // default: maxContextTokens
+ *         modalitiesIn?:           string[],  // default: ["text"]
+ *         modalitiesOut?:          string[],  // default: ["text"]
+ *         toolUse?:                boolean,   // default: false
+ *         structuredOutput?:       boolean,   // default: false
+ *         fineTunable?:            boolean,   // default: false
+ *         reasoningTier?:          string,    // none|basic|standard|advanced
+ *         citationSupport?:        boolean,   // default: false
+ *         promptCachingMaxTokens?: number,    // default: 0
+ *         costPer1kInputTokens?:   number,    // default: 0
+ *         costPer1kOutputTokens?:  number,    // default: 0
+ *         provider?:               string,
+ *         version?:                string,
+ *       }
+ *     },
+ *     audit?: boolean,                // default: true (route decisions)
+ *   }
+ *
+ * @example
+ *   var fleet = b.ai.capability.create({
+ *     models: {
+ *       "haiku":  { maxContextTokens: 200000, reasoningTier: "basic",
+ *                   costPer1kInputTokens: 0.001, costPer1kOutputTokens: 0.005 },
+ *       "opus":   { maxContextTokens: 200000, reasoningTier: "advanced",
+ *                   toolUse: true, modalitiesIn: ["text", "image"],
+ *                   costPer1kInputTokens: 0.015, costPer1kOutputTokens: 0.075 },
+ *     },
+ *   });
+ *   var pick = fleet.route({
+ *     requirements: { minContextTokens: 100000, toolUse: true,
+ *                     modalitiesIn: ["text", "image"] },
+ *     costBasis:    { inputTokens: 4000, outputTokens: 500 },
+ *   });
+ *   // → { modelId: "opus", descriptor: {...}, estimatedCost: 0.0975, reason: "cheapest-of-1" }
+ */
+function create(opts) {
+  validateOpts.requireObject(opts, "ai.capability.create", AiCapabilityError);
+  validateOpts(opts, ["models", "audit"], "ai.capability.create");
+
+  if (!opts.models || typeof opts.models !== "object" || Array.isArray(opts.models)) {
+    throw new AiCapabilityError("aiCapability/bad-models",
+      "ai.capability.create: models must be a plain object { modelId: descriptor }");
+  }
+  var ids = Object.keys(opts.models);
+  if (ids.length === 0) {
+    throw new AiCapabilityError("aiCapability/bad-models",
+      "ai.capability.create: models must declare at least one model");
+  }
+
+  var registry = new Map();
+  for (var i = 0; i < ids.length; i++) {
+    registry.set(ids[i], _normalizeDescriptor(ids[i], opts.models[ids[i]]));
+  }
+  var auditOn = opts.audit !== false;
+
+  function _emitAudit(action, outcome, metadata) {
+    if (!auditOn) return;
+    try {
+      audit().safeEmit({ action: action, outcome: outcome, metadata: metadata || {} });
+    } catch (_e) { /* audit best-effort — drop-silent */ }
+  }
+
+  function describe(modelId) {
+    var d = registry.get(modelId);
+    if (!d) {
+      throw new AiCapabilityError("aiCapability/unknown-model",
+        "ai.capability.describe: unknown model '" + modelId + "'");
+    }
+    return d;
+  }
+
+  function list() {
+    return Array.from(registry.keys());
+  }
+
+  function register(modelId, descriptor) {
+    validateOpts.requireNonEmptyString(modelId,
+      "ai.capability.register: modelId", AiCapabilityError, "aiCapability/bad-model");
+    registry.set(modelId, _normalizeDescriptor(modelId, descriptor));
+    return registry.get(modelId);
+  }
+
+  // Returns { ok, failures } — every unmet requirement names what was
+  // needed and what the model has, so a caller can surface a precise
+  // capability-mismatch reason instead of a bare boolean.
+  function _evaluate(descriptor, requirements) {
+    var failures = [];
+    function fail(requirement, need, have) {
+      failures.push({ requirement: requirement, need: need, have: have });
+    }
+    if (requirements.minContextTokens != null &&
+        descriptor.maxContextTokens < requirements.minContextTokens) {
+      fail("minContextTokens", requirements.minContextTokens, descriptor.maxContextTokens);
+    }
+    if (requirements.minOutputTokens != null &&
+        descriptor.maxOutputTokens < requirements.minOutputTokens) {
+      fail("minOutputTokens", requirements.minOutputTokens, descriptor.maxOutputTokens);
+    }
+    if (requirements.modalitiesIn != null) {
+      for (var a = 0; a < requirements.modalitiesIn.length; a++) {
+        if (descriptor.modalitiesIn.indexOf(requirements.modalitiesIn[a]) === -1) {
+          fail("modalitiesIn", requirements.modalitiesIn[a], descriptor.modalitiesIn);
+        }
+      }
+    }
+    if (requirements.modalitiesOut != null) {
+      for (var b = 0; b < requirements.modalitiesOut.length; b++) {
+        if (descriptor.modalitiesOut.indexOf(requirements.modalitiesOut[b]) === -1) {
+          fail("modalitiesOut", requirements.modalitiesOut[b], descriptor.modalitiesOut);
+        }
+      }
+    }
+    if (requirements.toolUse === true && descriptor.toolUse !== true) {
+      fail("toolUse", true, false);
+    }
+    if (requirements.structuredOutput === true && descriptor.structuredOutput !== true) {
+      fail("structuredOutput", true, false);
+    }
+    if (requirements.fineTunable === true && descriptor.fineTunable !== true) {
+      fail("fineTunable", true, false);
+    }
+    if (requirements.citationSupport === true && descriptor.citationSupport !== true) {
+      fail("citationSupport", true, false);
+    }
+    if (requirements.minReasoningTier != null &&
+        REASONING_TIERS.indexOf(descriptor.reasoningTier) <
+        REASONING_TIERS.indexOf(requirements.minReasoningTier)) {
+      fail("minReasoningTier", requirements.minReasoningTier, descriptor.reasoningTier);
+    }
+    if (requirements.minPromptCachingTokens != null &&
+        descriptor.promptCachingMaxTokens < requirements.minPromptCachingTokens) {
+      fail("minPromptCachingTokens", requirements.minPromptCachingTokens, descriptor.promptCachingMaxTokens);
+    }
+    return { ok: failures.length === 0, failures: failures };
+  }
+
+  function _validateRequirements(requirements) {
+    if (requirements == null) return {};
+    if (typeof requirements !== "object" || Array.isArray(requirements)) {
+      throw new AiCapabilityError("aiCapability/bad-requirements",
+        "ai.capability: requirements must be a plain object");
+    }
+    validateOpts(requirements, REQUIREMENT_KEYS, "ai.capability requirements");
+    if (requirements.minReasoningTier != null &&
+        REASONING_TIERS.indexOf(requirements.minReasoningTier) === -1) {
+      throw new AiCapabilityError("aiCapability/bad-requirements",
+        "ai.capability: minReasoningTier must be one of " + REASONING_TIERS.join(" / "));
+    }
+    if (requirements.modalitiesIn != null && !_isStringArray(requirements.modalitiesIn)) {
+      throw new AiCapabilityError("aiCapability/bad-requirements",
+        "ai.capability: requirements.modalitiesIn must be an array of non-empty strings");
+    }
+    if (requirements.modalitiesOut != null && !_isStringArray(requirements.modalitiesOut)) {
+      throw new AiCapabilityError("aiCapability/bad-requirements",
+        "ai.capability: requirements.modalitiesOut must be an array of non-empty strings");
+    }
+    // Numeric minimums are compared with `<` against the descriptor; a
+    // non-numeric value (NaN, "128k", a bad parse) makes that compare
+    // false and SILENTLY satisfies the requirement, so an undersized
+    // model could be selected. Reject non-finite / negative here so a
+    // malformed requirement fails fast instead of fail-open.
+    var numericMins = ["minContextTokens", "minOutputTokens", "minPromptCachingTokens"];
+    for (var ni = 0; ni < numericMins.length; ni++) {
+      var nk = numericMins[ni];
+      if (requirements[nk] != null && !_isNonNegFinite(requirements[nk])) {
+        throw new AiCapabilityError("aiCapability/bad-requirements",
+          "ai.capability: requirements." + nk + " must be a non-negative finite number");
+      }
+    }
+    // Boolean opt-in requirements are matched with `=== true`; a
+    // non-boolean (truthy 1, "false") would silently fail to require
+    // the capability. Reject non-booleans so the intent is explicit.
+    var booleanReqs = ["toolUse", "structuredOutput", "fineTunable", "citationSupport"];
+    for (var bi = 0; bi < booleanReqs.length; bi++) {
+      var bk = booleanReqs[bi];
+      if (requirements[bk] != null && typeof requirements[bk] !== "boolean") {
+        throw new AiCapabilityError("aiCapability/bad-requirements",
+          "ai.capability: requirements." + bk + " must be a boolean");
+      }
+    }
+    return requirements;
+  }
+
+  function satisfies(modelId, requirements) {
+    return _evaluate(describe(modelId), _validateRequirements(requirements));
+  }
+
+  // Per-call cost estimate. With a costBasis the estimate is the
+  // real per-call spend (input + output tokens at the model's rates);
+  // without one it is the sum of the per-1k rates — a stable proxy
+  // for "cheaper model" when the caller hasn't sized the request.
+  function _estimateCost(descriptor, costBasis) {
+    if (costBasis) {
+      var inTok = _isNonNegFinite(costBasis.inputTokens) ? costBasis.inputTokens : 0;
+      var outTok = _isNonNegFinite(costBasis.outputTokens) ? costBasis.outputTokens : 0;
+      return (inTok / COST_RATE_TOKEN_UNIT) * descriptor.costPer1kInputTokens +
+             (outTok / COST_RATE_TOKEN_UNIT) * descriptor.costPer1kOutputTokens;
+    }
+    return descriptor.costPer1kInputTokens + descriptor.costPer1kOutputTokens;
+  }
+
+  function route(routeOpts) {
+    routeOpts = routeOpts || {};
+    validateOpts(routeOpts, ["requirements", "fallback", "costBasis"], "ai.capability.route");
+    var requirements = _validateRequirements(routeOpts.requirements);
+    var costBasis = null;
+    if (routeOpts.costBasis != null) {
+      if (typeof routeOpts.costBasis !== "object" || Array.isArray(routeOpts.costBasis)) {
+        throw new AiCapabilityError("aiCapability/bad-requirements",
+          "ai.capability.route: costBasis must be a plain object { inputTokens, outputTokens }");
+      }
+      validateOpts(routeOpts.costBasis, ["inputTokens", "outputTokens"],
+        "ai.capability.route costBasis");
+      // A malformed costBasis field silently underprices a candidate
+      // and biases the "cheapest" choice toward the wrong model — fail
+      // fast instead. An absent field is fine (treated as 0 tokens on
+      // that side); a present-but-non-numeric field is rejected.
+      var cbFields = ["inputTokens", "outputTokens"];
+      for (var ci = 0; ci < cbFields.length; ci++) {
+        var ck = cbFields[ci];
+        if (routeOpts.costBasis[ck] != null && !_isNonNegFinite(routeOpts.costBasis[ck])) {
+          throw new AiCapabilityError("aiCapability/bad-requirements",
+            "ai.capability.route: costBasis." + ck + " must be a non-negative finite number");
+        }
+      }
+      costBasis = routeOpts.costBasis;
+    }
+
+    // Collect every satisfying model, then pick the cheapest. Tie
+    // break by model id (lexicographic) so the choice is deterministic
+    // across calls and across nodes.
+    var candidates = [];
+    var modelIds = Array.from(registry.keys());
+    for (var i = 0; i < modelIds.length; i++) {
+      var d = registry.get(modelIds[i]);
+      if (_evaluate(d, requirements).ok) {
+        candidates.push({ modelId: modelIds[i], descriptor: d, cost: _estimateCost(d, costBasis) });
+      }
+    }
+    candidates.sort(function (x, y) {
+      if (x.cost !== y.cost) return x.cost - y.cost;
+      return x.modelId < y.modelId ? -1 : (x.modelId > y.modelId ? 1 : 0);
+    });
+
+    if (candidates.length > 0) {
+      var pick = candidates[0];
+      _emitAudit("ai/capability-routed", "allowed", {
+        modelId: pick.modelId, candidateCount: candidates.length,
+        estimatedCost: pick.cost, requirements: requirements,
+      });
+      return {
+        modelId:       pick.modelId,
+        descriptor:    pick.descriptor,
+        estimatedCost: pick.cost,
+        reason:        "cheapest-of-" + candidates.length,
+      };
+    }
+
+    // No model satisfies the requirements.
+    if (routeOpts.fallback != null) {
+      var fb = registry.get(routeOpts.fallback);
+      if (!fb) {
+        throw new AiCapabilityError("aiCapability/unknown-model",
+          "ai.capability.route: fallback '" + routeOpts.fallback + "' is not a registered model");
+      }
+      _emitAudit("ai/capability-fallback", "allowed", {
+        modelId: routeOpts.fallback, requirements: requirements,
+      });
+      return {
+        modelId:       routeOpts.fallback,
+        descriptor:    fb,
+        estimatedCost: _estimateCost(fb, costBasis),
+        reason:        "fallback",
+      };
+    }
+
+    _emitAudit("ai/capability-no-candidate", "denied", { requirements: requirements });
+    throw new AiCapabilityError("aiCapability/no-candidate",
+      "ai.capability.route: no registered model satisfies the requirements " +
+      "and no fallback was supplied");
+  }
+
+  return {
+    describe:  describe,
+    list:      list,
+    register:  register,
+    satisfies: satisfies,
+    route:     route,
+  };
+}
+
+module.exports = {
+  create:             create,
+  REASONING_TIERS:    REASONING_TIERS,
+  AiCapabilityError:  AiCapabilityError,
+};

package/lib/ai-quota.js

@@ -0,0 +1,526 @@
+"use strict";
+/**
+ * @module b.ai.quota
+ * @nav    Compliance
+ * @title  AI usage quota
+ *
+ * @intro
+ *   Per-tenant, per-model usage budgets for AI inference endpoints.
+ *   OWASP LLM Top 10 2025 ranks <strong>LLM10: Unbounded
+ *   Consumption</strong> — the class that includes "denial of
+ *   wallet" (DoW), where an attacker drives a high volume of
+ *   pay-per-use inferences until the bill itself becomes the
+ *   attack — as a top application risk. A single misbehaving (or
+ *   compromised) tenant can saturate context windows, exhaust GPU
+ *   minutes, or run up an unbounded cloud-inference bill long
+ *   before a human notices.
+ *
+ *   This primitive enforces a hard ceiling per
+ *   <code>(tenant, model, dimension, period)</code>:
+ *
+ *   - <code>dimension</code> — what is being metered:
+ *     <code>"tokens"</code> (context + completion tokens),
+ *     <code>"requests"</code> (inference calls),
+ *     <code>"cost-usd"</code> (provider spend), or
+ *     <code>"compute-hours"</code> (GPU / accelerator time).
+ *   - <code>period</code> — the budget window, calendar-aligned in
+ *     UTC: <code>"second"</code>, <code>"minute"</code>,
+ *     <code>"hour"</code>, <code>"day"</code>, <code>"week"</code>
+ *     (Monday-aligned), or <code>"month"</code> (1st-of-month).
+ *   - <code>enforcement</code> — <code>"hard"</code> (default,
+ *     refuse the over-budget call), <code>"soft"</code> (admit but
+ *     report <code>allowed:false</code> so the caller decides), or
+ *     <code>"warn"</code> (admit + audit only).
+ *
+ *   <code>consume(tenant, model, amount)</code> is the single
+ *   atomic check-and-charge entry point: in <code>"hard"</code>
+ *   mode it reserves <code>amount</code> only if it fits under the
+ *   limit, otherwise it refuses without charging. There is no
+ *   separate "check then add" two-call shape to race against — the
+ *   reservation and the limit test happen in one operation.
+ *
+ *   <strong>Single-process by default; cross-node via store.</strong>
+ *   The in-memory counter is per-process. Multi-node deployments
+ *   that need an aggregate ceiling across the cluster supply an
+ *   <code>opts.store</code> adapter whose <code>reserve</code> (an
+ *   atomic conditional test-and-charge — "add only if current +
+ *   amount fits under the limit") and <code>add</code> are atomic on
+ *   the shared backend: a Redis Lua script, or a SQL
+ *   <code>UPDATE ... SET used = used + :amt WHERE used + :amt &lt;= :limit
+ *   RETURNING used</code>. The conditional reserve is what keeps
+ *   <code>hard</code> enforcement correct under cross-node
+ *   contention — there is no charge-then-refund window for a
+ *   concurrent call to observe. The framework records the active
+ *   cluster node id on every breach event so a denial-of-wallet
+ *   spike is attributable.
+ *
+ *   Limit resolution is most-specific-first:
+ *   <code>perTenantModel[t|m]</code> →
+ *   <code>perTenant[t]</code> → <code>perModel[m]</code> →
+ *   <code>limit</code> (the default). Tenant and model identifiers
+ *   are percent-encoded into the counter key so a hostile tenant
+ *   name cannot collide with another tenant's budget.
+ *
+ *   Audit emissions (drop-silent via <code>b.audit.safeEmit</code>):
+ *     - <code>ai/quota-applied</code>  — a consume succeeded.
+ *     - <code>ai/quota-exceeded</code> — a consume hit the ceiling
+ *       (refused under <code>"hard"</code>; reported under
+ *       <code>"soft"</code> / <code>"warn"</code>).
+ *
+ *   NIST AI RMF (AI 100-1) MANAGE 2.x ("AI system performance and
+ *   trustworthiness are monitored") and EU AI Act Art. 15
+ *   (accuracy, robustness and cybersecurity of high-risk systems —
+ *   resource-exhaustion resilience) map onto this primitive;
+ *   operators wire its emissions into the same audit chain auditors
+ *   read.
+ *
+ * @card
+ *   Per-tenant, per-model AI usage budgets (tokens / requests /
+ *   cost-usd / compute-hours) with atomic consume-and-check.
+ *   Defends OWASP LLM10 unbounded consumption / denial-of-wallet.
+ */
+
+var C = require("./constants");
+var lazyRequire = require("./lazy-require");
+var validateOpts = require("./validate-opts");
+var { defineClass } = require("./framework-error");
+
+var AiQuotaError = defineClass("AiQuotaError", { alwaysPermanent: true });
+
+var audit = lazyRequire(function () { return require("./audit"); });
+var observability = lazyRequire(function () { return require("./observability"); });
+var cluster = lazyRequire(function () { return require("./cluster"); });
+
+var DIMENSIONS   = ["tokens", "requests", "cost-usd", "compute-hours"];
+var PERIODS      = ["second", "minute", "hour", "day", "week", "month"];
+var ENFORCEMENTS = ["hard", "soft", "warn"];
+
+// ---- Calendar-aligned period windows (UTC) ----
+//
+// Fixed-duration periods (second / minute) align to the epoch, which
+// is itself UTC midnight, so a modulo is exact. Hour / day / week /
+// month align to human UTC boundaries via Date.UTC truncation —
+// week starts Monday, month starts on the 1st — so "100k tokens per
+// day" resets at 00:00 UTC, not at a rolling 24h offset from first
+// use.
+
+function _windowStartFor(period, now) {
+  var d = new Date(now);
+  switch (period) {
+    case "second": return now - (now % C.TIME.seconds(1));
+    case "minute": return now - (now % C.TIME.minutes(1));
+    case "hour":
+      return Date.UTC(d.getUTCFullYear(), d.getUTCMonth(), d.getUTCDate(), d.getUTCHours());
+    case "day":
+      return Date.UTC(d.getUTCFullYear(), d.getUTCMonth(), d.getUTCDate());
+    case "week": {
+      var dayMid = Date.UTC(d.getUTCFullYear(), d.getUTCMonth(), d.getUTCDate());
+      var dow = new Date(dayMid).getUTCDay();            // 0=Sun .. 6=Sat
+      var sinceMonday = (dow + 6) % 7;                   // 0=Mon .. 6=Sun
+      return dayMid - sinceMonday * C.TIME.days(1);
+    }
+    case "month":
+      return Date.UTC(d.getUTCFullYear(), d.getUTCMonth(), 1);
+    default:
+      // unreachable — period validated at create()
+      return now;
+  }
+}
+
+function _resetsAtFor(period, windowStart) {
+  var d = new Date(windowStart);
+  switch (period) {
+    case "second": return windowStart + C.TIME.seconds(1);
+    case "minute": return windowStart + C.TIME.minutes(1);
+    case "hour":   return windowStart + C.TIME.hours(1);
+    case "day":    return windowStart + C.TIME.days(1);
+    case "week":   return windowStart + C.TIME.weeks(1);
+    case "month":  return Date.UTC(d.getUTCFullYear(), d.getUTCMonth() + 1, 1);
+    default:       return windowStart;
+  }
+}
+
+// ---- Default in-memory atomic counter store ----
+//
+// Single-threaded JS makes each operation below one indivisible step,
+// so a concurrent caller never observes a partial update. Entries
+// self-expire at the window boundary; reads past expiry return 0 (a
+// fresh window). _keysWithPrefix backs the reset(tenant) enumeration
+// the default store can satisfy without an external scan.
+
+function _memoryStore() {
+  var m = new Map();                                     // key -> { value, expiresAt }
+  function _slot(key, windowMs) {
+    var now = Date.now();
+    var e = m.get(key);
+    if (!e || e.expiresAt <= now) {
+      e = { value: 0, expiresAt: now + windowMs };
+      m.set(key, e);
+    }
+    return e;
+  }
+  return {
+    // Atomic conditional reserve — tests current + amount <= limit and
+    // charges only if it fits, as one indivisible operation. Returns
+    // { allowed, used }; on refusal the amount is NOT charged, so a
+    // concurrent over-budget call cannot transiently inflate the
+    // counter and falsely deny a smaller call that should fit.
+    reserve: function (key, amount, limit, windowMs) {
+      var e = _slot(key, windowMs);
+      if (e.value + amount > limit) return { allowed: false, used: e.value };
+      e.value += amount;
+      return { allowed: true, used: e.value };
+    },
+    // Unconditional add — for soft / warn modes, which always charge.
+    add: function (key, amount, windowMs) {
+      var e = _slot(key, windowMs);
+      e.value += amount;
+      return e.value;
+    },
+    get: function (key) {
+      var e = m.get(key);
+      if (!e || e.expiresAt <= Date.now()) return 0;
+      return e.value;
+    },
+    reset: function (key) {
+      m.delete(key);
+    },
+    _keysWithPrefix: function (prefix) {
+      var out = [];
+      m.forEach(function (_e, k) { if (k.indexOf(prefix) === 0) out.push(k); });
+      return out;
+    },
+    _clear: function () { m.clear(); },
+  };
+}
+
+/**
+ * @primitive b.ai.quota.create
+ * @signature b.ai.quota.create(opts)
+ * @since     0.12.27
+ * @status    stable
+ * @compliance soc2, gdpr
+ * @related   b.tenantQuota.budget, b.ai.disclosure.chatbot
+ *
+ * Build a per-tenant AI usage-budget enforcer scoped to one
+ * <code>dimension</code> and one <code>period</code>. Returns an
+ * object exposing <code>consume(tenant, model, amount, opts?)</code>
+ * (the atomic check-and-charge), <code>check(tenant, model)</code>
+ * (read-only snapshot), <code>snapshot(tenant, model)</code> (alias
+ * of <code>check</code>), and <code>reset(tenant?, model?)</code>
+ * (drop the current window's counters).
+ *
+ * Spin up one enforcer per dimension you meter — e.g. a
+ * <code>"cost-usd"</code> monthly budget and a
+ * <code>"tokens"</code> per-minute burst cap can coexist as two
+ * <code>create()</code> calls sharing the same store.
+ *
+ * @opts
+ *   {
+ *     dimension:        string,    // required, one of:
+ *                                  //   "tokens" | "requests" |
+ *                                  //   "cost-usd" | "compute-hours"
+ *     period:           string,    // required, one of:
+ *                                  //   "second" | "minute" | "hour" |
+ *                                  //   "day" | "week" | "month"
+ *     limit:            number,    // required, default ceiling (> 0)
+ *     perTenant?:       { [tenantId: string]: number },
+ *     perModel?:        { [model: string]: number },
+ *     perTenantModel?:  { [tenantPipeModel: string]: number },
+ *                                  // key is `tenantId + "|" + model`
+ *     enforcement?:     string,    // "hard" (default) | "soft" | "warn"
+ *     store?:           object,    // { reserve, add, get, reset };
+ *                                  // default in-memory (per-process)
+ *     audit?:           boolean,   // default: true
+ *   }
+ *
+ * @example
+ *   var budget = b.ai.quota.create({
+ *     dimension:  "cost-usd",
+ *     period:     "month",
+ *     limit:      500,
+ *     perTenant:  { "tenant-vip": 5000 },
+ *     enforcement: "hard",
+ *   });
+ *   var r = await budget.consume("tenant-acme", "opus-4", 0.42);
+ *   // → { tenantId: "tenant-acme", model: "opus-4",
+ *   //     dimension: "cost-usd", period: "month", used: 0.42,
+ *   //     limit: 500, remaining: 499.58, allowed: true,
+ *   //     exceeded: false, windowStart: ..., resetsAt: ... }
+ */
+function create(opts) {
+  validateOpts.requireObject(opts, "ai.quota.create", AiQuotaError);
+  validateOpts(opts, [
+    "dimension", "period", "limit", "perTenant", "perModel",
+    "perTenantModel", "enforcement", "store", "audit",
+  ], "ai.quota.create");
+
+  var dimension = opts.dimension;
+  if (DIMENSIONS.indexOf(dimension) === -1) {
+    throw new AiQuotaError("aiQuota/bad-dimension",
+      "ai.quota.create: dimension must be one of " + DIMENSIONS.join(" / ") +
+      " (got " + JSON.stringify(dimension) + ")");
+  }
+
+  var period = opts.period;
+  if (PERIODS.indexOf(period) === -1) {
+    throw new AiQuotaError("aiQuota/bad-period",
+      "ai.quota.create: period must be one of " + PERIODS.join(" / ") +
+      " (got " + JSON.stringify(period) + ")");
+  }
+
+  if (typeof opts.limit !== "number" || !isFinite(opts.limit) || opts.limit <= 0) {
+    throw new AiQuotaError("aiQuota/bad-limit",
+      "ai.quota.create: limit must be a positive finite number");
+  }
+  var defaultLimit = opts.limit;
+
+  var perTenant      = _validateLimitMap(opts.perTenant, "perTenant");
+  var perModel       = _validateLimitMap(opts.perModel, "perModel");
+  var perTenantModel = _validateLimitMap(opts.perTenantModel, "perTenantModel");
+
+  var enforcement = (opts.enforcement == null) ? "hard" : opts.enforcement;
+  if (ENFORCEMENTS.indexOf(enforcement) === -1) {
+    throw new AiQuotaError("aiQuota/bad-enforcement",
+      "ai.quota.create: enforcement must be one of " + ENFORCEMENTS.join(" / ") +
+      " (got " + JSON.stringify(enforcement) + ")");
+  }
+
+  var store = opts.store || _memoryStore();
+  _validateStore(store);
+  var storeIsDefault = !opts.store;
+
+  var auditOn = opts.audit !== false;
+
+  function _limitFor(tenantId, model) {
+    var tmKey = tenantId + "|" + model;
+    if (Object.prototype.hasOwnProperty.call(perTenantModel, tmKey)) return perTenantModel[tmKey];
+    if (Object.prototype.hasOwnProperty.call(perTenant, tenantId))   return perTenant[tenantId];
+    if (Object.prototype.hasOwnProperty.call(perModel, model))       return perModel[model];
+    return defaultLimit;
+  }
+
+  // Counter key — tenant + model percent-encoded so a value
+  // containing the ":" separator cannot collide with another
+  // (tenant, model) pair's budget.
+  function _keyFor(tenantId, model, windowStart) {
+    return "aiq:" + dimension + ":" + period + ":" +
+      encodeURIComponent(tenantId) + ":" + encodeURIComponent(model) + ":" + windowStart;
+  }
+  function _keyPrefixForTenant(tenantId) {
+    return "aiq:" + dimension + ":" + period + ":" + encodeURIComponent(tenantId) + ":";
+  }
+
+  function _nodeId() {
+    try {
+      if (cluster().isClusterMode()) return cluster().currentNodeId();
+    } catch (_e) { /* cluster optional */ }
+    return null;
+  }
+
+  function _emitAudit(action, outcome, metadata) {
+    if (!auditOn) return;
+    try {
+      audit().safeEmit({ action: action, outcome: outcome, metadata: metadata || {} });
+    } catch (_e) { /* audit best-effort — drop-silent */ }
+  }
+
+  function _emitMetric(name, n) {
+    try { observability().safeEvent(name, n || 1, {}); }
+    catch (_e) { /* drop-silent */ }
+  }
+
+  // `mode` is the enforcement actually applied to this call (the
+  // per-call override when present, else the instance default) so the
+  // returned `enforcement` reflects how the call was evaluated.
+  function _result(tenantId, model, used, limit, windowStart, resetsAt, mode, allowed, exceeded) {
+    var remaining = limit - used;
+    return {
+      tenantId:    tenantId,
+      model:       model,
+      dimension:   dimension,
+      period:      period,
+      used:        used,
+      limit:       limit,
+      remaining:   remaining < 0 ? 0 : remaining,
+      allowed:     allowed,
+      exceeded:    exceeded,
+      enforcement: mode,
+      windowStart: windowStart,
+      resetsAt:    resetsAt,
+    };
+  }
+
+  function consume(tenantId, model, amount, consumeOpts) {
+    validateOpts.requireNonEmptyString(tenantId,
+      "ai.quota.consume: tenantId", AiQuotaError, "aiQuota/bad-tenant");
+    validateOpts.requireNonEmptyString(model,
+      "ai.quota.consume: model", AiQuotaError, "aiQuota/bad-model");
+    if (typeof amount !== "number" || !isFinite(amount) || amount < 0) {
+      throw new AiQuotaError("aiQuota/bad-amount",
+        "ai.quota.consume: amount must be a non-negative finite number");
+    }
+    consumeOpts = consumeOpts || {};
+    // Per-call enforcement override lets a single endpoint dial a
+    // softer mode for a trusted internal caller without a second
+    // enforcer; still validated against the allowlist.
+    var mode = (consumeOpts.enforcement == null) ? enforcement : consumeOpts.enforcement;
+    if (ENFORCEMENTS.indexOf(mode) === -1) {
+      throw new AiQuotaError("aiQuota/bad-enforcement",
+        "ai.quota.consume: enforcement override must be one of " + ENFORCEMENTS.join(" / "));
+    }
+
+    var now = Date.now();
+    var windowStart = _windowStartFor(period, now);
+    var resetsAt = _resetsAtFor(period, windowStart);
+    var windowMs = resetsAt - windowStart;
+    var limit = _limitFor(tenantId, model);
+    var key = _keyFor(tenantId, model, windowStart);
+
+    if (mode === "hard") {
+      // Atomic conditional reserve — the store tests current + amount
+      // <= limit and charges only if it fits, as one indivisible
+      // operation. Charging first and refunding the overage (a
+      // read-then-add or add-then-refund shape) would let a concurrent
+      // over-budget call transiently inflate the counter and falsely
+      // deny a smaller call that should fit; the conditional reserve
+      // never charges on refusal, so there is no transient to race.
+      var rv = store.reserve(key, amount, limit, windowMs);
+      if (rv.allowed) {
+        _emitAudit("ai/quota-applied", "allowed", {
+          tenantId: tenantId, model: model, dimension: dimension,
+          period: period, amount: amount, used: rv.used, limit: limit,
+          nodeId: _nodeId(),
+        });
+        _emitMetric("ai.quota.applied", 1);
+        return _result(tenantId, model, rv.used, limit, windowStart, resetsAt, mode, true, false);
+      }
+      _emitAudit("ai/quota-exceeded", "denied", {
+        tenantId: tenantId, model: model, dimension: dimension,
+        period: period, amount: amount, used: rv.used, limit: limit,
+        enforcement: mode, nodeId: _nodeId(),
+      });
+      _emitMetric("ai.quota.exceeded", 1);
+      throw new AiQuotaError("aiQuota/exceeded",
+        "ai.quota.consume: tenant '" + tenantId + "' model '" + model +
+        "' is at " + rv.used + " of " + limit + " " + dimension +
+        " this " + period + "; consuming " + amount + " would exceed the budget — call refused");
+    }
+
+    // soft / warn always charge — the call proceeds regardless of the
+    // ceiling; the mode only changes how the overage is reported.
+    var used = store.add(key, amount, windowMs);
+    if (used > limit) {
+      _emitAudit("ai/quota-exceeded", "allowed", {
+        tenantId: tenantId, model: model, dimension: dimension,
+        period: period, amount: amount, used: used, limit: limit,
+        enforcement: mode, nodeId: _nodeId(),
+      });
+      _emitMetric("ai.quota.exceeded", 1);
+      // soft reports allowed:false so the caller can choose to honor
+      // the ceiling; warn reports allowed:true (advisory only).
+      return _result(tenantId, model, used, limit, windowStart, resetsAt, mode, mode === "warn", true);
+    }
+    _emitAudit("ai/quota-applied", "allowed", {
+      tenantId: tenantId, model: model, dimension: dimension,
+      period: period, amount: amount, used: used, limit: limit,
+      nodeId: _nodeId(),
+    });
+    _emitMetric("ai.quota.applied", 1);
+    return _result(tenantId, model, used, limit, windowStart, resetsAt, mode, true, false);
+  }
+
+  function check(tenantId, model) {
+    validateOpts.requireNonEmptyString(tenantId,
+      "ai.quota.check: tenantId", AiQuotaError, "aiQuota/bad-tenant");
+    validateOpts.requireNonEmptyString(model,
+      "ai.quota.check: model", AiQuotaError, "aiQuota/bad-model");
+    var now = Date.now();
+    var windowStart = _windowStartFor(period, now);
+    var resetsAt = _resetsAtFor(period, windowStart);
+    var limit = _limitFor(tenantId, model);
+    var used = store.get(_keyFor(tenantId, model, windowStart));
+    return _result(tenantId, model, used, limit, windowStart, resetsAt, enforcement, used < limit, used >= limit);
+  }
+
+  function reset(tenantId, model) {
+    var now = Date.now();
+    var windowStart = _windowStartFor(period, now);
+    if (tenantId === undefined) {
+      // Clear everything. The default store supports a full clear;
+      // an external store gets a no-arg reset() if it offers one.
+      if (storeIsDefault) { store._clear(); return; }
+      if (typeof store.reset === "function") { store.reset(); return; }
+      return;
+    }
+    validateOpts.requireNonEmptyString(tenantId,
+      "ai.quota.reset: tenantId", AiQuotaError, "aiQuota/bad-tenant");
+    if (model !== undefined) {
+      validateOpts.requireNonEmptyString(model,
+        "ai.quota.reset: model", AiQuotaError, "aiQuota/bad-model");
+      store.reset(_keyFor(tenantId, model, windowStart));
+      return;
+    }
+    // tenant-wide reset needs key enumeration. The default in-memory
+    // store can scan its own keys; an external store would need a
+    // server-side prefix delete the framework can't portably issue.
+    if (storeIsDefault) {
+      var prefix = _keyPrefixForTenant(tenantId);
+      var keys = store._keysWithPrefix(prefix);
+      for (var i = 0; i < keys.length; i++) store.reset(keys[i]);
+      return;
+    }
+    throw new AiQuotaError("aiQuota/reset-unsupported",
+      "ai.quota.reset: tenant-wide reset with an external store requires " +
+      "an explicit model argument (per-key) or a store-side prefix delete");
+  }
+
+  return {
+    consume:   consume,
+    check:     check,
+    snapshot:  check,
+    reset:     reset,
+    dimension: dimension,
+    period:    period,
+  };
+}
+
+// Per-tenant / per-model / per-tenant-model limit-override maps are
+// validated at config time so a typo (negative cap, NaN) surfaces at
+// boot, not as a silent fall-through to the default ceiling.
+function _validateLimitMap(map, label) {
+  if (map == null) return {};
+  if (typeof map !== "object" || Array.isArray(map)) {
+    throw new AiQuotaError("aiQuota/bad-override",
+      "ai.quota.create: " + label + " must be a plain object { key: limit }");
+  }
+  var keys = Object.keys(map);
+  for (var i = 0; i < keys.length; i++) {
+    var v = map[keys[i]];
+    if (typeof v !== "number" || !isFinite(v) || v <= 0) {
+      throw new AiQuotaError("aiQuota/bad-override",
+        "ai.quota.create: " + label + "['" + keys[i] +
+        "'] must be a positive finite number");
+    }
+  }
+  return map;
+}
+
+function _validateStore(store) {
+  if (!store || typeof store !== "object" ||
+      typeof store.reserve !== "function" ||
+      typeof store.add !== "function" ||
+      typeof store.get !== "function" ||
+      typeof store.reset !== "function") {
+    throw new AiQuotaError("aiQuota/bad-store",
+      "ai.quota.create: store must expose reserve / add / get / reset functions");
+  }
+}
+
+module.exports = {
+  create:       create,
+  DIMENSIONS:   DIMENSIONS,
+  PERIODS:      PERIODS,
+  ENFORCEMENTS: ENFORCEMENTS,
+  AiQuotaError: AiQuotaError,
+};

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@blamejs/core",
-  "version": "0.12.26",
+  "version": "0.12.28",
   "description": "The Node framework that owns its stack.",
   "license": "Apache-2.0",
   "author": "blamejs contributors",

package/sbom.cdx.json

@@ -2,10 +2,10 @@
   "$schema": "http://cyclonedx.org/schema/bom-1.5.schema.json",
   "bomFormat": "CycloneDX",
   "specVersion": "1.5",
-  "serialNumber": "urn:uuid:f16ae992-0fc9-4ad6-b05b-b6dfd629e058",
+  "serialNumber": "urn:uuid:2bff79a1-ab38-4b20-8cdc-37b5e80872a3",
   "version": 1,
   "metadata": {
-    "timestamp": "2026-05-24T10:57:05.537Z",
+    "timestamp": "2026-05-24T16:44:14.267Z",
     "lifecycles": [
       {
         "phase": "build"
@@ -19,14 +19,14 @@
       }
     ],
     "component": {
-      "bom-ref": "@blamejs/core@0.12.26",
+      "bom-ref": "@blamejs/core@0.12.28",
       "type": "application",
       "name": "blamejs",
-      "version": "0.12.26",
+      "version": "0.12.28",
       "scope": "required",
       "author": "blamejs contributors",
       "description": "The Node framework that owns its stack.",
-      "purl": "pkg:npm/%40blamejs/core@0.12.26",
+      "purl": "pkg:npm/%40blamejs/core@0.12.28",
       "properties": [],
       "externalReferences": [
         {
@@ -54,7 +54,7 @@
   "components": [],
   "dependencies": [
     {
-      "ref": "@blamejs/core@0.12.26",
+      "ref": "@blamejs/core@0.12.28",
       "dependsOn": []
     }
   ]