@tangle-network/agent-app 0.47.24 → 0.48.1

package/dist/{chunk-42NBWR4U.js → chunk-SYWEMKZR.js}

@@ -3,7 +3,7 @@ import {
 } from "./chunk-ZSSCVT6H.js";
 import {
   parseJsonObjectBody
-} from "./chunk-3YUI7WNW.js";
+} from "./chunk-IAY2DMKW.js";
 import {
   mentionInputToPart,
   toChatMessageParts
@@ -1046,4 +1046,4 @@ export {
   createChatTurnRoutes,
   streamChatRouteAsSandboxEvents
 };
-//# sourceMappingURL=chunk-42NBWR4U.js.map
+//# sourceMappingURL=chunk-SYWEMKZR.js.map

package/dist/{chunk-QEPLXYOE.js → chunk-VHMZ3KUH.js}

@@ -4,7 +4,7 @@ import {
   boundsIntersect,
   elementAabb,
   scaleForPreset
-} from "./chunk-A2XMTCER.js";
+} from "./chunk-WYHFHTSQ.js";
 
 // src/design-canvas-react/engine/snap.ts
 var KIND_PRIORITY = {
@@ -447,4 +447,4 @@ export {
   buildInsertImageOp,
   DEFAULT_INSERT_TEMPLATES
 };
-//# sourceMappingURL=chunk-QEPLXYOE.js.map
+//# sourceMappingURL=chunk-VHMZ3KUH.js.map

package/dist/{chunk-A2XMTCER.js → chunk-WYHFHTSQ.js}

@@ -1,6 +1,6 @@
 import {
   assertMediaUrl
-} from "./chunk-3YUI7WNW.js";
+} from "./chunk-IAY2DMKW.js";
 
 // src/design-canvas/model.ts
 var SCENE_SCHEMA_VERSION = 1;
@@ -1001,4 +1001,4 @@ export {
   scalePageForChannelPreset,
   bleedAwareExportRect
 };
-//# sourceMappingURL=chunk-A2XMTCER.js.map
+//# sourceMappingURL=chunk-WYHFHTSQ.js.map

package/dist/design-canvas/index.js

@@ -31,12 +31,12 @@ import {
   validateSceneOperation,
   validateSceneOperations,
   validateSlotValue
-} from "../chunk-A2XMTCER.js";
+} from "../chunk-WYHFHTSQ.js";
 import {
   buildScopedMcpServerEntry,
   createMcpToolHandler
 } from "../chunk-6A7MYOUI.js";
-import "../chunk-3YUI7WNW.js";
+import "../chunk-IAY2DMKW.js";
 import "../chunk-TXD5HXLE.js";
 
 // src/design-canvas/operations.ts

package/dist/design-canvas-react/engine.js

@@ -20,7 +20,7 @@ import {
   nudgeDelta,
   resolveExportParams,
   resolveNodeCachePixelRatio
-} from "../chunk-QEPLXYOE.js";
+} from "../chunk-VHMZ3KUH.js";
 import {
   SCENE_COMMAND_HISTORY_LIMIT,
   addElementCommand,
@@ -39,12 +39,12 @@ import {
   setPageGuidesCommand,
   setPagePropsCommand,
   ungroupElementCommand
-} from "../chunk-G4CPEJNN.js";
+} from "../chunk-AKMQFCDL.js";
 import {
   bleedAwareExportBounds,
   scaleForPreset
-} from "../chunk-A2XMTCER.js";
-import "../chunk-3YUI7WNW.js";
+} from "../chunk-WYHFHTSQ.js";
+import "../chunk-IAY2DMKW.js";
 import "../chunk-TXD5HXLE.js";
 export {
   DEFAULT_INSERT_TEMPLATES,

package/dist/design-canvas-react/index.js

@@ -12,7 +12,7 @@ import {
   downloadDataUrl,
   exportDocumentJson,
   exportPageDataUrl
-} from "../chunk-HUYTN7RI.js";
+} from "../chunk-JF423ASS.js";
 import {
   BTN,
   BTN_ACTIVE,
@@ -41,7 +41,7 @@ import {
   buildRulerTicks,
   formatRulerLabel,
   selectTickStep
-} from "../chunk-54KJ54SY.js";
+} from "../chunk-JAT6SOPK.js";
 import "../chunk-Q2QKEV6J.js";
 import "../chunk-ZN5J47UX.js";
 import {
@@ -65,7 +65,7 @@ import {
   nudgeDelta,
   resolveExportParams,
   resolveNodeCachePixelRatio
-} from "../chunk-QEPLXYOE.js";
+} from "../chunk-VHMZ3KUH.js";
 import {
   SCENE_COMMAND_HISTORY_LIMIT,
   addElementCommand,
@@ -84,17 +84,17 @@ import {
   setPageGuidesCommand,
   setPagePropsCommand,
   ungroupElementCommand
-} from "../chunk-G4CPEJNN.js";
+} from "../chunk-AKMQFCDL.js";
 import {
   assertSceneMediaSrc,
   bleedAwareExportBounds,
   scaleForPreset
-} from "../chunk-A2XMTCER.js";
+} from "../chunk-WYHFHTSQ.js";
 import {
   DesignCanvasChromeLazy,
   DesignCanvasLazy
-} from "../chunk-WOZSOPUZ.js";
-import "../chunk-3YUI7WNW.js";
+} from "../chunk-KX2GA3L6.js";
+import "../chunk-IAY2DMKW.js";
 import "../chunk-TXD5HXLE.js";
 
 // src/design-canvas-react/components/CanvasInsertPanel.tsx

package/dist/design-canvas-react/lazy.js

@@ -1,7 +1,7 @@
 import {
   DesignCanvasChromeLazy,
   DesignCanvasLazy
-} from "../chunk-WOZSOPUZ.js";
+} from "../chunk-KX2GA3L6.js";
 export {
   DesignCanvasChromeLazy,
   DesignCanvasLazy

package/dist/eval-campaign/index.d.ts

@@ -27,6 +27,8 @@ import type { JudgeConfig, Scenario } from '@tangle-network/agent-eval/campaign'
 export interface EnsembleJudgeConfig<TArtifact, TScenario extends Scenario, D extends string> {
     /** Judge name — appears in traces and scorecards. */
     name: string;
+    /** Scoring revision for campaign caches. Change it when rubric, model, or callback settings change. */
+    judgeVersion?: JudgeConfig<TArtifact, TScenario>['judgeVersion'];
     /** Stable-ordered rubric dimensions. Drives the `JudgeDimension` list AND the
      *  reducer keys, so a judge that omits a dimension scores it 0 (never silently
      *  dropped). */
@@ -38,11 +40,9 @@ export interface EnsembleJudgeConfig<TArtifact, TScenario extends Scenario, D ex
      * `{ model, perDimension: null }` to record a judge failure WITHOUT killing
      * the ensemble; throw only on an unrecoverable error (the whole rep is then
      * treated as a failed judge).
+     * Use the supplied costLedger, costPhase, and costTags to record paid judge calls.
      */
-    scoreOne: (input: {
-        artifact: TArtifact;
-        scenario: TScenario;
-        signal: AbortSignal;
+    scoreOne: (input: Parameters<JudgeConfig<TArtifact, TScenario>['score']>[0] & {
         rep: number;
     }) => Promise<JudgeVerdict<D>>;
     /** Independent judge calls per artifact, reduced by `aggregateJudgeVerdicts`.

package/dist/eval-campaign/index.js

@@ -115,10 +115,11 @@ function buildEnsembleJudge(cfg) {
   }
   return {
     name: cfg.name,
+    ...cfg.judgeVersion === void 0 ? {} : { judgeVersion: cfg.judgeVersion },
     dimensions: cfg.rubric.map((key) => ({ key, description: cfg.describe?.(key) ?? key })),
-    async score({ artifact, scenario, signal }) {
+    async score(input) {
       const settled = await Promise.allSettled(
-        Array.from({ length: reps }, (_, rep) => cfg.scoreOne({ artifact, scenario, signal, rep }))
+        Array.from({ length: reps }, (_, rep) => cfg.scoreOne({ ...input, rep }))
       );
       const verdicts = settled.map(
         (r, rep) => r.status === "fulfilled" ? r.value : { model: `${cfg.name}-rep${rep}`, perDimension: null, rationale: String(r.reason) }

package/dist/eval-campaign/index.js.map

@@ -1 +1 @@
-{"version":3,"sources":["../../src/eval-campaign/index.ts","../../src/eval-campaign/trust-gate.ts"],"sourcesContent":["/**\n * Eval-campaign — the app-shell's curated surface for a product's\n * self-improvement loop, NOT a reimplementation.\n *\n * The loop ENGINE lives in `@tangle-network/agent-eval` (a peer dependency):\n * `selfImprove` already owns execution, scoring, data separation, release\n * decisions, durable provenance, and hosted ingest. Candidate search is always\n * explicit: pass an official optimization `method`, or pass a caller-owned\n * `SurfaceProposer`.\n *\n * This module adds the one piece `selfImprove` does not own and which every\n * multi-model product re-hand-rolls — the ensemble judge:\n *\n *   {@link buildEnsembleJudge} — turn a per-rubric `scoreOne` into a\n *   `JudgeConfig` that fans out N uncorrelated judge calls and reduces them via\n *   the substrate's `aggregateJudgeVerdicts` (survivor-mean, inter-rater spread,\n *   fail-loud on all-failed). A product writes its rubric + one judge call; the\n *   fan-out, partial-failure handling, and composite are the scaffold's.\n *\n * Everything else is a curated re-export so a product has ONE eval import:\n * `selfImprove` + release policies + optimization methods + their types. See\n * `.claude/skills/eval-campaign/SKILL.md` for the wiring contract.\n */\n\nimport {\n  aggregateJudgeVerdicts,\n  type JudgeVerdict,\n} from '@tangle-network/agent-eval'\nimport type {\n  JudgeConfig,\n  JudgeScore,\n  Scenario,\n} from '@tangle-network/agent-eval/campaign'\n\n/** Config for {@link buildEnsembleJudge}. `D` = the rubric's dimension union. */\nexport interface EnsembleJudgeConfig<TArtifact, TScenario extends Scenario, D extends string> {\n  /** Judge name — appears in traces and scorecards. */\n  name: string\n  /** Stable-ordered rubric dimensions. Drives the `JudgeDimension` list AND the\n   *  reducer keys, so a judge that omits a dimension scores it 0 (never silently\n   *  dropped). */\n  rubric: readonly D[]\n  /**\n   * Score ONE artifact on the rubric → a raw per-dimension verdict. Called\n   * `judgeReps` times per artifact; vary the model by `rep` for an uncorrelated\n   * ensemble (judges that share a base model share its bias). Return\n   * `{ model, perDimension: null }` to record a judge failure WITHOUT killing\n   * the ensemble; throw only on an unrecoverable error (the whole rep is then\n   * treated as a failed judge).\n   */\n  scoreOne: (input: {\n    artifact: TArtifact\n    scenario: TScenario\n    signal: AbortSignal\n    rep: number\n  }) => Promise<JudgeVerdict<D>>\n  /** Independent judge calls per artifact, reduced by `aggregateJudgeVerdicts`.\n   *  Default 1. Raise (with model variety in `scoreOne`) for inter-rater bands. */\n  judgeReps?: number\n  /** Per-dimension composite weights. Default: uniform over `rubric`. A partial\n   *  map selects-and-weights exactly the named dimensions. */\n  weights?: Partial<Record<D, number>>\n  /** Optional human-readable dimension descriptions. Default: the key itself. */\n  describe?: (dim: D) => string\n}\n\n/**\n * Build a `JudgeConfig` whose `score()` fans out `judgeReps` independent\n * `scoreOne` calls and reduces them with the substrate's\n * `aggregateJudgeVerdicts`. A single judge call failing does NOT fail the cell\n * (it is recorded and dropped); only ALL judges failing throws — which the\n * campaign records as a failed cell, never a silent zero.\n *\n * Pass the result straight to `selfImprove({ judge })` (or `runCampaign`).\n */\nexport function buildEnsembleJudge<TArtifact, TScenario extends Scenario, D extends string>(\n  cfg: EnsembleJudgeConfig<TArtifact, TScenario, D>,\n): JudgeConfig<TArtifact, TScenario> {\n  const reps = cfg.judgeReps ?? 1\n  if (reps < 1) {\n    throw new Error(`buildEnsembleJudge: judgeReps must be >= 1 (got ${reps})`)\n  }\n  if (cfg.rubric.length === 0) {\n    throw new Error('buildEnsembleJudge: rubric is empty')\n  }\n  return {\n    name: cfg.name,\n    dimensions: cfg.rubric.map((key) => ({ key, description: cfg.describe?.(key) ?? key })),\n    async score({ artifact, scenario, signal }): Promise<JudgeScore> {\n      const settled = await Promise.allSettled(\n        Array.from({ length: reps }, (_, rep) => cfg.scoreOne({ artifact, scenario, signal, rep })),\n      )\n      const verdicts: JudgeVerdict<D>[] = settled.map((r, rep) =>\n        r.status === 'fulfilled'\n          ? r.value\n          : { model: `${cfg.name}-rep${rep}`, perDimension: null, rationale: String(r.reason) },\n      )\n      // Throws iff EVERY rep failed → the campaign records a failed cell.\n      const agg = aggregateJudgeVerdicts(verdicts, cfg.rubric, cfg.weights)\n      return { composite: agg.composite, dimensions: agg.perDimension, notes: agg.rationale }\n    },\n  }\n}\n\n// ── Trust gate — the after-gate (\"is this result allowed to be believed\") ────\n// One level up from `aggregateJudgeVerdicts`: it audits the raters ACROSS items\n// and reports whether the composites are believable before a lift is reported.\nexport {\n  trustVerdicts,\n  type TrustItem,\n  type TrustThresholds,\n  type TrustVerdict,\n} from './trust-gate'\n\n// ── Curated re-exports — the one eval import for a product loop ──────────────\n// The loop engine + gates + drivers + the ensemble reducer, so a product wires\n// its self-improvement loop from a single module instead of reaching across\n// three agent-eval subpaths. All DOWNWARD imports (agent-app consumes the\n// substrate); the layering rule is preserved.\n\nexport { aggregateJudgeVerdicts } from '@tangle-network/agent-eval'\nexport type {\n  EnsembleAggregate,\n  JudgeVerdict,\n  RunRecord,\n} from '@tangle-network/agent-eval'\nexport {\n  compareOptimizationMethods,\n  defaultProductionGate,\n  externalTextOptimizationMethod,\n  gepaOptimizationMethod,\n  paretoSignificanceGate,\n  runCampaign,\n  skillOptOptimizationMethod,\n} from '@tangle-network/agent-eval/campaign'\nexport type {\n  CampaignResult,\n  CompareOptimizationMethodsOptions,\n  DispatchContext,\n  ExternalTextOptimizationMethodConfig,\n  Gate,\n  GepaOptimizationMethodConfig,\n  JudgeConfig,\n  JudgeDimension,\n  JudgeScore,\n  LabeledScenarioStore,\n  MutableSurface,\n  OptimizationMethod,\n  OptimizationMethodResult,\n  Scenario,\n  SkillOptOptimizationMethodConfig,\n  SurfaceProposer,\n} from '@tangle-network/agent-eval/campaign'\nexport { selfImprove } from '@tangle-network/agent-eval/contract'\nexport type {\n  SelfImproveBudget,\n  SelfImproveOptions,\n  SelfImproveResult,\n} from '@tangle-network/agent-eval/contract'\n","/**\n * Trust gate — decides whether an ensemble's scores are allowed to be BELIEVED,\n * one level up from {@link aggregateJudgeVerdicts} (which only reduces ONE\n * artifact's raters to a composite). A composite is a number; this is the check\n * that the number means anything. It is the code \"Enforced by\" for the\n * measurement-validation skill's after-gate (\"is this result allowed to be\n * believed\").\n *\n * Three checks, each fail-loud and named in `trustReasons`:\n *   (1) inter-rater reliability over the corpus ≥ `irrFloor` — raters that\n *       disagree no better than chance carry no signal to optimize against.\n *   (2) per-item rater spread ≤ `spreadCeiling` — for EACH item, raters must\n *       converge on THAT item.\n *   (3) surviving raters per item ≥ `minSurvivors` — a mean over one or two\n *       raters is an anecdote, not an ensemble.\n *\n * CRITICAL metric semantics — per-item spread is rater disagreement about the\n * SAME item: `max(score) − min(score)` across the raters that scored THAT item\n * (max over its dimensions), never pooled across different items or across the\n * baseline/candidate sides. Pooling reads a genuine quality gap BETWEEN items as\n * \"the raters split\" and so trips the gate exactly when the finding is largest —\n * the failure mode the after-gate exists to prevent. The corpus IRR (check 1)\n * leans on the substrate's `interRaterReliability`, whose expected-disagreement\n * denominator already pools across items, so genuine item-to-item variation\n * RAISES reliability rather than lowering it.\n */\n\nimport {\n  interRaterReliability,\n  type JudgeScore,\n  type JudgeVerdict,\n} from '@tangle-network/agent-eval'\n\n/** One item's raters: the per-judge verdicts {@link aggregateJudgeVerdicts}\n *  reduces, tagged with the item they scored so spread stays within-item. */\nexport interface TrustItem<D extends string = string> {\n  /** Stable item identifier — surfaces in `perItemSpread` and `trustReasons`. */\n  itemId: string\n  /** The raters' verdicts for THIS item (one per judge call). A failed judge\n   *  (`perDimension: null`) is dropped before spread/IRR, never folded as 0. */\n  verdicts: readonly JudgeVerdict<D>[]\n}\n\n/** Thresholds for {@link trustVerdicts}. All overridable; defaults are the\n *  conservative after-gate bar. */\nexport interface TrustThresholds {\n  /** Minimum corpus inter-rater reliability (Krippendorff-style α). Below this\n   *  the raters agree no better than chance. Default 0.2. */\n  irrFloor?: number\n  /** Maximum per-item rater spread (`max − min` over a single item's surviving\n   *  raters, across its dimensions). Above this the raters split ON THAT ITEM.\n   *  Default 0.5. */\n  spreadCeiling?: number\n  /** Minimum surviving (non-failed) raters required per item. Default 3. */\n  minSurvivors?: number\n}\n\n/** Result of the trust gate. `trustworthy` iff every check passed; `trustReasons`\n *  is empty iff `trustworthy`. */\nexport interface TrustVerdict {\n  /** True iff IRR ≥ floor AND every item's spread ≤ ceiling AND every item has\n   *  ≥ `minSurvivors` surviving raters. */\n  trustworthy: boolean\n  /** One entry per FAILED check, each naming its number + the offending value.\n   *  Empty iff `trustworthy`. */\n  trustReasons: string[]\n  /** Corpus inter-rater reliability actually measured (the check-1 value). */\n  interRaterReliability: number\n  /** Per-item spread (`max − min` over surviving raters, max over dimensions),\n   *  keyed by `itemId`. The check-2 input, surfaced for drill-down. */\n  perItemSpread: Record<string, number>\n}\n\nconst DEFAULT_IRR_FLOOR = 0.2\nconst DEFAULT_SPREAD_CEILING = 0.5\nconst DEFAULT_MIN_SURVIVORS = 3\n\n/** Surviving (non-failed) verdicts for an item — those with a real\n *  `perDimension` map. A failed judge carries no scores and is excluded from\n *  every statistic (it is NOT a zero rater). */\nfunction survivors<D extends string>(item: TrustItem<D>): JudgeVerdict<D>[] {\n  return item.verdicts.filter((v) => v.perDimension !== null)\n}\n\n/**\n * Within-item rater spread: for each dimension, `max − min` across the item's\n * surviving raters; the item's spread is the max over its dimensions (the worst\n * dimension the raters split on). Pooled ONLY within this one item — never\n * across items — so a quality gap between items cannot inflate it.\n */\nfunction itemSpread<D extends string>(survivorVerdicts: JudgeVerdict<D>[]): number {\n  if (survivorVerdicts.length < 2) return 0\n  const dims = new Set<string>()\n  for (const v of survivorVerdicts) {\n    for (const d of Object.keys(v.perDimension as Record<string, number>)) dims.add(d)\n  }\n  let worst = 0\n  for (const d of dims) {\n    let min = Infinity\n    let max = -Infinity\n    for (const v of survivorVerdicts) {\n      const score = (v.perDimension as Record<string, number>)[d]\n      if (score === undefined) continue\n      if (score < min) min = score\n      if (score > max) max = score\n    }\n    if (max > -Infinity && max - min > worst) worst = max - min\n  }\n  return worst\n}\n\n/**\n * Decide whether an ensemble's per-item verdicts are trustworthy enough to\n * believe a lift computed from them. Pure: no LLM, no I/O, no clock, no random —\n * the same `items` + `thresholds` always yield the same verdict.\n *\n * Sibling to {@link aggregateJudgeVerdicts}: that reduces ONE item's raters to a\n * composite; this audits the raters ACROSS items and reports whether the\n * composites are believable. Run it on the corpus of held-out items before\n * reporting any lift over their scores.\n *\n * @throws if `items` is empty — an empty corpus has no measurable trust, and a\n *   silent `trustworthy: true` over zero evidence is the exact lie the gate\n *   exists to refuse.\n */\nexport function trustVerdicts<D extends string>(\n  items: readonly TrustItem<D>[],\n  thresholds: TrustThresholds = {},\n): TrustVerdict {\n  if (items.length === 0) {\n    throw new Error('trustVerdicts: items is empty — no evidence to trust')\n  }\n  const irrFloor = thresholds.irrFloor ?? DEFAULT_IRR_FLOOR\n  const spreadCeiling = thresholds.spreadCeiling ?? DEFAULT_SPREAD_CEILING\n  const minSurvivors = thresholds.minSurvivors ?? DEFAULT_MIN_SURVIVORS\n\n  // Rater-major JudgeScore series for the substrate's IRR. Each item's surviving\n  // raters are assigned a stable column index so the same rater across items\n  // lines up; per (item, dimension) one JudgeScore per rater, in item-then-\n  // dimension order — the layout interRaterReliability chunks back into items.\n  const maxRaters = items.reduce((m, it) => Math.max(m, survivors(it).length), 0)\n  const raterSeries: JudgeScore[][] = Array.from({ length: maxRaters }, () => [])\n  const perItemSpread: Record<string, number> = {}\n  const splitItems: Array<{ itemId: string; spread: number }> = []\n  const starvedItems: Array<{ itemId: string; n: number }> = []\n\n  for (const item of items) {\n    const surv = survivors(item)\n    if (surv.length < minSurvivors) starvedItems.push({ itemId: item.itemId, n: surv.length })\n\n    const spread = itemSpread(surv)\n    perItemSpread[item.itemId] = spread\n    if (spread > spreadCeiling) splitItems.push({ itemId: item.itemId, spread })\n\n    if (surv.length >= 2) {\n      const dims = Array.from(\n        new Set(surv.flatMap((v) => Object.keys(v.perDimension as Record<string, number>))),\n      ).sort()\n      surv.forEach((v, raterIdx) => {\n        // raterIdx < surv.length ≤ maxRaters = raterSeries.length, so the column\n        // always exists; the ??= keeps the access provably defined for the type.\n        const column = (raterSeries[raterIdx] ??= [])\n        const pd = v.perDimension as Record<string, number>\n        for (const d of dims) {\n          const score = pd[d]\n          if (score === undefined) continue\n          column.push({\n            judgeName: v.model,\n            dimension: `${item.itemId}::${d}`,\n            score,\n            reasoning: v.rationale ?? '',\n          })\n        }\n      })\n    }\n  }\n\n  const irr = interRaterReliability(raterSeries)\n\n  const trustReasons: string[] = []\n  if (irr < irrFloor) {\n    trustReasons.push(`(1) IRR ${round(irr)} < ${irrFloor}`)\n  }\n  for (const { itemId, spread } of splitItems) {\n    trustReasons.push(`(2) item ${itemId} spread ${round(spread)} > ${spreadCeiling} — raters split`)\n  }\n  for (const { itemId, n } of starvedItems) {\n    trustReasons.push(`(3) item ${itemId}: ${n} surviving raters < ${minSurvivors}`)\n  }\n\n  return {\n    trustworthy: trustReasons.length === 0,\n    trustReasons,\n    interRaterReliability: irr,\n    perItemSpread,\n  }\n}\n\n/** Round to 2 decimals for stable, readable reason strings. */\nfunction round(n: number): number {\n  return Math.round(n * 100) / 100\n}\n"],"mappings":";AAwBA;AAAA,EACE;AAAA,OAEK;;;ACAP;AAAA,EACE;AAAA,OAGK;AA0CP,IAAM,oBAAoB;AAC1B,IAAM,yBAAyB;AAC/B,IAAM,wBAAwB;AAK9B,SAAS,UAA4B,MAAuC;AAC1E,SAAO,KAAK,SAAS,OAAO,CAAC,MAAM,EAAE,iBAAiB,IAAI;AAC5D;AAQA,SAAS,WAA6B,kBAA6C;AACjF,MAAI,iBAAiB,SAAS,EAAG,QAAO;AACxC,QAAM,OAAO,oBAAI,IAAY;AAC7B,aAAW,KAAK,kBAAkB;AAChC,eAAW,KAAK,OAAO,KAAK,EAAE,YAAsC,EAAG,MAAK,IAAI,CAAC;AAAA,EACnF;AACA,MAAI,QAAQ;AACZ,aAAW,KAAK,MAAM;AACpB,QAAI,MAAM;AACV,QAAI,MAAM;AACV,eAAW,KAAK,kBAAkB;AAChC,YAAM,QAAS,EAAE,aAAwC,CAAC;AAC1D,UAAI,UAAU,OAAW;AACzB,UAAI,QAAQ,IAAK,OAAM;AACvB,UAAI,QAAQ,IAAK,OAAM;AAAA,IACzB;AACA,QAAI,MAAM,aAAa,MAAM,MAAM,MAAO,SAAQ,MAAM;AAAA,EAC1D;AACA,SAAO;AACT;AAgBO,SAAS,cACd,OACA,aAA8B,CAAC,GACjB;AACd,MAAI,MAAM,WAAW,GAAG;AACtB,UAAM,IAAI,MAAM,2DAAsD;AAAA,EACxE;AACA,QAAM,WAAW,WAAW,YAAY;AACxC,QAAM,gBAAgB,WAAW,iBAAiB;AAClD,QAAM,eAAe,WAAW,gBAAgB;AAMhD,QAAM,YAAY,MAAM,OAAO,CAAC,GAAG,OAAO,KAAK,IAAI,GAAG,UAAU,EAAE,EAAE,MAAM,GAAG,CAAC;AAC9E,QAAM,cAA8B,MAAM,KAAK,EAAE,QAAQ,UAAU,GAAG,MAAM,CAAC,CAAC;AAC9E,QAAM,gBAAwC,CAAC;AAC/C,QAAM,aAAwD,CAAC;AAC/D,QAAM,eAAqD,CAAC;AAE5D,aAAW,QAAQ,OAAO;AACxB,UAAM,OAAO,UAAU,IAAI;AAC3B,QAAI,KAAK,SAAS,aAAc,cAAa,KAAK,EAAE,QAAQ,KAAK,QAAQ,GAAG,KAAK,OAAO,CAAC;AAEzF,UAAM,SAAS,WAAW,IAAI;AAC9B,kBAAc,KAAK,MAAM,IAAI;AAC7B,QAAI,SAAS,cAAe,YAAW,KAAK,EAAE,QAAQ,KAAK,QAAQ,OAAO,CAAC;AAE3E,QAAI,KAAK,UAAU,GAAG;AACpB,YAAM,OAAO,MAAM;AAAA,QACjB,IAAI,IAAI,KAAK,QAAQ,CAAC,MAAM,OAAO,KAAK,EAAE,YAAsC,CAAC,CAAC;AAAA,MACpF,EAAE,KAAK;AACP,WAAK,QAAQ,CAAC,GAAG,aAAa;AAG5B,cAAM,SAAU,YAAY,QAAQ,MAAM,CAAC;AAC3C,cAAM,KAAK,EAAE;AACb,mBAAW,KAAK,MAAM;AACpB,gBAAM,QAAQ,GAAG,CAAC;AAClB,cAAI,UAAU,OAAW;AACzB,iBAAO,KAAK;AAAA,YACV,WAAW,EAAE;AAAA,YACb,WAAW,GAAG,KAAK,MAAM,KAAK,CAAC;AAAA,YAC/B;AAAA,YACA,WAAW,EAAE,aAAa;AAAA,UAC5B,CAAC;AAAA,QACH;AAAA,MACF,CAAC;AAAA,IACH;AAAA,EACF;AAEA,QAAM,MAAM,sBAAsB,WAAW;AAE7C,QAAM,eAAyB,CAAC;AAChC,MAAI,MAAM,UAAU;AAClB,iBAAa,KAAK,WAAW,MAAM,GAAG,CAAC,MAAM,QAAQ,EAAE;AAAA,EACzD;AACA,aAAW,EAAE,QAAQ,OAAO,KAAK,YAAY;AAC3C,iBAAa,KAAK,YAAY,MAAM,WAAW,MAAM,MAAM,CAAC,MAAM,aAAa,sBAAiB;AAAA,EAClG;AACA,aAAW,EAAE,QAAQ,EAAE,KAAK,cAAc;AACxC,iBAAa,KAAK,YAAY,MAAM,KAAK,CAAC,uBAAuB,YAAY,EAAE;AAAA,EACjF;AAEA,SAAO;AAAA,IACL,aAAa,aAAa,WAAW;AAAA,IACrC;AAAA,IACA,uBAAuB;AAAA,IACvB;AAAA,EACF;AACF;AAGA,SAAS,MAAM,GAAmB;AAChC,SAAO,KAAK,MAAM,IAAI,GAAG,IAAI;AAC/B;;;ADjFA,SAAS,0BAAAA,+BAA8B;AAMvC;AAAA,EACE;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,OACK;AAmBP,SAAS,mBAAmB;AA9ErB,SAAS,mBACd,KACmC;AACnC,QAAM,OAAO,IAAI,aAAa;AAC9B,MAAI,OAAO,GAAG;AACZ,UAAM,IAAI,MAAM,mDAAmD,IAAI,GAAG;AAAA,EAC5E;AACA,MAAI,IAAI,OAAO,WAAW,GAAG;AAC3B,UAAM,IAAI,MAAM,qCAAqC;AAAA,EACvD;AACA,SAAO;AAAA,IACL,MAAM,IAAI;AAAA,IACV,YAAY,IAAI,OAAO,IAAI,CAAC,SAAS,EAAE,KAAK,aAAa,IAAI,WAAW,GAAG,KAAK,IAAI,EAAE;AAAA,IACtF,MAAM,MAAM,EAAE,UAAU,UAAU,OAAO,GAAwB;AAC/D,YAAM,UAAU,MAAM,QAAQ;AAAA,QAC5B,MAAM,KAAK,EAAE,QAAQ,KAAK,GAAG,CAAC,GAAG,QAAQ,IAAI,SAAS,EAAE,UAAU,UAAU,QAAQ,IAAI,CAAC,CAAC;AAAA,MAC5F;AACA,YAAM,WAA8B,QAAQ;AAAA,QAAI,CAAC,GAAG,QAClD,EAAE,WAAW,cACT,EAAE,QACF,EAAE,OAAO,GAAG,IAAI,IAAI,OAAO,GAAG,IAAI,cAAc,MAAM,WAAW,OAAO,EAAE,MAAM,EAAE;AAAA,MACxF;AAEA,YAAM,MAAM,uBAAuB,UAAU,IAAI,QAAQ,IAAI,OAAO;AACpE,aAAO,EAAE,WAAW,IAAI,WAAW,YAAY,IAAI,cAAc,OAAO,IAAI,UAAU;AAAA,IACxF;AAAA,EACF;AACF;","names":["aggregateJudgeVerdicts"]}
+{"version":3,"sources":["../../src/eval-campaign/index.ts","../../src/eval-campaign/trust-gate.ts"],"sourcesContent":["/**\n * Eval-campaign — the app-shell's curated surface for a product's\n * self-improvement loop, NOT a reimplementation.\n *\n * The loop ENGINE lives in `@tangle-network/agent-eval` (a peer dependency):\n * `selfImprove` already owns execution, scoring, data separation, release\n * decisions, durable provenance, and hosted ingest. Candidate search is always\n * explicit: pass an official optimization `method`, or pass a caller-owned\n * `SurfaceProposer`.\n *\n * This module adds the one piece `selfImprove` does not own and which every\n * multi-model product re-hand-rolls — the ensemble judge:\n *\n *   {@link buildEnsembleJudge} — turn a per-rubric `scoreOne` into a\n *   `JudgeConfig` that fans out N uncorrelated judge calls and reduces them via\n *   the substrate's `aggregateJudgeVerdicts` (survivor-mean, inter-rater spread,\n *   fail-loud on all-failed). A product writes its rubric + one judge call; the\n *   fan-out, partial-failure handling, and composite are the scaffold's.\n *\n * Everything else is a curated re-export so a product has ONE eval import:\n * `selfImprove` + release policies + optimization methods + their types. See\n * `.claude/skills/eval-campaign/SKILL.md` for the wiring contract.\n */\n\nimport {\n  aggregateJudgeVerdicts,\n  type JudgeVerdict,\n} from '@tangle-network/agent-eval'\nimport type {\n  JudgeConfig,\n  JudgeScore,\n  Scenario,\n} from '@tangle-network/agent-eval/campaign'\n\n/** Config for {@link buildEnsembleJudge}. `D` = the rubric's dimension union. */\nexport interface EnsembleJudgeConfig<TArtifact, TScenario extends Scenario, D extends string> {\n  /** Judge name — appears in traces and scorecards. */\n  name: string\n  /** Scoring revision for campaign caches. Change it when rubric, model, or callback settings change. */\n  judgeVersion?: JudgeConfig<TArtifact, TScenario>['judgeVersion']\n  /** Stable-ordered rubric dimensions. Drives the `JudgeDimension` list AND the\n   *  reducer keys, so a judge that omits a dimension scores it 0 (never silently\n   *  dropped). */\n  rubric: readonly D[]\n  /**\n   * Score ONE artifact on the rubric → a raw per-dimension verdict. Called\n   * `judgeReps` times per artifact; vary the model by `rep` for an uncorrelated\n   * ensemble (judges that share a base model share its bias). Return\n   * `{ model, perDimension: null }` to record a judge failure WITHOUT killing\n   * the ensemble; throw only on an unrecoverable error (the whole rep is then\n   * treated as a failed judge).\n   * Use the supplied costLedger, costPhase, and costTags to record paid judge calls.\n   */\n  scoreOne: (input: Parameters<JudgeConfig<TArtifact, TScenario>['score']>[0] & {\n    rep: number\n  }) => Promise<JudgeVerdict<D>>\n  /** Independent judge calls per artifact, reduced by `aggregateJudgeVerdicts`.\n   *  Default 1. Raise (with model variety in `scoreOne`) for inter-rater bands. */\n  judgeReps?: number\n  /** Per-dimension composite weights. Default: uniform over `rubric`. A partial\n   *  map selects-and-weights exactly the named dimensions. */\n  weights?: Partial<Record<D, number>>\n  /** Optional human-readable dimension descriptions. Default: the key itself. */\n  describe?: (dim: D) => string\n}\n\n/**\n * Build a `JudgeConfig` whose `score()` fans out `judgeReps` independent\n * `scoreOne` calls and reduces them with the substrate's\n * `aggregateJudgeVerdicts`. A single judge call failing does NOT fail the cell\n * (it is recorded and dropped); only ALL judges failing throws — which the\n * campaign records as a failed cell, never a silent zero.\n *\n * Pass the result straight to `selfImprove({ judge })` (or `runCampaign`).\n */\nexport function buildEnsembleJudge<TArtifact, TScenario extends Scenario, D extends string>(\n  cfg: EnsembleJudgeConfig<TArtifact, TScenario, D>,\n): JudgeConfig<TArtifact, TScenario> {\n  const reps = cfg.judgeReps ?? 1\n  if (reps < 1) {\n    throw new Error(`buildEnsembleJudge: judgeReps must be >= 1 (got ${reps})`)\n  }\n  if (cfg.rubric.length === 0) {\n    throw new Error('buildEnsembleJudge: rubric is empty')\n  }\n  return {\n    name: cfg.name,\n    ...(cfg.judgeVersion === undefined ? {} : { judgeVersion: cfg.judgeVersion }),\n    dimensions: cfg.rubric.map((key) => ({ key, description: cfg.describe?.(key) ?? key })),\n    async score(input): Promise<JudgeScore> {\n      const settled = await Promise.allSettled(\n        Array.from({ length: reps }, (_, rep) => cfg.scoreOne({ ...input, rep })),\n      )\n      const verdicts: JudgeVerdict<D>[] = settled.map((r, rep) =>\n        r.status === 'fulfilled'\n          ? r.value\n          : { model: `${cfg.name}-rep${rep}`, perDimension: null, rationale: String(r.reason) },\n      )\n      // Throws iff EVERY rep failed → the campaign records a failed cell.\n      const agg = aggregateJudgeVerdicts(verdicts, cfg.rubric, cfg.weights)\n      return { composite: agg.composite, dimensions: agg.perDimension, notes: agg.rationale }\n    },\n  }\n}\n\n// Agreement across items is distinct from evaluator accuracy against independent controls.\n// Consumers can opt into auditEvaluator from agent-eval/meta-eval for that separate evidence.\nexport {\n  trustVerdicts,\n  type TrustItem,\n  type TrustThresholds,\n  type TrustVerdict,\n} from './trust-gate'\n\n// ── Curated re-exports — the one eval import for a product loop ──────────────\n// The loop engine + gates + drivers + the ensemble reducer, so a product wires\n// its self-improvement loop from a single module instead of reaching across\n// three agent-eval subpaths. All DOWNWARD imports (agent-app consumes the\n// substrate); the layering rule is preserved.\n\nexport { aggregateJudgeVerdicts } from '@tangle-network/agent-eval'\nexport type {\n  EnsembleAggregate,\n  JudgeVerdict,\n  RunRecord,\n} from '@tangle-network/agent-eval'\nexport {\n  compareOptimizationMethods,\n  defaultProductionGate,\n  externalTextOptimizationMethod,\n  gepaOptimizationMethod,\n  paretoSignificanceGate,\n  runCampaign,\n  skillOptOptimizationMethod,\n} from '@tangle-network/agent-eval/campaign'\nexport type {\n  CampaignResult,\n  CompareOptimizationMethodsOptions,\n  DispatchContext,\n  ExternalTextOptimizationMethodConfig,\n  Gate,\n  GepaOptimizationMethodConfig,\n  JudgeConfig,\n  JudgeDimension,\n  JudgeScore,\n  LabeledScenarioStore,\n  MutableSurface,\n  OptimizationMethod,\n  OptimizationMethodResult,\n  Scenario,\n  SkillOptOptimizationMethodConfig,\n  SurfaceProposer,\n} from '@tangle-network/agent-eval/campaign'\nexport { selfImprove } from '@tangle-network/agent-eval/contract'\nexport type {\n  SelfImproveBudget,\n  SelfImproveOptions,\n  SelfImproveResult,\n} from '@tangle-network/agent-eval/contract'\n","/**\n * Summarize rater agreement, within-item spread, and surviving-judge coverage.\n * Consumers choose the thresholds that a trustworthy result must satisfy.\n * Agreement does not measure evaluator errors against independent controls.\n * Use agent-eval/meta-eval's auditEvaluator when the consumer needs that separate evidence.\n * Spread stays within each item so differences in task quality cannot mimic rater disagreement.\n */\n\nimport {\n  interRaterReliability,\n  type DimensionJudgeScore,\n  type JudgeVerdict,\n} from '@tangle-network/agent-eval'\n\n/** One item's raters: the per-judge verdicts {@link aggregateJudgeVerdicts}\n *  reduces, tagged with the item they scored so spread stays within-item. */\nexport interface TrustItem<D extends string = string> {\n  /** Stable item identifier — surfaces in `perItemSpread` and `trustReasons`. */\n  itemId: string\n  /** The raters' verdicts for THIS item (one per judge call). A failed judge\n   *  (`perDimension: null`) is dropped before spread/IRR, never folded as 0. */\n  verdicts: readonly JudgeVerdict<D>[]\n}\n\n/** Configurable agreement and coverage thresholds for {@link trustVerdicts}. */\nexport interface TrustThresholds {\n  /** Minimum corpus inter-rater reliability (Krippendorff-style α). Default 0.2. */\n  irrFloor?: number\n  /** Maximum per-item rater spread (`max − min` over a single item's surviving\n   *  raters, across its dimensions). Above this the raters split ON THAT ITEM.\n   *  Default 0.5. */\n  spreadCeiling?: number\n  /** Minimum surviving (non-failed) raters required per item. Default 3. */\n  minSurvivors?: number\n}\n\n/** Result of the trust gate. `trustworthy` iff every check passed; `trustReasons`\n *  is empty iff `trustworthy`. */\nexport interface TrustVerdict {\n  /** True iff IRR ≥ floor AND every item's spread ≤ ceiling AND every item has\n   *  ≥ `minSurvivors` surviving raters. */\n  trustworthy: boolean\n  /** One entry per FAILED check, each naming its number + the offending value.\n   *  Empty iff `trustworthy`. */\n  trustReasons: string[]\n  /** Corpus inter-rater reliability actually measured (the check-1 value). */\n  interRaterReliability: number\n  /** Per-item spread (`max − min` over surviving raters, max over dimensions),\n   *  keyed by `itemId`. The check-2 input, surfaced for drill-down. */\n  perItemSpread: Record<string, number>\n}\n\nconst DEFAULT_IRR_FLOOR = 0.2\nconst DEFAULT_SPREAD_CEILING = 0.5\nconst DEFAULT_MIN_SURVIVORS = 3\n\n/** Surviving (non-failed) verdicts for an item — those with a real\n *  `perDimension` map. A failed judge carries no scores and is excluded from\n *  every statistic (it is NOT a zero rater). */\nfunction survivors<D extends string>(item: TrustItem<D>): JudgeVerdict<D>[] {\n  return item.verdicts.filter((v) => v.perDimension !== null)\n}\n\n/**\n * Within-item rater spread: for each dimension, `max − min` across the item's\n * surviving raters; the item's spread is the max over its dimensions (the worst\n * dimension the raters split on). Pooled ONLY within this one item — never\n * across items — so a quality gap between items cannot inflate it.\n */\nfunction itemSpread<D extends string>(survivorVerdicts: JudgeVerdict<D>[]): number {\n  if (survivorVerdicts.length < 2) return 0\n  const dims = new Set<string>()\n  for (const v of survivorVerdicts) {\n    for (const d of Object.keys(v.perDimension as Record<string, number>)) dims.add(d)\n  }\n  let worst = 0\n  for (const d of dims) {\n    let min = Infinity\n    let max = -Infinity\n    for (const v of survivorVerdicts) {\n      const score = (v.perDimension as Record<string, number>)[d]\n      if (score === undefined) continue\n      if (score < min) min = score\n      if (score > max) max = score\n    }\n    if (max > -Infinity && max - min > worst) worst = max - min\n  }\n  return worst\n}\n\n/**\n * Check an ensemble against its configured agreement and coverage thresholds. Pure: no LLM, no I/O, no clock, no random —\n * the same `items` + `thresholds` always yield the same verdict.\n *\n * Sibling to {@link aggregateJudgeVerdicts}: that reduces ONE item's raters to a\n * composite; this summarizes agreement across the supplied items.\n * A passing result does not establish evaluator accuracy or authorize a release.\n *\n * @throws if `items` is empty — an empty corpus has no measurable trust, and a\n *   silent `trustworthy: true` over zero evidence is the exact lie the gate\n *   exists to refuse.\n */\nexport function trustVerdicts<D extends string>(\n  items: readonly TrustItem<D>[],\n  thresholds: TrustThresholds = {},\n): TrustVerdict {\n  if (items.length === 0) {\n    throw new Error('trustVerdicts: items is empty — no evidence to trust')\n  }\n  const irrFloor = thresholds.irrFloor ?? DEFAULT_IRR_FLOOR\n  const spreadCeiling = thresholds.spreadCeiling ?? DEFAULT_SPREAD_CEILING\n  const minSurvivors = thresholds.minSurvivors ?? DEFAULT_MIN_SURVIVORS\n\n  // Rater-major JudgeScore series for the substrate's IRR. Each item's surviving\n  // raters are assigned a stable column index so the same rater across items\n  // lines up; per (item, dimension) one JudgeScore per rater, in item-then-\n  // dimension order — the layout interRaterReliability chunks back into items.\n  const maxRaters = items.reduce((m, it) => Math.max(m, survivors(it).length), 0)\n  const raterSeries: DimensionJudgeScore[][] = Array.from({ length: maxRaters }, () => [])\n  const perItemSpread: Record<string, number> = {}\n  const splitItems: Array<{ itemId: string; spread: number }> = []\n  const starvedItems: Array<{ itemId: string; n: number }> = []\n\n  for (const item of items) {\n    const surv = survivors(item)\n    if (surv.length < minSurvivors) starvedItems.push({ itemId: item.itemId, n: surv.length })\n\n    const spread = itemSpread(surv)\n    perItemSpread[item.itemId] = spread\n    if (spread > spreadCeiling) splitItems.push({ itemId: item.itemId, spread })\n\n    if (surv.length >= 2) {\n      const dims = Array.from(\n        new Set(surv.flatMap((v) => Object.keys(v.perDimension as Record<string, number>))),\n      ).sort()\n      surv.forEach((v, raterIdx) => {\n        // raterIdx < surv.length ≤ maxRaters = raterSeries.length, so the column\n        // always exists; the ??= keeps the access provably defined for the type.\n        const column = (raterSeries[raterIdx] ??= [])\n        const pd = v.perDimension as Record<string, number>\n        for (const d of dims) {\n          const score = pd[d]\n          if (score === undefined) continue\n          column.push({\n            judgeName: v.model,\n            dimension: `${item.itemId}::${d}`,\n            score,\n            reasoning: v.rationale ?? '',\n          })\n        }\n      })\n    }\n  }\n\n  const irr = interRaterReliability(raterSeries)\n\n  const trustReasons: string[] = []\n  if (irr < irrFloor) {\n    trustReasons.push(`(1) IRR ${round(irr)} < ${irrFloor}`)\n  }\n  for (const { itemId, spread } of splitItems) {\n    trustReasons.push(`(2) item ${itemId} spread ${round(spread)} > ${spreadCeiling} — raters split`)\n  }\n  for (const { itemId, n } of starvedItems) {\n    trustReasons.push(`(3) item ${itemId}: ${n} surviving raters < ${minSurvivors}`)\n  }\n\n  return {\n    trustworthy: trustReasons.length === 0,\n    trustReasons,\n    interRaterReliability: irr,\n    perItemSpread,\n  }\n}\n\n/** Round to 2 decimals for stable, readable reason strings. */\nfunction round(n: number): number {\n  return Math.round(n * 100) / 100\n}\n"],"mappings":";AAwBA;AAAA,EACE;AAAA,OAEK;;;ACnBP;AAAA,EACE;AAAA,OAGK;AAwCP,IAAM,oBAAoB;AAC1B,IAAM,yBAAyB;AAC/B,IAAM,wBAAwB;AAK9B,SAAS,UAA4B,MAAuC;AAC1E,SAAO,KAAK,SAAS,OAAO,CAAC,MAAM,EAAE,iBAAiB,IAAI;AAC5D;AAQA,SAAS,WAA6B,kBAA6C;AACjF,MAAI,iBAAiB,SAAS,EAAG,QAAO;AACxC,QAAM,OAAO,oBAAI,IAAY;AAC7B,aAAW,KAAK,kBAAkB;AAChC,eAAW,KAAK,OAAO,KAAK,EAAE,YAAsC,EAAG,MAAK,IAAI,CAAC;AAAA,EACnF;AACA,MAAI,QAAQ;AACZ,aAAW,KAAK,MAAM;AACpB,QAAI,MAAM;AACV,QAAI,MAAM;AACV,eAAW,KAAK,kBAAkB;AAChC,YAAM,QAAS,EAAE,aAAwC,CAAC;AAC1D,UAAI,UAAU,OAAW;AACzB,UAAI,QAAQ,IAAK,OAAM;AACvB,UAAI,QAAQ,IAAK,OAAM;AAAA,IACzB;AACA,QAAI,MAAM,aAAa,MAAM,MAAM,MAAO,SAAQ,MAAM;AAAA,EAC1D;AACA,SAAO;AACT;AAcO,SAAS,cACd,OACA,aAA8B,CAAC,GACjB;AACd,MAAI,MAAM,WAAW,GAAG;AACtB,UAAM,IAAI,MAAM,2DAAsD;AAAA,EACxE;AACA,QAAM,WAAW,WAAW,YAAY;AACxC,QAAM,gBAAgB,WAAW,iBAAiB;AAClD,QAAM,eAAe,WAAW,gBAAgB;AAMhD,QAAM,YAAY,MAAM,OAAO,CAAC,GAAG,OAAO,KAAK,IAAI,GAAG,UAAU,EAAE,EAAE,MAAM,GAAG,CAAC;AAC9E,QAAM,cAAuC,MAAM,KAAK,EAAE,QAAQ,UAAU,GAAG,MAAM,CAAC,CAAC;AACvF,QAAM,gBAAwC,CAAC;AAC/C,QAAM,aAAwD,CAAC;AAC/D,QAAM,eAAqD,CAAC;AAE5D,aAAW,QAAQ,OAAO;AACxB,UAAM,OAAO,UAAU,IAAI;AAC3B,QAAI,KAAK,SAAS,aAAc,cAAa,KAAK,EAAE,QAAQ,KAAK,QAAQ,GAAG,KAAK,OAAO,CAAC;AAEzF,UAAM,SAAS,WAAW,IAAI;AAC9B,kBAAc,KAAK,MAAM,IAAI;AAC7B,QAAI,SAAS,cAAe,YAAW,KAAK,EAAE,QAAQ,KAAK,QAAQ,OAAO,CAAC;AAE3E,QAAI,KAAK,UAAU,GAAG;AACpB,YAAM,OAAO,MAAM;AAAA,QACjB,IAAI,IAAI,KAAK,QAAQ,CAAC,MAAM,OAAO,KAAK,EAAE,YAAsC,CAAC,CAAC;AAAA,MACpF,EAAE,KAAK;AACP,WAAK,QAAQ,CAAC,GAAG,aAAa;AAG5B,cAAM,SAAU,YAAY,QAAQ,MAAM,CAAC;AAC3C,cAAM,KAAK,EAAE;AACb,mBAAW,KAAK,MAAM;AACpB,gBAAM,QAAQ,GAAG,CAAC;AAClB,cAAI,UAAU,OAAW;AACzB,iBAAO,KAAK;AAAA,YACV,WAAW,EAAE;AAAA,YACb,WAAW,GAAG,KAAK,MAAM,KAAK,CAAC;AAAA,YAC/B;AAAA,YACA,WAAW,EAAE,aAAa;AAAA,UAC5B,CAAC;AAAA,QACH;AAAA,MACF,CAAC;AAAA,IACH;AAAA,EACF;AAEA,QAAM,MAAM,sBAAsB,WAAW;AAE7C,QAAM,eAAyB,CAAC;AAChC,MAAI,MAAM,UAAU;AAClB,iBAAa,KAAK,WAAW,MAAM,GAAG,CAAC,MAAM,QAAQ,EAAE;AAAA,EACzD;AACA,aAAW,EAAE,QAAQ,OAAO,KAAK,YAAY;AAC3C,iBAAa,KAAK,YAAY,MAAM,WAAW,MAAM,MAAM,CAAC,MAAM,aAAa,sBAAiB;AAAA,EAClG;AACA,aAAW,EAAE,QAAQ,EAAE,KAAK,cAAc;AACxC,iBAAa,KAAK,YAAY,MAAM,KAAK,CAAC,uBAAuB,YAAY,EAAE;AAAA,EACjF;AAEA,SAAO;AAAA,IACL,aAAa,aAAa,WAAW;AAAA,IACrC;AAAA,IACA,uBAAuB;AAAA,IACvB;AAAA,EACF;AACF;AAGA,SAAS,MAAM,GAAmB;AAChC,SAAO,KAAK,MAAM,IAAI,GAAG,IAAI;AAC/B;;;AD1DA,SAAS,0BAAAA,+BAA8B;AAMvC;AAAA,EACE;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,OACK;AAmBP,SAAS,mBAAmB;AA9ErB,SAAS,mBACd,KACmC;AACnC,QAAM,OAAO,IAAI,aAAa;AAC9B,MAAI,OAAO,GAAG;AACZ,UAAM,IAAI,MAAM,mDAAmD,IAAI,GAAG;AAAA,EAC5E;AACA,MAAI,IAAI,OAAO,WAAW,GAAG;AAC3B,UAAM,IAAI,MAAM,qCAAqC;AAAA,EACvD;AACA,SAAO;AAAA,IACL,MAAM,IAAI;AAAA,IACV,GAAI,IAAI,iBAAiB,SAAY,CAAC,IAAI,EAAE,cAAc,IAAI,aAAa;AAAA,IAC3E,YAAY,IAAI,OAAO,IAAI,CAAC,SAAS,EAAE,KAAK,aAAa,IAAI,WAAW,GAAG,KAAK,IAAI,EAAE;AAAA,IACtF,MAAM,MAAM,OAA4B;AACtC,YAAM,UAAU,MAAM,QAAQ;AAAA,QAC5B,MAAM,KAAK,EAAE,QAAQ,KAAK,GAAG,CAAC,GAAG,QAAQ,IAAI,SAAS,EAAE,GAAG,OAAO,IAAI,CAAC,CAAC;AAAA,MAC1E;AACA,YAAM,WAA8B,QAAQ;AAAA,QAAI,CAAC,GAAG,QAClD,EAAE,WAAW,cACT,EAAE,QACF,EAAE,OAAO,GAAG,IAAI,IAAI,OAAO,GAAG,IAAI,cAAc,MAAM,WAAW,OAAO,EAAE,MAAM,EAAE;AAAA,MACxF;AAEA,YAAM,MAAM,uBAAuB,UAAU,IAAI,QAAQ,IAAI,OAAO;AACpE,aAAO,EAAE,WAAW,IAAI,WAAW,YAAY,IAAI,cAAc,OAAO,IAAI,UAAU;AAAA,IACxF;AAAA,EACF;AACF;","names":["aggregateJudgeVerdicts"]}

package/dist/eval-campaign/trust-gate.d.ts

@@ -1,28 +1,9 @@
 /**
- * Trust gate — decides whether an ensemble's scores are allowed to be BELIEVED,
- * one level up from {@link aggregateJudgeVerdicts} (which only reduces ONE
- * artifact's raters to a composite). A composite is a number; this is the check
- * that the number means anything. It is the code "Enforced by" for the
- * measurement-validation skill's after-gate ("is this result allowed to be
- * believed").
- *
- * Three checks, each fail-loud and named in `trustReasons`:
- *   (1) inter-rater reliability over the corpus ≥ `irrFloor` — raters that
- *       disagree no better than chance carry no signal to optimize against.
- *   (2) per-item rater spread ≤ `spreadCeiling` — for EACH item, raters must
- *       converge on THAT item.
- *   (3) surviving raters per item ≥ `minSurvivors` — a mean over one or two
- *       raters is an anecdote, not an ensemble.
- *
- * CRITICAL metric semantics — per-item spread is rater disagreement about the
- * SAME item: `max(score) − min(score)` across the raters that scored THAT item
- * (max over its dimensions), never pooled across different items or across the
- * baseline/candidate sides. Pooling reads a genuine quality gap BETWEEN items as
- * "the raters split" and so trips the gate exactly when the finding is largest —
- * the failure mode the after-gate exists to prevent. The corpus IRR (check 1)
- * leans on the substrate's `interRaterReliability`, whose expected-disagreement
- * denominator already pools across items, so genuine item-to-item variation
- * RAISES reliability rather than lowering it.
+ * Summarize rater agreement, within-item spread, and surviving-judge coverage.
+ * Consumers choose the thresholds that a trustworthy result must satisfy.
+ * Agreement does not measure evaluator errors against independent controls.
+ * Use agent-eval/meta-eval's auditEvaluator when the consumer needs that separate evidence.
+ * Spread stays within each item so differences in task quality cannot mimic rater disagreement.
  */
 import { type JudgeVerdict } from '@tangle-network/agent-eval';
 /** One item's raters: the per-judge verdicts {@link aggregateJudgeVerdicts}
@@ -34,11 +15,9 @@ export interface TrustItem<D extends string = string> {
      *  (`perDimension: null`) is dropped before spread/IRR, never folded as 0. */
     verdicts: readonly JudgeVerdict<D>[];
 }
-/** Thresholds for {@link trustVerdicts}. All overridable; defaults are the
- *  conservative after-gate bar. */
+/** Configurable agreement and coverage thresholds for {@link trustVerdicts}. */
 export interface TrustThresholds {
-    /** Minimum corpus inter-rater reliability (Krippendorff-style α). Below this
-     *  the raters agree no better than chance. Default 0.2. */
+    /** Minimum corpus inter-rater reliability (Krippendorff-style α). Default 0.2. */
     irrFloor?: number;
     /** Maximum per-item rater spread (`max − min` over a single item's surviving
      *  raters, across its dimensions). Above this the raters split ON THAT ITEM.
@@ -63,14 +42,12 @@ export interface TrustVerdict {
     perItemSpread: Record<string, number>;
 }
 /**
- * Decide whether an ensemble's per-item verdicts are trustworthy enough to
- * believe a lift computed from them. Pure: no LLM, no I/O, no clock, no random —
+ * Check an ensemble against its configured agreement and coverage thresholds. Pure: no LLM, no I/O, no clock, no random —
  * the same `items` + `thresholds` always yield the same verdict.
  *
  * Sibling to {@link aggregateJudgeVerdicts}: that reduces ONE item's raters to a
- * composite; this audits the raters ACROSS items and reports whether the
- * composites are believable. Run it on the corpus of held-out items before
- * reporting any lift over their scores.
+ * composite; this summarizes agreement across the supplied items.
+ * A passing result does not establish evaluator accuracy or authorize a release.
  *
  * @throws if `items` is empty — an empty corpus has no measurable trust, and a
  *   silent `trustworthy: true` over zero evidence is the exact lie the gate

package/dist/platform/index.js

@@ -13,12 +13,12 @@ import {
   resolveTangleSsoAccount,
   signSessionCookieValue,
   verifySignedSsoState
-} from "../chunk-TXM4SS46.js";
+} from "../chunk-CSMJM7YT.js";
 import {
   resolveTangleDevOrUserKey,
   resolveTangleExecutionEnvironment
 } from "../chunk-JML7WKWU.js";
-import "../chunk-3YUI7WNW.js";
+import "../chunk-IAY2DMKW.js";
 import "../chunk-TXD5HXLE.js";
 
 // src/platform/hub.ts

package/dist/public-consultation/index.js

@@ -1,9 +1,9 @@
 import {
   createChatTurnRoutes,
   streamChatRouteAsSandboxEvents
-} from "../chunk-42NBWR4U.js";
+} from "../chunk-SYWEMKZR.js";
 import "../chunk-ZSSCVT6H.js";
-import "../chunk-3YUI7WNW.js";
+import "../chunk-IAY2DMKW.js";
 import "../chunk-4PZE7XAM.js";
 import "../chunk-ZVEEWGDK.js";
 import "../chunk-X47R2IVO.js";

package/dist/sequences/index.js

@@ -18,7 +18,7 @@ import {
 } from "../chunk-6A7MYOUI.js";
 import {
   assertMediaUrl
-} from "../chunk-3YUI7WNW.js";
+} from "../chunk-IAY2DMKW.js";
 import "../chunk-TXD5HXLE.js";
 
 // src/sequences/operations.ts

package/dist/web/core.d.ts

@@ -0,0 +1,84 @@
+/**
+ * Web-boundary utilities every agent app's routes hand-roll: JSON body parsing
+ * + narrowing, request-context extraction (real client IP behind Cloudflare),
+ * a KV-backed sliding-window rate limiter, the free-route budget policy built
+ * on it, and security response headers. Pure mechanism — no DB, no domain. The
+ * KV is a structural interface so this needs no `@cloudflare/workers-types`
+ * dependency.
+ */
+export * from './rate-limit';
+export * from './free-route-limit';
+export type JsonObject = Record<string, unknown>;
+/** Parse + object-narrow a Request body. `[body, null]` on success, `[null,
+ *  errorResponse]` on a non-object body (callers `if (err) return err`). */
+export declare function parseJsonObjectBody(request: Request): Promise<[JsonObject, null] | [null, Response]>;
+/** Narrow one required string field, 400 if missing/empty. */
+export declare function requireString(body: JsonObject, field: string): string | Response;
+/** Define the context of a request including IP address, user agent, timestamp, and request ID */
+export interface RequestContext {
+    ipAddress: string;
+    userAgent: string;
+    timestamp: string;
+    requestId: string;
+}
+/** Extract request context for audit trails. Uses `CF-Connecting-IP` for the
+ *  real client IP behind Cloudflare. */
+export declare function extractRequestContext(request: Request): RequestContext;
+/** Define options for configuring cookie attributes and behavior */
+export interface CookieOptions {
+    name: string;
+    /** Default '/'. */
+    path?: string;
+    /** Default true. */
+    httpOnly?: boolean;
+    /** Adds the `Secure` attribute. Default false. */
+    secure?: boolean;
+    /** Default 'Lax'. */
+    sameSite?: 'Lax' | 'Strict' | 'None';
+    maxAgeSeconds?: number;
+}
+/** Serialize a Set-Cookie header value: `name=encodeURIComponent(value)` plus
+ *  attributes in Path / HttpOnly / SameSite / Max-Age / Secure order.
+ *  Throws on `SameSite=None` without `secure` — browsers silently drop that
+ *  combination, which would otherwise fail invisibly. */
+export declare function serializeCookie(value: string, opts: CookieOptions): string;
+/** Set-Cookie header value that deletes the cookie (empty value, Max-Age=0). */
+export declare function clearCookieHeader(opts: Omit<CookieOptions, 'maxAgeSeconds'>): string;
+/** Read + decode one cookie from a Cookie request header; null when absent. */
+export declare function readCookieValue(cookieHeader: string | null, name: string): string | null;
+/** Define options for configuring security-related HTTP headers including disclaimers and retention labels */
+export interface SecurityHeaderOptions {
+    /** Product disclaimer (e.g. "AI-powered tool. Not legal advice."). Omitted if absent. */
+    disclaimer?: string;
+    /** Data-retention label (e.g. "7-years"). Omitted if absent. */
+    retention?: string;
+    /** Extra headers to set. */
+    extra?: Record<string, string>;
+}
+/** Canonical generic response headers used by {@link addSecurityHeaders}.
+ * Exported so static-asset hosts can apply the same policy without copying
+ * values that silently drift from Worker/API responses. */
+export declare const STANDARD_SECURITY_HEADERS: Readonly<{
+    readonly 'Strict-Transport-Security': 'max-age=31536000; includeSubDomains; preload';
+    readonly 'X-Content-Type-Options': 'nosniff';
+    readonly 'X-Frame-Options': 'SAMEORIGIN';
+    readonly 'Referrer-Policy': 'same-origin';
+    readonly 'X-XSS-Protection': '1; mode=block';
+}>;
+/** Set standard security headers on a response (HSTS, nosniff, frame-options,
+ *  referrer-policy, XSS) + optional product disclaimer/retention. The security
+ *  set is generic; the disclaimer/retention are the product's. */
+export declare function addSecurityHeaders(response: Response, opts?: SecurityHeaderOptions): Response;
+/**
+ * Canonical media-reference boundary shared by every surface that persists a
+ * media url (sequences clips, design-canvas image/video src). The ONE rule:
+ * remote `http(s)` or a rooted `/api/` path are allowed; everything else is
+ * rejected, with a named reason for known-bad local/inline schemes so the
+ * thrown message is actionable for an LLM planner. The url is trimmed before
+ * the scheme check so leading whitespace cannot smuggle a rejected scheme past
+ * a naive `startsWith`.
+ *
+ * @param what - noun for the error message (e.g. 'media url', 'src').
+ */
+export declare function assertMediaUrl(url: string, what?: string): void;
+export { isWorkspaceFileExportable } from './file-export';

package/dist/web/index.d.ts

@@ -1,84 +1,3 @@
-/**
- * Web-boundary utilities every agent app's routes hand-roll: JSON body parsing
- * + narrowing, request-context extraction (real client IP behind Cloudflare),
- * a KV-backed sliding-window rate limiter, the free-route budget policy built
- * on it, and security response headers. Pure mechanism — no DB, no domain. The
- * KV is a structural interface so this needs no `@cloudflare/workers-types`
- * dependency.
- */
-export * from './rate-limit';
-export * from './free-route-limit';
-export type JsonObject = Record<string, unknown>;
-/** Parse + object-narrow a Request body. `[body, null]` on success, `[null,
- *  errorResponse]` on a non-object body (callers `if (err) return err`). */
-export declare function parseJsonObjectBody(request: Request): Promise<[JsonObject, null] | [null, Response]>;
-/** Narrow one required string field, 400 if missing/empty. */
-export declare function requireString(body: JsonObject, field: string): string | Response;
-/** Define the context of a request including IP address, user agent, timestamp, and request ID */
-export interface RequestContext {
-    ipAddress: string;
-    userAgent: string;
-    timestamp: string;
-    requestId: string;
-}
-/** Extract request context for audit trails. Uses `CF-Connecting-IP` for the
- *  real client IP behind Cloudflare. */
-export declare function extractRequestContext(request: Request): RequestContext;
-/** Define options for configuring cookie attributes and behavior */
-export interface CookieOptions {
-    name: string;
-    /** Default '/'. */
-    path?: string;
-    /** Default true. */
-    httpOnly?: boolean;
-    /** Adds the `Secure` attribute. Default false. */
-    secure?: boolean;
-    /** Default 'Lax'. */
-    sameSite?: 'Lax' | 'Strict' | 'None';
-    maxAgeSeconds?: number;
-}
-/** Serialize a Set-Cookie header value: `name=encodeURIComponent(value)` plus
- *  attributes in Path / HttpOnly / SameSite / Max-Age / Secure order.
- *  Throws on `SameSite=None` without `secure` — browsers silently drop that
- *  combination, which would otherwise fail invisibly. */
-export declare function serializeCookie(value: string, opts: CookieOptions): string;
-/** Set-Cookie header value that deletes the cookie (empty value, Max-Age=0). */
-export declare function clearCookieHeader(opts: Omit<CookieOptions, 'maxAgeSeconds'>): string;
-/** Read + decode one cookie from a Cookie request header; null when absent. */
-export declare function readCookieValue(cookieHeader: string | null, name: string): string | null;
-/** Define options for configuring security-related HTTP headers including disclaimers and retention labels */
-export interface SecurityHeaderOptions {
-    /** Product disclaimer (e.g. "AI-powered tool. Not legal advice."). Omitted if absent. */
-    disclaimer?: string;
-    /** Data-retention label (e.g. "7-years"). Omitted if absent. */
-    retention?: string;
-    /** Extra headers to set. */
-    extra?: Record<string, string>;
-}
-/** Canonical generic response headers used by {@link addSecurityHeaders}.
- * Exported so static-asset hosts can apply the same policy without copying
- * values that silently drift from Worker/API responses. */
-export declare const STANDARD_SECURITY_HEADERS: Readonly<{
-    readonly 'Strict-Transport-Security': 'max-age=31536000; includeSubDomains; preload';
-    readonly 'X-Content-Type-Options': 'nosniff';
-    readonly 'X-Frame-Options': 'SAMEORIGIN';
-    readonly 'Referrer-Policy': 'same-origin';
-    readonly 'X-XSS-Protection': '1; mode=block';
-}>;
-/** Set standard security headers on a response (HSTS, nosniff, frame-options,
- *  referrer-policy, XSS) + optional product disclaimer/retention. The security
- *  set is generic; the disclaimer/retention are the product's. */
-export declare function addSecurityHeaders(response: Response, opts?: SecurityHeaderOptions): Response;
-/**
- * Canonical media-reference boundary shared by every surface that persists a
- * media url (sequences clips, design-canvas image/video src). The ONE rule:
- * remote `http(s)` or a rooted `/api/` path are allowed; everything else is
- * rejected, with a named reason for known-bad local/inline schemes so the
- * thrown message is actionable for an LLM planner. The url is trimmed before
- * the scheme check so leading whitespace cannot smuggle a rejected scheme past
- * a naive `startsWith`.
- *
- * @param what - noun for the error message (e.g. 'media url', 'src').
- */
-export declare function assertMediaUrl(url: string, what?: string): void;
-export { isWorkspaceFileExportable } from './file-export';
+/** Browser-safe application-boundary helpers. No React or execution-engine peers. */
+export * from './core';
+export * from './message-groups';

package/dist/web/index.js

@@ -10,12 +10,13 @@ import {
   clearCookieHeader,
   extractRequestContext,
   freeRouteLimitResponse,
+  groupConversationMessages,
   parseJsonObjectBody,
   readCookieValue,
   requireString,
   serializeCookie,
   withFreeRouteLimit
-} from "../chunk-3YUI7WNW.js";
+} from "../chunk-IAY2DMKW.js";
 import {
   isWorkspaceFileExportable
 } from "../chunk-TXD5HXLE.js";
@@ -31,6 +32,7 @@ export {
   clearCookieHeader,
   extractRequestContext,
   freeRouteLimitResponse,
+  groupConversationMessages,
   isWorkspaceFileExportable,
   parseJsonObjectBody,
   readCookieValue,

package/dist/web/message-groups.d.ts

@@ -0,0 +1,21 @@
+/** Renderer-neutral attribution. Not a transcript merger or an execution state machine. */
+export interface ConversationGroupItem {
+    id?: string | number;
+    kind?: string;
+    role?: string;
+    /** A different assistant/persona must never inherit the previous speaker's label. */
+    speakerId?: string;
+    /** Distinct threads must not be grouped if their rows share a viewport. */
+    conversationId?: string;
+}
+export type GroupedConversationItem<T> = T & {
+    isContinuation?: boolean;
+    groupId?: string;
+};
+/**
+ * Annotate assistant rows until an actual user message, speaker or conversation change.
+ * Tool/progress notices do not interrupt the group. The first visible assistant is
+ * always labeled, including when a renderer has paged earlier history away.
+ * Pass display order; stored content, IDs, timestamps and input objects are untouched.
+ */
+export declare function groupConversationMessages<T extends ConversationGroupItem>(items?: readonly T[]): GroupedConversationItem<T>[];