@tangle-network/agent-eval 0.42.0 → 0.43.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -2,8 +2,8 @@ import {
2
2
  BENCHMARK_SPLIT_SEED,
3
3
  deterministicSplit,
4
4
  routing_exports
5
- } from "../chunk-6QDKWHLS.js";
6
- import "../chunk-PZ5AY32C.js";
5
+ } from "../chunk-MHQPVHXU.js";
6
+ import "../chunk-NSBPE2FW.js";
7
7
  export {
8
8
  BENCHMARK_SPLIT_SEED,
9
9
  deterministicSplit,
@@ -10,7 +10,7 @@ import {
10
10
  TraceEmitter
11
11
  } from "../chunk-TVVP3ZZQ.js";
12
12
  import "../chunk-QYJT52YW.js";
13
- import "../chunk-PZ5AY32C.js";
13
+ import "../chunk-NSBPE2FW.js";
14
14
 
15
15
  // src/builder-eval/builder-session.ts
16
16
  var BuilderSession = class {
@@ -245,6 +245,42 @@ declare class FsLabeledScenarioStore implements LabeledScenarioStore {
245
245
  private pathForSource;
246
246
  }
247
247
 
248
+ /**
249
+ * @experimental
250
+ *
251
+ * `CampaignStorage` — the filesystem seam `runCampaign` writes through
252
+ * (run/cell dirs, the resumability cache, per-cell artifacts, trace spans).
253
+ *
254
+ * The default (`fsCampaignStorage`) is the Node filesystem — identical
255
+ * behavior to the inline `node:fs` calls it replaces, so existing CLI
256
+ * consumers are unaffected. `inMemoryCampaignStorage` keeps everything in a
257
+ * `Map`, so the substrate runs in environments WITHOUT a filesystem
258
+ * (Cloudflare Workers, Deno Deploy, other edge runtimes) — the campaign
259
+ * still produces its `CampaignResult` (cells + aggregates) in memory;
260
+ * artifacts/traces simply aren't persisted to disk.
261
+ *
262
+ * Paths are opaque keys to the in-memory adapter — it does not parse them,
263
+ * so the same `join(...)`-built paths work unchanged across both adapters.
264
+ */
265
+ interface CampaignStorage {
266
+ /** Ensure a directory exists (recursive). No-op for in-memory. */
267
+ ensureDir(dir: string): void;
268
+ /** Does this path exist (as a written file or an ensured dir)? */
269
+ exists(path: string): boolean;
270
+ /** Read a UTF-8 file; `undefined` when missing or unreadable. */
271
+ read(path: string): string | undefined;
272
+ /** Write a file (string or bytes). Parent dir is assumed ensured. */
273
+ write(path: string, content: string | Uint8Array): void;
274
+ }
275
+ /** Node-filesystem storage — the default. Lazily requires `node:fs` so the
276
+ * module imports cleanly in non-Node runtimes (where the caller passes
277
+ * `inMemoryCampaignStorage` instead and never constructs this). */
278
+ declare function fsCampaignStorage(): CampaignStorage;
279
+ /** In-memory storage for filesystem-less runtimes. Artifacts + trace spans
280
+ * live in a `Map` for the duration of the run; the `CampaignResult` is
281
+ * fully populated, but nothing is persisted to disk. */
282
+ declare function inMemoryCampaignStorage(): CampaignStorage;
283
+
248
284
  /**
249
285
  * @experimental
250
286
  *
@@ -290,6 +326,12 @@ interface RunCampaignOptions<TScenario extends Scenario, TArtifact> {
290
326
  now?: () => Date;
291
327
  /** Test seam — override per-cell trace writer factory. */
292
328
  buildTraceWriter?: (cellId: string, dir: string) => CampaignTraceWriter;
329
+ /** Storage backend for run/cell dirs, the resumability cache, artifacts,
330
+ * and trace spans. Default: the Node filesystem (`fsCampaignStorage`).
331
+ * Pass `inMemoryCampaignStorage()` to run in a filesystem-less runtime
332
+ * (Cloudflare Workers, Deno, edge) — the `CampaignResult` is still
333
+ * produced; artifacts/traces just aren't persisted to disk. */
334
+ storage?: CampaignStorage;
293
335
  }
294
336
  declare function runCampaign<TScenario extends Scenario, TArtifact>(opts: RunCampaignOptions<TScenario, TArtifact>): Promise<CampaignResult<TArtifact, TScenario>>;
295
337
 
@@ -476,4 +518,4 @@ declare function gitWorktreeAdapter(opts: GitWorktreeAdapterOptions): WorktreeAd
476
518
  * as a ref under the adapter's worktree dir. */
477
519
  declare function resolveWorktreePath(surface: CodeSurface, worktreeDir?: string): string;
478
520
 
479
- export { CampaignResult, CampaignTraceWriter, CodeSurface, type DefaultProductionGateOptions, DispatchFn, type EvolutionaryDriverOptions, FsLabeledScenarioStore, type FsLabeledScenarioStoreOptions, Gate, GateResult, GenerationRecord, type GepaDriverOptions, type GitWorktreeAdapterOptions, type HeldOutGateOptions, ImprovementDriver, JudgeConfig, LabeledScenarioRecord, LabeledScenarioSampleArgs, LabeledScenarioStore, LabeledScenarioStoreError, LabeledScenarioWrite, MutableSurface, Mutator, type OpenAutoPrOptions, type OpenAutoPrResult, type RunCampaignOptions, type RunEvalOptions, type RunImprovementLoopOptions, type RunImprovementLoopResult, type RunOptimizationOptions, type RunOptimizationResult, Scenario, type Worktree, type WorktreeAdapter, WorktreeAdapterError, composeGate, defaultProductionGate, evolutionaryDriver, gepaDriver, gitWorktreeAdapter, heldOutGate, openAutoPr, resolveWorktreePath, runCampaign, runEval, runImprovementLoop, runOptimization, surfaceHash };
521
+ export { CampaignResult, type CampaignStorage, CampaignTraceWriter, CodeSurface, type DefaultProductionGateOptions, DispatchFn, type EvolutionaryDriverOptions, FsLabeledScenarioStore, type FsLabeledScenarioStoreOptions, Gate, GateResult, GenerationRecord, type GepaDriverOptions, type GitWorktreeAdapterOptions, type HeldOutGateOptions, ImprovementDriver, JudgeConfig, LabeledScenarioRecord, LabeledScenarioSampleArgs, LabeledScenarioStore, LabeledScenarioStoreError, LabeledScenarioWrite, MutableSurface, Mutator, type OpenAutoPrOptions, type OpenAutoPrResult, type RunCampaignOptions, type RunEvalOptions, type RunImprovementLoopOptions, type RunImprovementLoopResult, type RunOptimizationOptions, type RunOptimizationResult, Scenario, type Worktree, type WorktreeAdapter, WorktreeAdapterError, composeGate, defaultProductionGate, evolutionaryDriver, fsCampaignStorage, gepaDriver, gitWorktreeAdapter, heldOutGate, inMemoryCampaignStorage, openAutoPr, resolveWorktreePath, runCampaign, runEval, runImprovementLoop, runOptimization, surfaceHash };
@@ -1,6 +1,8 @@
1
1
  import {
2
+ fsCampaignStorage,
3
+ inMemoryCampaignStorage,
2
4
  runCampaign
3
- } from "../chunk-YNMCYUWT.js";
5
+ } from "../chunk-RXK7FXLV.js";
4
6
  import {
5
7
  buildReflectionPrompt,
6
8
  parseReflectionResponse,
@@ -17,7 +19,7 @@ import {
17
19
  } from "../chunk-VXNVVBZO.js";
18
20
  import "../chunk-PC4UYEBM.js";
19
21
  import "../chunk-QYJT52YW.js";
20
- import "../chunk-PZ5AY32C.js";
22
+ import "../chunk-NSBPE2FW.js";
21
23
 
22
24
  // src/campaign/auto-pr.ts
23
25
  import { execSync } from "child_process";
@@ -753,7 +755,7 @@ async function runImprovementLoop(opts) {
753
755
  throw new Error("runImprovementLoop: autoOnPromote='pr' requires ghOwner + ghRepo.");
754
756
  }
755
757
  const optimization = await runOptimization(opts);
756
- const { runCampaign: runCampaign2 } = await import("../run-campaign-KEJK5KFT.js");
758
+ const { runCampaign: runCampaign2 } = await import("../run-campaign-GNDO66B4.js");
757
759
  const baselineOnHoldout = await runCampaign2({
758
760
  ...opts,
759
761
  scenarios: opts.holdoutScenarios,
@@ -893,9 +895,11 @@ export {
893
895
  composeGate,
894
896
  defaultProductionGate,
895
897
  evolutionaryDriver,
898
+ fsCampaignStorage,
896
899
  gepaDriver,
897
900
  gitWorktreeAdapter,
898
901
  heldOutGate,
902
+ inMemoryCampaignStorage,
899
903
  openAutoPr,
900
904
  resolveWorktreePath,
901
905
  runCampaign,
@@ -1 +1 @@
1
- {"version":3,"sources":["../../src/campaign/auto-pr.ts","../../src/campaign/drivers/evolutionary.ts","../../src/campaign/drivers/gepa.ts","../../src/campaign/gates/compose.ts","../../src/campaign/gates/default-production-gate.ts","../../src/campaign/gates/heldout-gate.ts","../../src/campaign/labeled-store/fs-adapter.ts","../../src/campaign/presets/run-eval.ts","../../src/campaign/presets/run-optimization.ts","../../src/campaign/presets/run-improvement-loop.ts","../../src/campaign/worktree/index.ts"],"sourcesContent":["/**\n * @experimental\n *\n * `openAutoPr` — thin shell-out helper for the `runImprovementLoop` preset's\n * `autoOnPromote: 'pr'` mode. Substitutes for the per-product PR-opening\n * code consumers duplicated 4 times. The PR body includes the campaign's\n * manifest hash, gate verdict, and scorecard summary so reviewers can see\n * exactly what was promoted + why.\n *\n * NOT a deploy mechanism — this only OPENS a PR. The human reviews + merges.\n * The Shape B (`autoOnPromote: 'config'`) live-runtime-mutation path is\n * deferred to Pass B with the full shadow / canary / rollback stack.\n */\n\nimport { execSync } from 'node:child_process'\nimport { writeFileSync } from 'node:fs'\nimport { tmpdir } from 'node:os'\nimport { join } from 'node:path'\nimport type { CampaignResult, GateResult, Scenario } from './types'\n\nexport interface OpenAutoPrOptions<TArtifact, TScenario extends Scenario> {\n /** Campaign result to attach to the PR. */\n result: CampaignResult<TArtifact, TScenario>\n /** Gate verdict explaining the promotion. Substrate refuses to open a PR\n * when `gate.decision !== 'ship'` — fails loud. */\n gate: GateResult\n /** Promoted surface diff — typically the new system prompt addendum or\n * full profile diff. Substrate writes it as the PR body. */\n promotedDiff: string\n /** GH owner/repo target (e.g., `tangle-network/gtm-agent`). */\n ghOwner: string\n ghRepo: string\n /** Branch name for the PR. Default `auto/<manifestHash[:12]>`. */\n branch?: string\n /** PR title. Default includes manifest hash. */\n title?: string\n /** Whether to actually open the PR or just dry-run. Default reads\n * `GH_AUTO_PR_TOKEN` env — present = open, absent = dry-run. */\n dryRun?: boolean\n /** Test seam — substitute `gh pr create` invocation. */\n ghExec?: (args: string[]) => { stdout: string; stderr: string; status: number }\n}\n\nexport interface OpenAutoPrResult {\n opened: boolean\n prUrl?: string\n dryRun: boolean\n reason: string\n}\n\nexport function openAutoPr<TArtifact, TScenario extends Scenario>(\n options: OpenAutoPrOptions<TArtifact, TScenario>,\n): OpenAutoPrResult {\n if (options.gate.decision !== 'ship') {\n return {\n opened: false,\n dryRun: false,\n reason: `gate verdict was \"${options.gate.decision}\" — refusing to open PR`,\n }\n }\n\n const dryRun = options.dryRun ?? !process.env.GH_AUTO_PR_TOKEN\n const branch = options.branch ?? `auto/${options.result.manifestHash.slice(0, 12)}`\n const title =\n options.title ?? `auto: campaign ${options.result.manifestHash.slice(0, 8)} promoted by gate`\n\n const body = renderPrBody(options.result, options.gate, options.promotedDiff)\n const bodyPath = join(tmpdir(), `auto-pr-body-${Date.now()}.md`)\n writeFileSync(bodyPath, body)\n\n if (dryRun) {\n return {\n opened: false,\n dryRun: true,\n reason: `dry-run (GH_AUTO_PR_TOKEN not set). Would create PR on ${options.ghOwner}/${options.ghRepo} branch ${branch}. Body at ${bodyPath}.`,\n }\n }\n\n const ghExec = options.ghExec ?? defaultGhExec\n const result = ghExec([\n 'pr',\n 'create',\n '--repo',\n `${options.ghOwner}/${options.ghRepo}`,\n '--head',\n branch,\n '--title',\n title,\n '--body-file',\n bodyPath,\n ])\n if (result.status !== 0) {\n return {\n opened: false,\n dryRun: false,\n reason: `gh pr create failed (exit ${result.status}): ${result.stderr.slice(0, 400)}`,\n }\n }\n const prUrl = result.stdout.trim()\n return { opened: true, prUrl, dryRun: false, reason: 'PR opened' }\n}\n\nfunction renderPrBody<TArtifact, TScenario extends Scenario>(\n result: CampaignResult<TArtifact, TScenario>,\n gate: GateResult,\n diff: string,\n): string {\n const lines: string[] = []\n lines.push(`## Automated promotion by \\`runImprovementLoop\\``)\n lines.push('')\n lines.push(`**Manifest**: \\`${result.manifestHash}\\``)\n lines.push(`**Seed**: ${result.seed}`)\n lines.push(`**Duration**: ${Math.round(result.durationMs / 1000)}s`)\n lines.push(\n `**Cells**: executed ${result.aggregates.cellsExecuted}, cached ${result.aggregates.cellsCached}, skipped ${result.aggregates.cellsSkipped}, failed ${result.aggregates.cellsFailed}`,\n )\n lines.push(`**Total spend**: $${result.aggregates.totalCostUsd.toFixed(2)}`)\n lines.push('')\n lines.push(`### Gate verdict: \\`${gate.decision}\\``)\n lines.push('')\n for (const reason of gate.reasons) lines.push(`- ${reason}`)\n if (gate.delta !== undefined) lines.push(`- delta: ${gate.delta.toFixed(3)}`)\n lines.push('')\n lines.push('### Contributing gates')\n lines.push('')\n lines.push('| gate | passed | detail |')\n lines.push('|---|---|---|')\n for (const c of gate.contributingGates) {\n const detail =\n typeof c.detail === 'object'\n ? JSON.stringify(c.detail).slice(0, 80)\n : String(c.detail).slice(0, 80)\n lines.push(`| ${c.name} | ${c.passed ? '✓' : '✗'} | ${detail} |`)\n }\n lines.push('')\n lines.push('### Promoted surface')\n lines.push('')\n lines.push('```diff')\n lines.push(diff.slice(0, 8000))\n lines.push('```')\n lines.push('')\n lines.push('### By-judge aggregates')\n lines.push('')\n lines.push('| judge | mean | ci95 | n |')\n lines.push('|---|---|---|---|')\n for (const [name, agg] of Object.entries(result.aggregates.byJudge)) {\n lines.push(\n `| ${name} | ${agg.mean.toFixed(3)} | [${agg.ci95[0].toFixed(3)}, ${agg.ci95[1].toFixed(3)}] | ${agg.n} |`,\n )\n }\n return lines.join('\\n')\n}\n\nfunction defaultGhExec(args: string[]): { stdout: string; stderr: string; status: number } {\n try {\n const stdout = execSync(`gh ${args.map(quoteArg).join(' ')}`, {\n env: { ...process.env, GH_TOKEN: process.env.GH_AUTO_PR_TOKEN ?? process.env.GH_TOKEN ?? '' },\n stdio: ['ignore', 'pipe', 'pipe'],\n }).toString('utf8')\n return { stdout, stderr: '', status: 0 }\n } catch (err) {\n const e = err as { status?: number; stderr?: Buffer; stdout?: Buffer }\n return {\n stdout: e.stdout?.toString('utf8') ?? '',\n stderr: e.stderr?.toString('utf8') ?? '',\n status: e.status ?? 1,\n }\n }\n}\n\nfunction quoteArg(arg: string): string {\n if (/^[a-zA-Z0-9_/\\-:.@]+$/.test(arg)) return arg\n return `\"${arg.replace(/\"/g, '\\\\\"')}\"`\n}\n","/**\n * @experimental\n *\n * `evolutionaryDriver` — adapts a stateless `Mutator` (population mutation:\n * GEPA / AxGEPA / reflective-mutation) into an `ImprovementDriver`. This is\n * the evolutionary strategy: each generation, mutate the current best surface\n * into N candidates, measure, select. No generation memory beyond the current\n * surface; the loop body handles ranking + promotion.\n *\n * The reflective alternative is agent-runtime's `improvementDriver` with a\n * `reflectiveGenerator` / `agenticGenerator`: it reasons over the report +\n * trace findings to propose targeted edits rather than blind mutations. Both\n * conform to `ImprovementDriver`; the improvement loop is identical regardless\n * of which drives it.\n */\n\nimport type { ImprovementDriver, Mutator } from '../types'\n\nexport interface EvolutionaryDriverOptions<TFindings = unknown> {\n mutator: Mutator<TFindings>\n /** External findings fed to the mutator each generation. Default: []. */\n findings?: TFindings[]\n}\n\nexport function evolutionaryDriver<TFindings = unknown>(\n opts: EvolutionaryDriverOptions<TFindings>,\n): ImprovementDriver<TFindings> {\n return {\n kind: `evolutionary:${opts.mutator.kind}`,\n async propose({ currentSurface, findings, populationSize, signal }) {\n return opts.mutator.mutate({\n findings: findings.length > 0 ? findings : (opts.findings ?? []),\n currentSurface,\n populationSize,\n signal,\n })\n },\n }\n}\n","/**\n * @experimental\n *\n * `gepaDriver` — a reflective `ImprovementDriver` for prompt-tier surfaces.\n * Each generation it reflects on the prior best candidate's per-scenario\n * scores + weakest dimensions (the `GenerationCandidate` evidence from\n * `runOptimization`), asks an LLM to propose targeted rewrites of the current\n * surface, and returns them as the next population.\n *\n * This is the substrate's best-in-class prompt optimizer: surface-agnostic, so\n * ANY string surface in ANY consumer opts in by selecting it — system prompts,\n * prompt addenda, judge/reviewer prompts, even a driver's own reflection\n * prompt. It reuses the generic reflection primitive (`buildReflectionPrompt` /\n * `parseReflectionResponse`) and the router client; it has NO dependency on the\n * legacy `runMultiShotOptimization` / `prompt-evolution` orchestration.\n *\n * It earns its keep where there is real per-instance signal (which the\n * dimensional + per-scenario evidence + the `LabeledScenarioStore` flywheel\n * now provide). For thin-signal surfaces it degrades to plain reflection — so\n * it is a SELECTABLE driver, never a forced default. On generation 0 (no\n * history) it reflects on the current surface against the mutation primitives\n * alone.\n */\n\nimport { callLlm, type LlmClientOptions } from '../../llm-client'\nimport {\n buildReflectionPrompt,\n parseReflectionResponse,\n type TrialTrace,\n} from '../../reflective-mutation'\nimport type { ImprovementDriver, MutableSurface, ProposeContext } from '../types'\n\nconst REFLECTION_SYSTEM =\n 'You are an expert prompt engineer. Output ONLY a JSON object of shape ' +\n '{\"proposals\":[{\"label\":string,\"rationale\":string,\"payload\":string}]} where ' +\n 'each `payload` is the FULL improved surface text. No prose outside the JSON.'\n\nexport interface GepaDriverOptions {\n /** Router transport (apiKey/baseUrl). */\n llm: LlmClientOptions\n /** Model that performs the reflection. */\n model: string\n /** What is being optimized — appears in the reflection prompt for orientation. */\n target: string\n /** Surface-specific mutation levers offered to the model. */\n mutationPrimitives?: string[]\n /** Top/bottom scenarios surfaced as evidence each generation. Default 3. */\n evidenceK?: number\n /** Reflection sampling temperature. Default 0.7. */\n temperature?: number\n /** Reflection max tokens. Default 6000. */\n maxTokens?: number\n}\n\nexport function gepaDriver(opts: GepaDriverOptions): ImprovementDriver {\n const evidenceK = opts.evidenceK ?? 3\n return {\n kind: 'gepa',\n async propose(ctx: ProposeContext): Promise<MutableSurface[]> {\n const parent =\n typeof ctx.currentSurface === 'string'\n ? ctx.currentSurface\n : JSON.stringify(ctx.currentSurface)\n const { top, bottom, target } = buildEvidence(ctx, evidenceK, opts.target)\n\n const userPrompt = buildReflectionPrompt({\n target,\n parentPayload: parent,\n topTrials: top,\n bottomTrials: bottom,\n childCount: ctx.populationSize,\n mutationPrimitives: opts.mutationPrimitives,\n })\n\n const result = await callLlm(\n {\n model: opts.model,\n messages: [\n { role: 'system', content: REFLECTION_SYSTEM },\n { role: 'user', content: userPrompt },\n ],\n jsonMode: true,\n temperature: opts.temperature ?? 0.7,\n maxTokens: opts.maxTokens ?? 6000,\n },\n opts.llm,\n )\n\n const proposals = parseReflectionResponse(result.content, ctx.populationSize)\n const out: MutableSurface[] = []\n for (const proposal of proposals) {\n const text = typeof proposal.payload === 'string' ? proposal.payload.trim() : ''\n if (text && text !== parent && !out.includes(text)) out.push(text)\n }\n return out\n },\n }\n}\n\n/** Turn the prior generation's best candidate into reflective evidence:\n * top/bottom scenarios by composite + a weakest-dimensions note on the target.\n * Empty on generation 0 — the model reflects on the surface alone. */\nfunction buildEvidence(\n ctx: ProposeContext,\n evidenceK: number,\n baseTarget: string,\n): { top: TrialTrace[]; bottom: TrialTrace[]; target: string } {\n const last = ctx.history.at(-1)\n if (!last || last.candidates.length === 0) {\n return { top: [], bottom: [], target: baseTarget }\n }\n const best = [...last.candidates].sort((a, b) => b.composite - a.composite)[0]\n if (!best) return { top: [], bottom: [], target: baseTarget }\n\n const byScore = [...best.scenarios].sort((a, b) => b.composite - a.composite)\n const toTrace = (s: { scenarioId: string; composite: number }): TrialTrace => ({\n id: s.scenarioId,\n score: s.composite,\n })\n const top = byScore.slice(0, evidenceK).map(toTrace)\n const bottom = byScore.slice(-evidenceK).reverse().map(toTrace)\n\n const weakest = Object.entries(best.dimensions)\n .sort((a, b) => a[1] - b[1])\n .slice(0, 3)\n .map(([dim, value]) => `${dim} (${value.toFixed(2)})`)\n const target =\n weakest.length > 0 ? `${baseTarget} — weakest dimensions: ${weakest.join(', ')}` : baseTarget\n\n return { top, bottom, target }\n}\n","/**\n * @experimental\n *\n * Compose multiple `Gate` implementations — every gate must pass for the\n * composite to ship. Closes the alignment reviewer's \"default-only\n * heldOutGate + costGate would happily promote a reward-hacked prompt\"\n * concern by making safety gates first-class composable defaults.\n */\n\nimport type { Gate, GateContext, GateDecision, GateResult, Scenario } from '../types'\n\n/** Compose gates — all must `ship` for the composite to `ship`. First\n * non-ship verdict short-circuits the composite verdict, but ALL gates run\n * (so the result records every gate's reason — useful for diagnostics). */\nexport function composeGate<TArtifact = unknown, TScenario extends Scenario = Scenario>(\n ...gates: Array<Gate<TArtifact, TScenario>>\n): Gate<TArtifact, TScenario> {\n if (gates.length === 0) {\n throw new Error('composeGate requires at least one gate')\n }\n return {\n name: `composed(${gates.map((g) => g.name).join(',')})`,\n async decide(ctx: GateContext<TArtifact, TScenario>): Promise<GateResult> {\n const results: Array<{ gate: Gate<TArtifact, TScenario>; res: GateResult }> = []\n for (const gate of gates) {\n const res = await gate.decide(ctx)\n results.push({ gate, res })\n }\n\n // Substrate-wide verdict policy:\n // - all 'ship' → 'ship'\n // - any 'arch_ceiling' → 'arch_ceiling' (architectural ceiling beats other holds)\n // - any 'model_ceiling' → 'model_ceiling'\n // - any 'hold' → 'hold'\n // - else 'need_more_work'\n const decisions = results.map((r) => r.res.decision)\n const overall: GateDecision = decisions.every((d) => d === 'ship')\n ? 'ship'\n : decisions.includes('arch_ceiling')\n ? 'arch_ceiling'\n : decisions.includes('model_ceiling')\n ? 'model_ceiling'\n : decisions.includes('hold')\n ? 'hold'\n : 'need_more_work'\n\n const contributing = results.flatMap((r) =>\n r.res.contributingGates.length > 0\n ? r.res.contributingGates\n : [{ name: r.gate.name, passed: r.res.decision === 'ship', detail: r.res }],\n )\n\n const reasons = results.flatMap((r) =>\n r.res.reasons.map((reason) => `[${r.gate.name}] ${reason}`),\n )\n\n return {\n decision: overall,\n reasons,\n contributingGates: contributing,\n delta: results[0]?.res.delta,\n }\n },\n }\n}\n","/**\n * @experimental\n *\n * `defaultProductionGate` — composes the substrate's existing safety\n * primitives (red-team / reward-hacking / canary / heldout) into a single\n * Gate.decide shape. Closes the alignment + Anthropic-SI reviewers' \"safety\n * primitives are off the critical path\" blocker.\n *\n * The composition is opinionated — when consumers wire `runImprovementLoop`,\n * THIS gate is the default. Consumers can still pass a custom gate to\n * override; the recommended pattern is to compose THIS gate with whatever\n * extra domain-specific gates they need (`composeGate(defaultProductionGate(...), customGate)`).\n */\n\nimport type { RunRecord } from '@tangle-network/agent-runtime'\nimport type { CanaryReport } from '../../canary'\nimport { runCanaries } from '../../canary'\nimport type { RedTeamCase } from '../../red-team'\nimport { scoreRedTeamOutput } from '../../red-team'\nimport type { RewardHackingReport } from '../../rl/reward-hacking'\nimport { detectRewardHacking } from '../../rl/reward-hacking'\nimport type { Gate, GateContext, GateResult, Scenario } from '../types'\n\nexport interface DefaultProductionGateOptions {\n /** Required: scenarios held out from training; substrate compares\n * candidate-on-holdout vs baseline-on-holdout. */\n holdoutScenarios: Scenario[]\n /** Minimum mean-composite improvement required to ship. Default 0.5. */\n deltaThreshold?: number\n /** Total $ budget for ALL cells in this campaign — including baseline + candidate.\n * Composite verdict refuses to ship when spend exceeded budget. */\n budgetUsd?: number\n /** Red-team cases to probe candidate outputs against. When omitted the\n * substrate uses `DEFAULT_RED_TEAM_CORPUS`. Provide a domain-specific\n * battery for tighter coverage. */\n redTeamBattery?: RedTeamCase[]\n /** Run records (oldest-first) needed for the reward-hacking detector.\n * Substrate populates from prior production-loop generations. */\n recentRuns?: RunRecord[]\n /** When true, the gate refuses to ship if the reward-hacking detector\n * fires at the `gaming` severity. Default true. */\n blockOnRewardHackingGaming?: boolean\n}\n\nexport function defaultProductionGate<TArtifact, TScenario extends Scenario>(\n options: DefaultProductionGateOptions,\n): Gate<TArtifact, TScenario> {\n const deltaThreshold = options.deltaThreshold ?? 0.5\n const blockOnGaming = options.blockOnRewardHackingGaming ?? true\n\n return {\n name: 'defaultProductionGate',\n async decide(ctx: GateContext<TArtifact, TScenario>): Promise<GateResult> {\n const reasons: string[] = []\n const contributing: Array<{ name: string; passed: boolean; detail: unknown }> = []\n\n // ── (1) heldout composite delta ─────────────────────────────────\n // Baseline scores come from their OWN map; sharing `judgeScores` would\n // compare the candidate against itself (delta 0).\n const baselineComposite = meanComposite(\n ctx.baselineArtifacts,\n ctx.baselineJudgeScores ?? ctx.judgeScores,\n options.holdoutScenarios,\n )\n const candidateComposite = meanComposite(\n ctx.candidateArtifacts,\n ctx.judgeScores,\n options.holdoutScenarios,\n )\n const delta = candidateComposite - baselineComposite\n const heldoutPass = delta >= deltaThreshold\n contributing.push({\n name: 'heldout-delta',\n passed: heldoutPass,\n detail: { baselineComposite, candidateComposite, delta, deltaThreshold },\n })\n if (!heldoutPass) {\n reasons.push(`heldout delta ${delta.toFixed(3)} < threshold ${deltaThreshold}`)\n }\n\n // ── (2) budget gate ─────────────────────────────────────────────\n const budgetPass =\n options.budgetUsd === undefined ||\n ctx.cost.candidate + ctx.cost.baseline <= options.budgetUsd\n contributing.push({\n name: 'budget',\n passed: budgetPass,\n detail: {\n candidateUsd: ctx.cost.candidate,\n baselineUsd: ctx.cost.baseline,\n budgetUsd: options.budgetUsd,\n },\n })\n if (!budgetPass) {\n reasons.push(\n `spend ${(ctx.cost.candidate + ctx.cost.baseline).toFixed(2)} > budget ${options.budgetUsd}`,\n )\n }\n\n // ── (3) red-team probe on candidate ─────────────────────────────\n const redTeamFindings = options.redTeamBattery\n ? probeRedTeam(ctx.candidateArtifacts, options.redTeamBattery)\n : { passed: true, findings: [] }\n contributing.push({\n name: 'red-team',\n passed: redTeamFindings.passed,\n detail: {\n failures: redTeamFindings.findings.length,\n sample: redTeamFindings.findings.slice(0, 3),\n },\n })\n if (!redTeamFindings.passed) {\n reasons.push(`red-team probe failed (${redTeamFindings.findings.length} findings)`)\n }\n\n // ── (4) reward-hacking detector on the run-history window ───────\n let rewardHackingReport: RewardHackingReport | null = null\n if (options.recentRuns && options.recentRuns.length >= 10) {\n rewardHackingReport = detectRewardHacking({ runs: options.recentRuns })\n }\n // reward-hacking severity is numeric (0..1). \"gaming\" threshold per\n // detectRewardHacking defaults = 0.6. Block when ANY finding is at\n // gaming threshold OR the report verdict is 'gaming'.\n const gamingThreshold = 0.6\n const gamingFindings = (rewardHackingReport?.findings ?? []).filter(\n (f) => f.severity >= gamingThreshold,\n )\n const rewardHackingPass =\n !rewardHackingReport ||\n !blockOnGaming ||\n (gamingFindings.length === 0 && rewardHackingReport.verdict !== 'gaming')\n contributing.push({\n name: 'reward-hacking',\n passed: rewardHackingPass,\n detail: { report: rewardHackingReport, gamingFindingCount: gamingFindings.length },\n })\n if (!rewardHackingPass) {\n reasons.push(\n `reward-hacking detector flagged ${gamingFindings.length} gaming-severity findings (verdict=${rewardHackingReport!.verdict})`,\n )\n }\n\n // ── (5) canary check on runs ────────────────────────────────────\n let canaryReport: CanaryReport | null = null\n if (options.recentRuns && options.recentRuns.length >= 10) {\n canaryReport = runCanaries(options.recentRuns, {})\n }\n // CanarySeverity is 'info' | 'warn' | 'error' — block on 'error'.\n const errorAlerts = (canaryReport?.alerts ?? []).filter((a) => a.severity === 'error')\n const canaryPass = errorAlerts.length === 0\n contributing.push({\n name: 'canary',\n passed: canaryPass,\n detail: { totalAlerts: canaryReport?.alerts.length ?? 0, errorAlerts: errorAlerts.length },\n })\n if (!canaryPass) {\n reasons.push(`canary error alerts: ${errorAlerts.length}`)\n }\n\n // ── Verdict ─────────────────────────────────────────────────────\n const allPassed = contributing.every((c) => c.passed)\n const decision = allPassed ? 'ship' : 'hold'\n\n return {\n decision,\n reasons: reasons.length > 0 ? reasons : ['all gates passed'],\n contributingGates: contributing,\n delta,\n }\n },\n }\n}\n\nfunction meanComposite<TArtifact, TScenario extends Scenario>(\n artifacts: Map<string, TArtifact> | undefined,\n judgeScoresByCell: Map<string, Record<string, { composite: number }>>,\n scenarios: TScenario[],\n): number {\n if (!artifacts || artifacts.size === 0) return 0\n const scenarioIds = new Set(scenarios.map((s) => s.id))\n const composites: number[] = []\n for (const [cellId, scores] of judgeScoresByCell) {\n const scenarioId = cellId.split(':')[0] ?? ''\n if (!scenarioIds.has(scenarioId)) continue\n const cellComposites = Object.values(scores).map((s) => s.composite)\n if (cellComposites.length === 0) continue\n composites.push(cellComposites.reduce((a, b) => a + b, 0) / cellComposites.length)\n }\n if (composites.length === 0) return 0\n return composites.reduce((a, b) => a + b, 0) / composites.length\n}\n\nfunction probeRedTeam<TArtifact>(\n artifacts: Map<string, TArtifact>,\n battery: RedTeamCase[],\n): { passed: boolean; findings: Array<{ scenarioId: string; reason: string }> } {\n const findings: Array<{ scenarioId: string; reason: string }> = []\n for (const [_cellId, artifact] of artifacts) {\n const text = extractText(artifact)\n if (text === undefined) continue\n for (const rtCase of battery) {\n const finding = scoreRedTeamOutput(text, [], rtCase)\n if (!finding.passed) {\n findings.push({ scenarioId: rtCase.id, reason: finding.reason ?? 'red-team probe failed' })\n }\n }\n }\n return { passed: findings.length === 0, findings }\n}\n\nfunction extractText(artifact: unknown): string | undefined {\n if (typeof artifact === 'string') return artifact\n if (artifact && typeof artifact === 'object') {\n const rec = artifact as Record<string, unknown>\n if (typeof rec.text === 'string') return rec.text\n if (typeof rec.output === 'string') return rec.output\n if (typeof rec.content === 'string') return rec.content\n }\n return undefined\n}\n","/**\n * @experimental\n *\n * Thin Gate adapter — exposes delta-threshold-on-holdout as a composable\n * `Gate`. Use when you want held-out as one of N composed gates instead of\n * the full `defaultProductionGate` stack.\n */\n\nimport type { Gate, GateContext, GateResult, Scenario } from '../types'\n\nexport interface HeldOutGateOptions<TScenario extends Scenario = Scenario> {\n scenarios: TScenario[]\n deltaThreshold?: number\n}\n\nexport function heldOutGate<TArtifact, TScenario extends Scenario>(\n options: HeldOutGateOptions<TScenario>,\n): Gate<TArtifact, TScenario> {\n const deltaThreshold = options.deltaThreshold ?? 0.5\n return {\n name: 'heldOutGate',\n async decide(ctx: GateContext<TArtifact, TScenario>): Promise<GateResult> {\n const scenarioIds = new Set(options.scenarios.map((s) => s.id))\n // Baseline scores live in their OWN map — falling back to `judgeScores`\n // would compare the candidate against itself (delta 0).\n const baseline = meanForScenarios(ctx.baselineJudgeScores ?? ctx.judgeScores, scenarioIds)\n const candidate = meanForScenarios(ctx.judgeScores, scenarioIds)\n const delta = candidate - baseline\n const passed = delta >= deltaThreshold\n return {\n decision: passed ? 'ship' : 'hold',\n reasons: passed\n ? [`held-out delta ${delta.toFixed(3)} ≥ ${deltaThreshold}`]\n : [`held-out delta ${delta.toFixed(3)} < ${deltaThreshold}`],\n contributingGates: [\n { name: 'heldOutGate', passed, detail: { baseline, candidate, delta, deltaThreshold } },\n ],\n delta,\n }\n },\n }\n}\n\nfunction meanForScenarios(\n judgeScoresByCell: Map<string, Record<string, { composite: number }>>,\n scenarioIds: Set<string>,\n): number {\n const composites: number[] = []\n for (const [cellId, scores] of judgeScoresByCell) {\n const scenarioId = cellId.split(':')[0] ?? ''\n if (!scenarioIds.has(scenarioId)) continue\n const vals = Object.values(scores).map((s) => s.composite)\n if (vals.length > 0) composites.push(vals.reduce((a, b) => a + b, 0) / vals.length)\n }\n return composites.length === 0 ? 0 : composites.reduce((a, b) => a + b, 0) / composites.length\n}\n","/**\n * @experimental\n *\n * Filesystem `LabeledScenarioStore` adapter. The default capture sink for\n * traces + eval artifacts. Production deployments typically swap for a\n * Turso/SQLite adapter (same interface).\n *\n * Records land as one JSONL file per source under `<root>/<source>.jsonl`.\n * Each line is a `LabeledScenarioRecord`. Append-only — no in-place edits.\n *\n * Safety properties enforced at write-time:\n *\n * - **Provenance required**: writes without `source`, `sourceVersionHash`,\n * `capturedAt`, `redactionStatus` are rejected. Closes the alignment\n * reviewer's data-poisoning gap.\n * - **Per-source rate limits**: optional `rateLimitBucket` + `maxWritesPerMinute`\n * stops a single tenant/source from flooding the store.\n *\n * Safety properties enforced at sample-time:\n *\n * - **Required split + capturedBefore**: substrate refuses to sample without\n * an explicit `split` ('train' | 'test') AND a temporal cutoff. Eliminates\n * accidental train/test contamination.\n * - **Default training-source filter**: when the store is sampled with\n * `split: 'train'`, production-trace records are EXCLUDED unless the\n * caller passes `filter.source: 'production-trace'` explicitly. Closes\n * the contamination-by-default gap flagged by the senior eval engineer.\n */\n\nimport { createHash } from 'node:crypto'\nimport { existsSync, mkdirSync, readFileSync, writeFileSync } from 'node:fs'\nimport { join } from 'node:path'\nimport type {\n LabeledScenarioRecord,\n LabeledScenarioSampleArgs,\n LabeledScenarioSource,\n LabeledScenarioStore,\n LabeledScenarioWrite,\n} from '../types'\n\nexport interface FsLabeledScenarioStoreOptions {\n /** Root directory for JSONL files. Created if missing. */\n root: string\n /** Per-source rate limit. When set, writes exceeding the cap are rejected\n * with a typed error. Default: no limit. */\n maxWritesPerMinutePerBucket?: number\n /** Test seam — override `Date.now()` for deterministic tests. */\n now?: () => number\n}\n\nexport class LabeledScenarioStoreError extends Error {\n constructor(\n public readonly code: string,\n message: string,\n ) {\n super(message)\n this.name = 'LabeledScenarioStoreError'\n }\n}\n\ninterface RateLimitState {\n bucket: string\n windowStartMs: number\n count: number\n}\n\nexport class FsLabeledScenarioStore implements LabeledScenarioStore {\n private readonly now: () => number\n private readonly rateLimits = new Map<string, RateLimitState>()\n\n constructor(private readonly options: FsLabeledScenarioStoreOptions) {\n if (!existsSync(options.root)) mkdirSync(options.root, { recursive: true })\n this.now = options.now ?? Date.now\n }\n\n async observe(write: LabeledScenarioWrite): Promise<void> {\n this.assertProvenance(write)\n this.assertRateLimit(write)\n const record = this.toRecord(write)\n const path = this.pathForSource(write.source)\n const line = `${JSON.stringify(record)}\\n`\n // Append atomically. For high-throughput a writev-friendly buffered\n // implementation lands in the Turso adapter; FS adapter is for tests +\n // local dev + small workloads.\n appendLine(path, line)\n }\n\n async sample(args: LabeledScenarioSampleArgs): Promise<LabeledScenarioRecord[]> {\n if (!args.split) {\n throw new LabeledScenarioStoreError(\n 'split_required',\n 'sample() requires an explicit `split` (train | test) — substrate refuses ambiguous reads',\n )\n }\n if (!args.capturedBefore) {\n throw new LabeledScenarioStoreError(\n 'capturedBefore_required',\n 'sample() requires an explicit `capturedBefore` timestamp for temporal-split discipline',\n )\n }\n\n const all: LabeledScenarioRecord[] = []\n for (const source of ALL_SOURCES) {\n // Default training-source filter: when sampling train, EXCLUDE\n // production-trace records unless the caller asks for them.\n if (args.split === 'train' && source === 'production-trace') {\n const explicit = sourceFilterContains(args.filter?.source, 'production-trace')\n if (!explicit) continue\n }\n const path = this.pathForSource(source)\n if (!existsSync(path)) continue\n const lines = readFileSync(path, 'utf8').split('\\n').filter(Boolean)\n for (const line of lines) {\n let record: LabeledScenarioRecord\n try {\n record = JSON.parse(line) as LabeledScenarioRecord\n } catch {\n continue\n }\n if (!matchesFilter(record, args, source)) continue\n all.push(record)\n }\n }\n\n // Deterministic order: by capturedAt ascending, then recordHash.\n all.sort((a, b) => {\n if (a.capturedAt !== b.capturedAt) return a.capturedAt.localeCompare(b.capturedAt)\n return a.recordHash.localeCompare(b.recordHash)\n })\n\n return all.slice(0, args.count)\n }\n\n async size(): Promise<{ train: number; test: number; bySource: Record<string, number> }> {\n const bySource: Record<string, number> = {}\n let total = 0\n for (const source of ALL_SOURCES) {\n const path = this.pathForSource(source)\n if (!existsSync(path)) {\n bySource[source] = 0\n continue\n }\n const count = readFileSync(path, 'utf8').split('\\n').filter(Boolean).length\n bySource[source] = count\n total += count\n }\n // FS adapter doesn't track split assignments per-record (split is\n // computed at sample-time based on `capturedBefore`). For size(), we\n // report `train`+`test` as the same total — split is a sampling concept.\n return { train: total, test: total, bySource }\n }\n\n private assertProvenance(write: LabeledScenarioWrite): void {\n if (!write.source) {\n throw new LabeledScenarioStoreError(\n 'missing_source',\n 'LabeledScenarioWrite requires `source`',\n )\n }\n if (!write.sourceVersionHash || write.sourceVersionHash.length === 0) {\n throw new LabeledScenarioStoreError(\n 'missing_source_version',\n 'LabeledScenarioWrite requires `sourceVersionHash` (git sha or substrate version)',\n )\n }\n if (!write.capturedAt) {\n throw new LabeledScenarioStoreError(\n 'missing_captured_at',\n 'LabeledScenarioWrite requires `capturedAt` ISO timestamp',\n )\n }\n if (!write.redactionStatus) {\n throw new LabeledScenarioStoreError(\n 'missing_redaction_status',\n 'LabeledScenarioWrite requires explicit `redactionStatus` — raw / redacted-pii / redacted-secrets / fully-redacted',\n )\n }\n if (!ALL_SOURCES.includes(write.source)) {\n throw new LabeledScenarioStoreError(\n 'unknown_source',\n `LabeledScenarioWrite.source must be one of: ${ALL_SOURCES.join(', ')}`,\n )\n }\n }\n\n private assertRateLimit(write: LabeledScenarioWrite): void {\n const cap = this.options.maxWritesPerMinutePerBucket\n if (!cap || !write.rateLimitBucket) return\n const now = this.now()\n const windowMs = 60_000\n let state = this.rateLimits.get(write.rateLimitBucket)\n if (!state || now - state.windowStartMs >= windowMs) {\n state = { bucket: write.rateLimitBucket, windowStartMs: now, count: 0 }\n this.rateLimits.set(write.rateLimitBucket, state)\n }\n if (state.count >= cap) {\n throw new LabeledScenarioStoreError(\n 'rate_limit_exceeded',\n `LabeledScenarioStore: bucket ${write.rateLimitBucket} exceeded ${cap} writes/min`,\n )\n }\n state.count += 1\n }\n\n private toRecord(write: LabeledScenarioWrite): LabeledScenarioRecord {\n const recordHash = sha256(\n JSON.stringify({\n id: write.scenario.id,\n src: write.source,\n at: write.capturedAt,\n ver: write.sourceVersionHash,\n }),\n )\n // FS adapter assigns split at sample-time, but we cache a hint here\n // based on capturedAt vs the world's \"now\" — sampler overrides this.\n return {\n ...write,\n recordHash,\n split: 'train',\n }\n }\n\n private pathForSource(source: string): string {\n return join(this.options.root, `${source}.jsonl`)\n }\n}\n\nconst ALL_SOURCES: LabeledScenarioWrite['source'][] = [\n 'production-trace',\n 'eval-run',\n 'manual',\n 'red-team',\n 'synthetic',\n]\n\nfunction sourceFilterContains(\n filter: LabeledScenarioSource | LabeledScenarioSource[] | undefined,\n needle: LabeledScenarioSource,\n): boolean {\n if (!filter) return false\n if (Array.isArray(filter)) return filter.includes(needle)\n return filter === needle\n}\n\nfunction matchesFilter(\n record: LabeledScenarioRecord,\n args: LabeledScenarioSampleArgs,\n source: string,\n): boolean {\n // Temporal cutoff — train must be capturedAt < capturedBefore.\n if (args.split === 'train' && record.capturedAt >= args.capturedBefore) return false\n if (args.split === 'test' && record.capturedAt < args.capturedBefore) return false\n\n const f = args.filter\n if (!f) return true\n if (f.kind && record.scenario.kind !== f.kind) return false\n if (f.source) {\n const sources = Array.isArray(f.source) ? f.source : [f.source]\n if (!sources.includes(source as never)) return false\n }\n if (f.minComposite !== undefined || f.maxComposite !== undefined) {\n const composites = Object.values(record.judgeScores).map((s) => s.composite)\n const max = composites.length === 0 ? 0 : Math.max(...composites)\n if (f.minComposite !== undefined && max < f.minComposite) return false\n if (f.maxComposite !== undefined && max > f.maxComposite) return false\n }\n return true\n}\n\nfunction sha256(input: string): string {\n return createHash('sha256').update(input).digest('hex').slice(0, 16)\n}\n\nfunction appendLine(path: string, line: string): void {\n if (existsSync(path)) {\n const existing = readFileSync(path, 'utf8')\n writeFileSync(path, existing + line)\n } else {\n writeFileSync(path, line)\n }\n}\n","/**\n * @experimental\n *\n * `runEval` — the simplest preset over `runCampaign`. No optimizer, no\n * gate, no auto-PR. Just: run scenarios through dispatch, score with\n * judges, return CampaignResult.\n *\n * The 80% case for consumers who want a scorecard, not an improvement loop.\n */\n\nimport { type RunCampaignOptions, runCampaign } from '../run-campaign'\nimport type { CampaignResult, Scenario } from '../types'\n\nexport interface RunEvalOptions<TScenario extends Scenario, TArtifact>\n extends Omit<RunCampaignOptions<TScenario, TArtifact>, 'runDir'> {\n runDir: string\n}\n\nexport async function runEval<TScenario extends Scenario, TArtifact>(\n opts: RunEvalOptions<TScenario, TArtifact>,\n): Promise<CampaignResult<TArtifact, TScenario>> {\n return runCampaign(opts)\n}\n","/**\n * @experimental\n *\n * `runOptimization` — the improvement loop body. Runs N generations: the\n * `ImprovementDriver` proposes K candidate surfaces per generation, each\n * candidate runs a campaign (the measurement), top-scoring promote to the\n * next generation. Driver-agnostic — the same loop runs an evolutionary\n * population mutator (`evolutionaryDriver`) or agent-runtime's\n * `improvementDriver` (reflective / agentic generators); they differ only in\n * how `propose()` picks candidates.\n *\n * This is `runLoop`'s shape (plan → measure → decide) specialized to surface\n * improvement: `driver.propose` = plan, `runCampaign` = the measurement (which\n * runs the worker behind `dispatch`), the mean-composite ranking = the\n * validator, `driver.decide` = the stop check.\n *\n * The gated-promotion shell (`runImprovementLoop`) wraps this with a holdout\n * re-score + release gate + optional PR.\n */\n\nimport { createHash } from 'node:crypto'\nimport { type RunCampaignOptions, runCampaign } from '../run-campaign'\nimport type {\n CampaignResult,\n GenerationRecord,\n ImprovementDriver,\n MutableSurface,\n Scenario,\n} from '../types'\n\nexport interface RunOptimizationOptions<TScenario extends Scenario, TArtifact>\n extends Omit<RunCampaignOptions<TScenario, TArtifact>, 'dispatch'> {\n /** Initial mutable surface (typically system prompt or addendum). */\n baselineSurface: MutableSurface\n /** Dispatcher that takes the CURRENT surface + scenario → artifact. */\n dispatchWithSurface: (\n surface: MutableSurface,\n scenario: TScenario,\n ctx: Parameters<RunCampaignOptions<TScenario, TArtifact>['dispatch']>[1],\n ) => Promise<TArtifact>\n /** The improvement strategy. Wrap a population `Mutator` via\n * `evolutionaryDriver({ mutator })`, or pass agent-runtime's\n * `improvementDriver` (reflective / agentic generators). */\n driver: ImprovementDriver\n populationSize: number\n maxGenerations: number\n /** How many top-scoring candidates carry to the next generation. Default 2. */\n promoteTopK?: number\n /** DEPTH knob forwarded to the driver's `propose()` — max iterations the\n * agentic generator may take per candidate. */\n maxImprovementShots?: number\n /** Phase-2 research report forwarded to `propose()` (analyst findings +\n * diff). Opaque here; the driver types it. */\n report?: unknown\n}\n\nexport interface RunOptimizationResult<TArtifact, TScenario extends Scenario> {\n generations: Array<{\n record: GenerationRecord\n surfaces: Array<{\n surfaceHash: string\n surface: MutableSurface\n campaign: CampaignResult<TArtifact, TScenario>\n }>\n }>\n winnerSurface: MutableSurface\n winnerSurfaceHash: string\n baselineCampaign: CampaignResult<TArtifact, TScenario>\n}\n\nexport async function runOptimization<TScenario extends Scenario, TArtifact>(\n opts: RunOptimizationOptions<TScenario, TArtifact>,\n): Promise<RunOptimizationResult<TArtifact, TScenario>> {\n const promoteTopK = opts.promoteTopK ?? 2\n\n // Baseline run\n const baselineCampaign = await runCampaign<TScenario, TArtifact>({\n ...opts,\n dispatch: (scenario, ctx) => opts.dispatchWithSurface(opts.baselineSurface, scenario, ctx),\n runDir: `${opts.runDir}/baseline`,\n })\n\n const generations: RunOptimizationResult<TArtifact, TScenario>['generations'] = []\n const history: GenerationRecord[] = []\n let currentSurfaces: MutableSurface[] = [opts.baselineSurface]\n let winnerSurface = opts.baselineSurface\n let winnerSurfaceHash = surfaceHash(opts.baselineSurface)\n let winnerComposite = meanComposite(baselineCampaign)\n\n for (let gen = 0; gen < opts.maxGenerations; gen++) {\n // Decide: the driver may stop early based on accumulated history.\n if (opts.driver.decide?.({ history }).stop) break\n\n // Plan: the driver proposes N candidates from the current best surface,\n // the accumulated generation history, and any external findings.\n const candidates = await opts.driver.propose({\n currentSurface: currentSurfaces[0] ?? opts.baselineSurface,\n history,\n findings: [],\n populationSize: opts.populationSize,\n generation: gen,\n signal: new AbortController().signal,\n report: opts.report,\n dataset: opts.labeledStore && opts.labeledStore !== 'off' ? opts.labeledStore : undefined,\n maxImprovementShots: opts.maxImprovementShots,\n })\n\n // Run each candidate as its own campaign.\n const surfaceResults: Array<{\n surfaceHash: string\n surface: MutableSurface\n campaign: CampaignResult<TArtifact, TScenario>\n composite: number\n }> = []\n for (let i = 0; i < candidates.length; i++) {\n const surface = candidates[i] as MutableSurface\n const hash = surfaceHash(surface)\n const campaign = await runCampaign<TScenario, TArtifact>({\n ...opts,\n dispatch: (scenario, ctx) => opts.dispatchWithSurface(surface, scenario, ctx),\n runDir: `${opts.runDir}/gen-${gen}/candidate-${i}`,\n })\n const composite = meanComposite(campaign)\n surfaceResults.push({ surfaceHash: hash, surface, campaign, composite })\n }\n\n // Rank, promote top-K.\n surfaceResults.sort((a, b) => b.composite - a.composite)\n const promoted = surfaceResults.slice(0, promoteTopK)\n currentSurfaces = promoted.map((p) => p.surface)\n const top = surfaceResults[0]\n if (top && top.composite > winnerComposite) {\n winnerSurface = top.surface\n winnerSurfaceHash = top.surfaceHash\n winnerComposite = top.composite\n }\n\n const record: GenerationRecord = {\n generationIndex: gen,\n candidates: surfaceResults.map((s) => {\n const breakdown = candidateBreakdown(s.campaign)\n return {\n surfaceHash: s.surfaceHash,\n composite: s.composite,\n ci95: [s.composite, s.composite] as [number, number],\n dimensions: breakdown.dimensions,\n scenarios: breakdown.scenarios,\n }\n }),\n promoted: promoted.map((p) => p.surfaceHash),\n }\n history.push(record)\n generations.push({\n record,\n surfaces: surfaceResults.map((s) => ({\n surfaceHash: s.surfaceHash,\n surface: s.surface,\n campaign: s.campaign,\n })),\n })\n }\n\n return {\n generations,\n winnerSurface,\n winnerSurfaceHash,\n baselineCampaign,\n }\n}\n\nexport function surfaceHash(surface: MutableSurface): string {\n // Prompt/tool surfaces (string) hash by content; code surfaces hash by the\n // worktree + base ref pair (the content lives in git, not in the string).\n const material =\n typeof surface === 'string'\n ? surface\n : JSON.stringify({\n kind: surface.kind,\n worktreeRef: surface.worktreeRef,\n baseRef: surface.baseRef ?? null,\n })\n return createHash('sha256').update(material).digest('hex').slice(0, 16)\n}\n\nfunction meanComposite<TArtifact, TScenario extends Scenario>(\n campaign: CampaignResult<TArtifact, TScenario>,\n): number {\n const composites: number[] = []\n for (const cell of campaign.cells) {\n const cellComposites = Object.values(cell.judgeScores).map((s) => s.composite)\n if (cellComposites.length > 0) {\n composites.push(cellComposites.reduce((a, b) => a + b, 0) / cellComposites.length)\n }\n }\n return composites.length === 0 ? 0 : composites.reduce((a, b) => a + b, 0) / composites.length\n}\n\n/** Per-candidate evidence a reflective driver grounds its next proposal on:\n * mean score per judge dimension + per-scenario composite. */\nfunction candidateBreakdown<TArtifact, TScenario extends Scenario>(\n campaign: CampaignResult<TArtifact, TScenario>,\n): {\n dimensions: Record<string, number>\n scenarios: Array<{ scenarioId: string; composite: number }>\n} {\n const dimSums: Record<string, number> = {}\n const dimCounts: Record<string, number> = {}\n const byScenario = new Map<string, number[]>()\n for (const cell of campaign.cells) {\n const judgeScores = Object.values(cell.judgeScores)\n if (judgeScores.length === 0) continue\n const cellComposite = judgeScores.reduce((a, s) => a + s.composite, 0) / judgeScores.length\n const arr = byScenario.get(cell.scenarioId) ?? []\n arr.push(cellComposite)\n byScenario.set(cell.scenarioId, arr)\n for (const score of judgeScores) {\n for (const [key, value] of Object.entries(score.dimensions)) {\n dimSums[key] = (dimSums[key] ?? 0) + value\n dimCounts[key] = (dimCounts[key] ?? 0) + 1\n }\n }\n }\n const dimensions: Record<string, number> = {}\n for (const key of Object.keys(dimSums)) {\n const count = dimCounts[key] ?? 0\n dimensions[key] = count > 0 ? (dimSums[key] ?? 0) / count : 0\n }\n const scenarios = [...byScenario.entries()].map(([scenarioId, comps]) => ({\n scenarioId,\n composite: comps.reduce((a, b) => a + b, 0) / comps.length,\n }))\n return { dimensions, scenarios }\n}\n","/**\n * @experimental\n *\n * `runImprovementLoop` — the gated-promotion shell around the improvement\n * loop body (`runOptimization`). Drives candidate surfaces via the\n * `ImprovementDriver`, re-scores the winner against the baseline on a\n * holdout set, runs the release gate, and optionally opens a PR.\n *\n * Role vocabulary (see docs/design/loop-taxonomy.md):\n * - DRIVER = the `ImprovementDriver` (evolutionary GEPA mutator OR\n * reflective analyst). Proposes candidate SURFACES — the\n * worker's system prompt / tool config — NOT conversation\n * turns.\n * - MEASUREMENT= `runCampaign`. Scores one surface by running the worker\n * (via `dispatch`) over scenarios and judging the output.\n * - WORKER = the agent harness in the sandbox, invoked behind the\n * topology-opaque `dispatch` seam — never referenced here.\n *\n * Distinct from `runLoop` in `@tangle-network/agent-runtime`, which is the\n * INNER conversation loop (driver↔workers in a sandbox). `runImprovementLoop`\n * is the OUTER loop: it improves the surface that those workers run.\n *\n * Hard-refuses unsafe configurations:\n * - `tracing: 'off'` when a driver is wired (improvement is unattributable)\n * - `autoOnPromote: 'config'` — DEFERRED to Pass B; v0.40 only ships\n * `'pr'` and `'none'`.\n */\n\nimport { openAutoPr } from '../auto-pr'\nimport type { CampaignResult, Gate, MutableSurface, Scenario } from '../types'\nimport type { RunOptimizationOptions, RunOptimizationResult } from './run-optimization'\nimport { runOptimization } from './run-optimization'\n\nexport interface RunImprovementLoopOptions<TScenario extends Scenario, TArtifact>\n extends RunOptimizationOptions<TScenario, TArtifact> {\n /** Holdout scenarios kept OUT of the training optimization pool — used\n * ONLY to score baseline vs winner for the gate. */\n holdoutScenarios: TScenario[]\n /** Promotion gate. Substrate strongly recommends `defaultProductionGate`\n * for production wiring (composes red-team / reward-hacking / canary /\n * heldout). */\n gate: Gate<TArtifact, TScenario>\n /** What to do when the gate ships:\n * - `'pr'`: open a PR via `openAutoPr`\n * - `'none'`: just report — caller decides what to do with the winner\n * v0.40 does NOT support `'config'` (live-runtime self-mutation) —\n * deferred to Pass B behind safety stack. */\n autoOnPromote: 'pr' | 'none'\n /** GH owner / repo for the auto-PR. Required when autoOnPromote === 'pr'. */\n ghOwner?: string\n ghRepo?: string\n /** Optional render override — substrate writes a diff-shaped surface; pass\n * a function to format the promoted surface differently. */\n renderPromotedDiff?: (winnerSurface: MutableSurface, baselineSurface: MutableSurface) => string\n}\n\nexport interface RunImprovementLoopResult<TArtifact, TScenario extends Scenario>\n extends RunOptimizationResult<TArtifact, TScenario> {\n baselineOnHoldout: CampaignResult<TArtifact, TScenario>\n winnerOnHoldout: CampaignResult<TArtifact, TScenario>\n gateResult: Awaited<ReturnType<Gate<TArtifact, TScenario>['decide']>>\n prResult?: ReturnType<typeof openAutoPr>\n}\n\nexport async function runImprovementLoop<TScenario extends Scenario, TArtifact>(\n opts: RunImprovementLoopOptions<TScenario, TArtifact>,\n): Promise<RunImprovementLoopResult<TArtifact, TScenario>> {\n // ── Safety pre-flight ─────────────────────────────────────────────\n // biome-ignore lint/suspicious/noExplicitAny: Pass A reserved field for Pass B Shape B\n if ((opts as any).autoOnPromote === 'config') {\n throw new Error(\n \"runImprovementLoop: autoOnPromote='config' is deferred to Pass B (requires shadow deploy + rollback + ensemble judges). Use 'pr' or 'none' in v0.40.\",\n )\n }\n // Refuse tracing=off whenever a driver is wired. An improvement loop\n // without traces is unattributable — its candidate surfaces cannot be\n // cited back to the spans that motivated them, and the dataset flywheel\n // (LabeledScenarioStore) that GEPA optimizes against goes unfed.\n if (opts.tracing === 'off' && opts.driver) {\n throw new Error(\n \"runImprovementLoop: tracing='off' is forbidden when a driver is wired. The improvement loop without traces is unattributable; candidate surfaces cannot be cited back to spans and the optimization dataset goes unfed.\",\n )\n }\n if (opts.autoOnPromote === 'pr' && (!opts.ghOwner || !opts.ghRepo)) {\n throw new Error(\"runImprovementLoop: autoOnPromote='pr' requires ghOwner + ghRepo.\")\n }\n\n // ── (1) optimization loop produces a winner ────────────────────────\n const optimization = await runOptimization(opts)\n\n // ── (2) baseline + winner re-scored on the holdout set ─────────────\n const { runCampaign } = await import('../run-campaign')\n\n const baselineOnHoldout = await runCampaign<TScenario, TArtifact>({\n ...opts,\n scenarios: opts.holdoutScenarios,\n dispatch: (scenario, ctx) => opts.dispatchWithSurface(opts.baselineSurface, scenario, ctx),\n runDir: `${opts.runDir}/holdout-baseline`,\n })\n\n const winnerOnHoldout = await runCampaign<TScenario, TArtifact>({\n ...opts,\n scenarios: opts.holdoutScenarios,\n dispatch: (scenario, ctx) =>\n opts.dispatchWithSurface(optimization.winnerSurface, scenario, ctx),\n runDir: `${opts.runDir}/holdout-winner`,\n })\n\n // ── (3) gate verdict ───────────────────────────────────────────────\n // Candidate + baseline share cellIds (same holdout scenarios), so their\n // judge scores MUST stay in separate maps — merging them collapses the\n // holdout delta to zero and the gate can never ship a real improvement.\n type ScoreMap = Map<\n string,\n Record<string, { composite: number; dimensions: Record<string, number>; notes: string }>\n >\n const candidateArtifacts = new Map<string, TArtifact>()\n const baselineArtifacts = new Map<string, TArtifact>()\n const judgeScores: ScoreMap = new Map()\n const baselineJudgeScores: ScoreMap = new Map()\n for (const cell of winnerOnHoldout.cells) {\n candidateArtifacts.set(cell.cellId, cell.artifact)\n judgeScores.set(cell.cellId, cell.judgeScores)\n }\n for (const cell of baselineOnHoldout.cells) {\n baselineArtifacts.set(cell.cellId, cell.artifact)\n baselineJudgeScores.set(cell.cellId, cell.judgeScores)\n }\n\n const gateResult = await opts.gate.decide({\n candidateArtifacts,\n baselineArtifacts,\n judgeScores,\n baselineJudgeScores,\n scenarios: opts.holdoutScenarios,\n cost: {\n candidate: winnerOnHoldout.aggregates.totalCostUsd,\n baseline: baselineOnHoldout.aggregates.totalCostUsd,\n },\n signal: new AbortController().signal,\n })\n\n // ── (4) auto-PR when gate ships ────────────────────────────────────\n let prResult: ReturnType<typeof openAutoPr> | undefined\n if (opts.autoOnPromote === 'pr' && gateResult.decision === 'ship') {\n const render = opts.renderPromotedDiff ?? defaultRenderDiff\n const promotedDiff = render(optimization.winnerSurface, opts.baselineSurface)\n prResult = openAutoPr({\n result: winnerOnHoldout,\n gate: gateResult,\n promotedDiff,\n ghOwner: opts.ghOwner!,\n ghRepo: opts.ghRepo!,\n })\n }\n\n return {\n ...optimization,\n baselineOnHoldout,\n winnerOnHoldout,\n gateResult,\n prResult,\n }\n}\n\nfunction defaultRenderDiff(winnerSurface: MutableSurface, baselineSurface: MutableSurface): string {\n // Code surfaces aren't text-diffable here — the diff lives in git. Render\n // the worktree/base refs + summary so the PR body points at the change.\n if (typeof winnerSurface !== 'string' || typeof baselineSurface !== 'string') {\n const fmt = (s: MutableSurface): string =>\n typeof s === 'string'\n ? '(prompt surface)'\n : `worktree=${s.worktreeRef}${s.baseRef ? ` base=${s.baseRef}` : ''}${s.summary ? `\\n${s.summary}` : ''}`\n return `--- baseline\\n${fmt(baselineSurface)}\\n+++ winner\\n${fmt(winnerSurface)}`\n }\n const lines: string[] = []\n lines.push('--- baseline')\n lines.push('+++ winner')\n for (const l of baselineSurface.split('\\n')) lines.push(`- ${l}`)\n for (const l of winnerSurface.split('\\n')) lines.push(`+ ${l}`)\n return lines.join('\\n')\n}\n","/**\n * @experimental\n *\n * VCS-pluggable worktree adapter. One improvement = one worktree, PR-like\n * (multiple commits allowed). A code-tier driver's `propose()` creates a\n * worktree, an agent commits the change into it, and `finalize()` returns a\n * `CodeSurface{ worktreeRef }` the measurement checks out to run the worker\n * against the changed code. On promotion the worktree becomes the PR branch.\n *\n * The interface is VCS-agnostic so a future `jj` ([jj-vcs](https://github.com/jj-vcs/jj))\n * adapter can slot in without touching driver code. Only the git adapter\n * ships today. See `docs/design/self-improvement-engine.md`.\n */\n\nimport { execFileSync } from 'node:child_process'\nimport { existsSync } from 'node:fs'\nimport { basename, isAbsolute, join } from 'node:path'\nimport type { CodeSurface } from '../types'\n\nexport interface Worktree {\n /** Absolute path to the checked-out worktree directory. */\n path: string\n /** The branch the worktree is on (becomes the PR branch on promotion). */\n branch: string\n /** The ref the worktree was forked from. */\n baseRef: string\n}\n\nexport interface WorktreeAdapter {\n /** Create an isolated worktree on a fresh branch off `baseRef`. */\n create(opts: { baseRef: string; label: string }): Promise<Worktree>\n /** Commit any pending changes in the worktree, then return a CodeSurface\n * pointing at it. The agent has already written its change into\n * `worktree.path` by the time this is called. */\n finalize(worktree: Worktree, summary: string): Promise<CodeSurface>\n /** Remove the worktree (and its branch) — called for losing candidates. */\n discard(worktree: Worktree): Promise<void>\n}\n\nexport class WorktreeAdapterError extends Error {\n constructor(\n message: string,\n readonly cause?: unknown,\n ) {\n super(message)\n this.name = 'WorktreeAdapterError'\n }\n}\n\nexport interface GitWorktreeAdapterOptions {\n /** Repo root the worktrees fork from. */\n repoRoot: string\n /** Directory worktrees are created under. Default: `<repoRoot>/.worktrees`. */\n worktreeDir?: string\n /** Branch-name prefix. Default: `improve`. */\n branchPrefix?: string\n /** Test seam — defaults to a real `git` runner. */\n git?: (args: string[], cwd: string) => string\n}\n\nfunction defaultGit(args: string[], cwd: string): string {\n try {\n return execFileSync('git', args, { cwd, encoding: 'utf8' }).trim()\n } catch (err) {\n const stderr =\n err && typeof err === 'object' && 'stderr' in err\n ? String((err as { stderr: unknown }).stderr)\n : ''\n throw new WorktreeAdapterError(`git ${args.join(' ')} failed: ${stderr || String(err)}`, err)\n }\n}\n\n/** Slugify a label into a branch-safe segment. */\nfunction slug(label: string): string {\n return (\n label\n .toLowerCase()\n .replace(/[^a-z0-9]+/g, '-')\n .replace(/^-+|-+$/g, '')\n .slice(0, 48) || 'candidate'\n )\n}\n\nexport function gitWorktreeAdapter(opts: GitWorktreeAdapterOptions): WorktreeAdapter {\n const git = opts.git ?? defaultGit\n const worktreeDir = opts.worktreeDir ?? join(opts.repoRoot, '.worktrees')\n const branchPrefix = opts.branchPrefix ?? 'improve'\n\n return {\n async create({ baseRef, label }) {\n const id = `${slug(label)}-${Date.now().toString(36)}-${Math.random().toString(36).slice(2, 6)}`\n const branch = `${branchPrefix}/${id}`\n const path = join(worktreeDir, id)\n git(['worktree', 'add', '-b', branch, path, baseRef], opts.repoRoot)\n return { path, branch, baseRef }\n },\n\n async finalize(worktree, summary) {\n // Stage + commit any pending changes the agent left in the worktree.\n // A no-op commit is refused by git, so only commit when the tree is dirty.\n const status = git(['status', '--porcelain'], worktree.path)\n if (status.length > 0) {\n git(['add', '-A'], worktree.path)\n git(['commit', '-m', summary], worktree.path)\n }\n return {\n kind: 'code',\n worktreeRef: worktree.path,\n baseRef: worktree.baseRef,\n summary,\n }\n },\n\n async discard(worktree) {\n // Remove the worktree, then delete its branch. Force-remove because the\n // worktree may hold uncommitted experiment state we're discarding.\n git(['worktree', 'remove', '--force', worktree.path], opts.repoRoot)\n git(['branch', '-D', worktree.branch], opts.repoRoot)\n },\n }\n}\n\n/** Resolve a `CodeSurface`'s worktreeRef to a directory the measurement can\n * run the worker in. A path ref is returned as-is; anything else is treated\n * as a ref under the adapter's worktree dir. */\nexport function resolveWorktreePath(surface: CodeSurface, worktreeDir?: string): string {\n if (isAbsolute(surface.worktreeRef) && existsSync(surface.worktreeRef)) return surface.worktreeRef\n if (worktreeDir) return join(worktreeDir, basename(surface.worktreeRef))\n return surface.worktreeRef\n}\n"],"mappings":";;;;;;;;;;;;;;;;;;;;;;AAcA,SAAS,gBAAgB;AACzB,SAAS,qBAAqB;AAC9B,SAAS,cAAc;AACvB,SAAS,YAAY;AAiCd,SAAS,WACd,SACkB;AAClB,MAAI,QAAQ,KAAK,aAAa,QAAQ;AACpC,WAAO;AAAA,MACL,QAAQ;AAAA,MACR,QAAQ;AAAA,MACR,QAAQ,qBAAqB,QAAQ,KAAK,QAAQ;AAAA,IACpD;AAAA,EACF;AAEA,QAAM,SAAS,QAAQ,UAAU,CAAC,QAAQ,IAAI;AAC9C,QAAM,SAAS,QAAQ,UAAU,QAAQ,QAAQ,OAAO,aAAa,MAAM,GAAG,EAAE,CAAC;AACjF,QAAM,QACJ,QAAQ,SAAS,kBAAkB,QAAQ,OAAO,aAAa,MAAM,GAAG,CAAC,CAAC;AAE5E,QAAM,OAAO,aAAa,QAAQ,QAAQ,QAAQ,MAAM,QAAQ,YAAY;AAC5E,QAAM,WAAW,KAAK,OAAO,GAAG,gBAAgB,KAAK,IAAI,CAAC,KAAK;AAC/D,gBAAc,UAAU,IAAI;AAE5B,MAAI,QAAQ;AACV,WAAO;AAAA,MACL,QAAQ;AAAA,MACR,QAAQ;AAAA,MACR,QAAQ,0DAA0D,QAAQ,OAAO,IAAI,QAAQ,MAAM,WAAW,MAAM,aAAa,QAAQ;AAAA,IAC3I;AAAA,EACF;AAEA,QAAM,SAAS,QAAQ,UAAU;AACjC,QAAM,SAAS,OAAO;AAAA,IACpB;AAAA,IACA;AAAA,IACA;AAAA,IACA,GAAG,QAAQ,OAAO,IAAI,QAAQ,MAAM;AAAA,IACpC;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,EACF,CAAC;AACD,MAAI,OAAO,WAAW,GAAG;AACvB,WAAO;AAAA,MACL,QAAQ;AAAA,MACR,QAAQ;AAAA,MACR,QAAQ,6BAA6B,OAAO,MAAM,MAAM,OAAO,OAAO,MAAM,GAAG,GAAG,CAAC;AAAA,IACrF;AAAA,EACF;AACA,QAAM,QAAQ,OAAO,OAAO,KAAK;AACjC,SAAO,EAAE,QAAQ,MAAM,OAAO,QAAQ,OAAO,QAAQ,YAAY;AACnE;AAEA,SAAS,aACP,QACA,MACA,MACQ;AACR,QAAM,QAAkB,CAAC;AACzB,QAAM,KAAK,kDAAkD;AAC7D,QAAM,KAAK,EAAE;AACb,QAAM,KAAK,mBAAmB,OAAO,YAAY,IAAI;AACrD,QAAM,KAAK,aAAa,OAAO,IAAI,EAAE;AACrC,QAAM,KAAK,iBAAiB,KAAK,MAAM,OAAO,aAAa,GAAI,CAAC,GAAG;AACnE,QAAM;AAAA,IACJ,uBAAuB,OAAO,WAAW,aAAa,YAAY,OAAO,WAAW,WAAW,aAAa,OAAO,WAAW,YAAY,YAAY,OAAO,WAAW,WAAW;AAAA,EACrL;AACA,QAAM,KAAK,qBAAqB,OAAO,WAAW,aAAa,QAAQ,CAAC,CAAC,EAAE;AAC3E,QAAM,KAAK,EAAE;AACb,QAAM,KAAK,uBAAuB,KAAK,QAAQ,IAAI;AACnD,QAAM,KAAK,EAAE;AACb,aAAW,UAAU,KAAK,QAAS,OAAM,KAAK,KAAK,MAAM,EAAE;AAC3D,MAAI,KAAK,UAAU,OAAW,OAAM,KAAK,YAAY,KAAK,MAAM,QAAQ,CAAC,CAAC,EAAE;AAC5E,QAAM,KAAK,EAAE;AACb,QAAM,KAAK,wBAAwB;AACnC,QAAM,KAAK,EAAE;AACb,QAAM,KAAK,4BAA4B;AACvC,QAAM,KAAK,eAAe;AAC1B,aAAW,KAAK,KAAK,mBAAmB;AACtC,UAAM,SACJ,OAAO,EAAE,WAAW,WAChB,KAAK,UAAU,EAAE,MAAM,EAAE,MAAM,GAAG,EAAE,IACpC,OAAO,EAAE,MAAM,EAAE,MAAM,GAAG,EAAE;AAClC,UAAM,KAAK,KAAK,EAAE,IAAI,MAAM,EAAE,SAAS,WAAM,QAAG,MAAM,MAAM,IAAI;AAAA,EAClE;AACA,QAAM,KAAK,EAAE;AACb,QAAM,KAAK,sBAAsB;AACjC,QAAM,KAAK,EAAE;AACb,QAAM,KAAK,SAAS;AACpB,QAAM,KAAK,KAAK,MAAM,GAAG,GAAI,CAAC;AAC9B,QAAM,KAAK,KAAK;AAChB,QAAM,KAAK,EAAE;AACb,QAAM,KAAK,yBAAyB;AACpC,QAAM,KAAK,EAAE;AACb,QAAM,KAAK,6BAA6B;AACxC,QAAM,KAAK,mBAAmB;AAC9B,aAAW,CAAC,MAAM,GAAG,KAAK,OAAO,QAAQ,OAAO,WAAW,OAAO,GAAG;AACnE,UAAM;AAAA,MACJ,KAAK,IAAI,MAAM,IAAI,KAAK,QAAQ,CAAC,CAAC,OAAO,IAAI,KAAK,CAAC,EAAE,QAAQ,CAAC,CAAC,KAAK,IAAI,KAAK,CAAC,EAAE,QAAQ,CAAC,CAAC,OAAO,IAAI,CAAC;AAAA,IACxG;AAAA,EACF;AACA,SAAO,MAAM,KAAK,IAAI;AACxB;AAEA,SAAS,cAAc,MAAoE;AACzF,MAAI;AACF,UAAM,SAAS,SAAS,MAAM,KAAK,IAAI,QAAQ,EAAE,KAAK,GAAG,CAAC,IAAI;AAAA,MAC5D,KAAK,EAAE,GAAG,QAAQ,KAAK,UAAU,QAAQ,IAAI,oBAAoB,QAAQ,IAAI,YAAY,GAAG;AAAA,MAC5F,OAAO,CAAC,UAAU,QAAQ,MAAM;AAAA,IAClC,CAAC,EAAE,SAAS,MAAM;AAClB,WAAO,EAAE,QAAQ,QAAQ,IAAI,QAAQ,EAAE;AAAA,EACzC,SAAS,KAAK;AACZ,UAAM,IAAI;AACV,WAAO;AAAA,MACL,QAAQ,EAAE,QAAQ,SAAS,MAAM,KAAK;AAAA,MACtC,QAAQ,EAAE,QAAQ,SAAS,MAAM,KAAK;AAAA,MACtC,QAAQ,EAAE,UAAU;AAAA,IACtB;AAAA,EACF;AACF;AAEA,SAAS,SAAS,KAAqB;AACrC,MAAI,wBAAwB,KAAK,GAAG,EAAG,QAAO;AAC9C,SAAO,IAAI,IAAI,QAAQ,MAAM,KAAK,CAAC;AACrC;;;ACrJO,SAAS,mBACd,MAC8B;AAC9B,SAAO;AAAA,IACL,MAAM,gBAAgB,KAAK,QAAQ,IAAI;AAAA,IACvC,MAAM,QAAQ,EAAE,gBAAgB,UAAU,gBAAgB,OAAO,GAAG;AAClE,aAAO,KAAK,QAAQ,OAAO;AAAA,QACzB,UAAU,SAAS,SAAS,IAAI,WAAY,KAAK,YAAY,CAAC;AAAA,QAC9D;AAAA,QACA;AAAA,QACA;AAAA,MACF,CAAC;AAAA,IACH;AAAA,EACF;AACF;;;ACNA,IAAM,oBACJ;AAqBK,SAAS,WAAW,MAA4C;AACrE,QAAM,YAAY,KAAK,aAAa;AACpC,SAAO;AAAA,IACL,MAAM;AAAA,IACN,MAAM,QAAQ,KAAgD;AAC5D,YAAM,SACJ,OAAO,IAAI,mBAAmB,WAC1B,IAAI,iBACJ,KAAK,UAAU,IAAI,cAAc;AACvC,YAAM,EAAE,KAAK,QAAQ,OAAO,IAAI,cAAc,KAAK,WAAW,KAAK,MAAM;AAEzE,YAAM,aAAa,sBAAsB;AAAA,QACvC;AAAA,QACA,eAAe;AAAA,QACf,WAAW;AAAA,QACX,cAAc;AAAA,QACd,YAAY,IAAI;AAAA,QAChB,oBAAoB,KAAK;AAAA,MAC3B,CAAC;AAED,YAAM,SAAS,MAAM;AAAA,QACnB;AAAA,UACE,OAAO,KAAK;AAAA,UACZ,UAAU;AAAA,YACR,EAAE,MAAM,UAAU,SAAS,kBAAkB;AAAA,YAC7C,EAAE,MAAM,QAAQ,SAAS,WAAW;AAAA,UACtC;AAAA,UACA,UAAU;AAAA,UACV,aAAa,KAAK,eAAe;AAAA,UACjC,WAAW,KAAK,aAAa;AAAA,QAC/B;AAAA,QACA,KAAK;AAAA,MACP;AAEA,YAAM,YAAY,wBAAwB,OAAO,SAAS,IAAI,cAAc;AAC5E,YAAM,MAAwB,CAAC;AAC/B,iBAAW,YAAY,WAAW;AAChC,cAAM,OAAO,OAAO,SAAS,YAAY,WAAW,SAAS,QAAQ,KAAK,IAAI;AAC9E,YAAI,QAAQ,SAAS,UAAU,CAAC,IAAI,SAAS,IAAI,EAAG,KAAI,KAAK,IAAI;AAAA,MACnE;AACA,aAAO;AAAA,IACT;AAAA,EACF;AACF;AAKA,SAAS,cACP,KACA,WACA,YAC6D;AAC7D,QAAM,OAAO,IAAI,QAAQ,GAAG,EAAE;AAC9B,MAAI,CAAC,QAAQ,KAAK,WAAW,WAAW,GAAG;AACzC,WAAO,EAAE,KAAK,CAAC,GAAG,QAAQ,CAAC,GAAG,QAAQ,WAAW;AAAA,EACnD;AACA,QAAM,OAAO,CAAC,GAAG,KAAK,UAAU,EAAE,KAAK,CAAC,GAAG,MAAM,EAAE,YAAY,EAAE,SAAS,EAAE,CAAC;AAC7E,MAAI,CAAC,KAAM,QAAO,EAAE,KAAK,CAAC,GAAG,QAAQ,CAAC,GAAG,QAAQ,WAAW;AAE5D,QAAM,UAAU,CAAC,GAAG,KAAK,SAAS,EAAE,KAAK,CAAC,GAAG,MAAM,EAAE,YAAY,EAAE,SAAS;AAC5E,QAAM,UAAU,CAAC,OAA8D;AAAA,IAC7E,IAAI,EAAE;AAAA,IACN,OAAO,EAAE;AAAA,EACX;AACA,QAAM,MAAM,QAAQ,MAAM,GAAG,SAAS,EAAE,IAAI,OAAO;AACnD,QAAM,SAAS,QAAQ,MAAM,CAAC,SAAS,EAAE,QAAQ,EAAE,IAAI,OAAO;AAE9D,QAAM,UAAU,OAAO,QAAQ,KAAK,UAAU,EAC3C,KAAK,CAAC,GAAG,MAAM,EAAE,CAAC,IAAI,EAAE,CAAC,CAAC,EAC1B,MAAM,GAAG,CAAC,EACV,IAAI,CAAC,CAAC,KAAK,KAAK,MAAM,GAAG,GAAG,KAAK,MAAM,QAAQ,CAAC,CAAC,GAAG;AACvD,QAAM,SACJ,QAAQ,SAAS,IAAI,GAAG,UAAU,+BAA0B,QAAQ,KAAK,IAAI,CAAC,KAAK;AAErF,SAAO,EAAE,KAAK,QAAQ,OAAO;AAC/B;;;ACpHO,SAAS,eACX,OACyB;AAC5B,MAAI,MAAM,WAAW,GAAG;AACtB,UAAM,IAAI,MAAM,wCAAwC;AAAA,EAC1D;AACA,SAAO;AAAA,IACL,MAAM,YAAY,MAAM,IAAI,CAAC,MAAM,EAAE,IAAI,EAAE,KAAK,GAAG,CAAC;AAAA,IACpD,MAAM,OAAO,KAA6D;AACxE,YAAM,UAAwE,CAAC;AAC/E,iBAAW,QAAQ,OAAO;AACxB,cAAM,MAAM,MAAM,KAAK,OAAO,GAAG;AACjC,gBAAQ,KAAK,EAAE,MAAM,IAAI,CAAC;AAAA,MAC5B;AAQA,YAAM,YAAY,QAAQ,IAAI,CAAC,MAAM,EAAE,IAAI,QAAQ;AACnD,YAAM,UAAwB,UAAU,MAAM,CAAC,MAAM,MAAM,MAAM,IAC7D,SACA,UAAU,SAAS,cAAc,IAC/B,iBACA,UAAU,SAAS,eAAe,IAChC,kBACA,UAAU,SAAS,MAAM,IACvB,SACA;AAEV,YAAM,eAAe,QAAQ;AAAA,QAAQ,CAAC,MACpC,EAAE,IAAI,kBAAkB,SAAS,IAC7B,EAAE,IAAI,oBACN,CAAC,EAAE,MAAM,EAAE,KAAK,MAAM,QAAQ,EAAE,IAAI,aAAa,QAAQ,QAAQ,EAAE,IAAI,CAAC;AAAA,MAC9E;AAEA,YAAM,UAAU,QAAQ;AAAA,QAAQ,CAAC,MAC/B,EAAE,IAAI,QAAQ,IAAI,CAAC,WAAW,IAAI,EAAE,KAAK,IAAI,KAAK,MAAM,EAAE;AAAA,MAC5D;AAEA,aAAO;AAAA,QACL,UAAU;AAAA,QACV;AAAA,QACA,mBAAmB;AAAA,QACnB,OAAO,QAAQ,CAAC,GAAG,IAAI;AAAA,MACzB;AAAA,IACF;AAAA,EACF;AACF;;;ACpBO,SAAS,sBACd,SAC4B;AAC5B,QAAM,iBAAiB,QAAQ,kBAAkB;AACjD,QAAM,gBAAgB,QAAQ,8BAA8B;AAE5D,SAAO;AAAA,IACL,MAAM;AAAA,IACN,MAAM,OAAO,KAA6D;AACxE,YAAM,UAAoB,CAAC;AAC3B,YAAM,eAA0E,CAAC;AAKjF,YAAM,oBAAoB;AAAA,QACxB,IAAI;AAAA,QACJ,IAAI,uBAAuB,IAAI;AAAA,QAC/B,QAAQ;AAAA,MACV;AACA,YAAM,qBAAqB;AAAA,QACzB,IAAI;AAAA,QACJ,IAAI;AAAA,QACJ,QAAQ;AAAA,MACV;AACA,YAAM,QAAQ,qBAAqB;AACnC,YAAM,cAAc,SAAS;AAC7B,mBAAa,KAAK;AAAA,QAChB,MAAM;AAAA,QACN,QAAQ;AAAA,QACR,QAAQ,EAAE,mBAAmB,oBAAoB,OAAO,eAAe;AAAA,MACzE,CAAC;AACD,UAAI,CAAC,aAAa;AAChB,gBAAQ,KAAK,iBAAiB,MAAM,QAAQ,CAAC,CAAC,gBAAgB,cAAc,EAAE;AAAA,MAChF;AAGA,YAAM,aACJ,QAAQ,cAAc,UACtB,IAAI,KAAK,YAAY,IAAI,KAAK,YAAY,QAAQ;AACpD,mBAAa,KAAK;AAAA,QAChB,MAAM;AAAA,QACN,QAAQ;AAAA,QACR,QAAQ;AAAA,UACN,cAAc,IAAI,KAAK;AAAA,UACvB,aAAa,IAAI,KAAK;AAAA,UACtB,WAAW,QAAQ;AAAA,QACrB;AAAA,MACF,CAAC;AACD,UAAI,CAAC,YAAY;AACf,gBAAQ;AAAA,UACN,UAAU,IAAI,KAAK,YAAY,IAAI,KAAK,UAAU,QAAQ,CAAC,CAAC,aAAa,QAAQ,SAAS;AAAA,QAC5F;AAAA,MACF;AAGA,YAAM,kBAAkB,QAAQ,iBAC5B,aAAa,IAAI,oBAAoB,QAAQ,cAAc,IAC3D,EAAE,QAAQ,MAAM,UAAU,CAAC,EAAE;AACjC,mBAAa,KAAK;AAAA,QAChB,MAAM;AAAA,QACN,QAAQ,gBAAgB;AAAA,QACxB,QAAQ;AAAA,UACN,UAAU,gBAAgB,SAAS;AAAA,UACnC,QAAQ,gBAAgB,SAAS,MAAM,GAAG,CAAC;AAAA,QAC7C;AAAA,MACF,CAAC;AACD,UAAI,CAAC,gBAAgB,QAAQ;AAC3B,gBAAQ,KAAK,0BAA0B,gBAAgB,SAAS,MAAM,YAAY;AAAA,MACpF;AAGA,UAAI,sBAAkD;AACtD,UAAI,QAAQ,cAAc,QAAQ,WAAW,UAAU,IAAI;AACzD,8BAAsB,oBAAoB,EAAE,MAAM,QAAQ,WAAW,CAAC;AAAA,MACxE;AAIA,YAAM,kBAAkB;AACxB,YAAM,kBAAkB,qBAAqB,YAAY,CAAC,GAAG;AAAA,QAC3D,CAAC,MAAM,EAAE,YAAY;AAAA,MACvB;AACA,YAAM,oBACJ,CAAC,uBACD,CAAC,iBACA,eAAe,WAAW,KAAK,oBAAoB,YAAY;AAClE,mBAAa,KAAK;AAAA,QAChB,MAAM;AAAA,QACN,QAAQ;AAAA,QACR,QAAQ,EAAE,QAAQ,qBAAqB,oBAAoB,eAAe,OAAO;AAAA,MACnF,CAAC;AACD,UAAI,CAAC,mBAAmB;AACtB,gBAAQ;AAAA,UACN,mCAAmC,eAAe,MAAM,sCAAsC,oBAAqB,OAAO;AAAA,QAC5H;AAAA,MACF;AAGA,UAAI,eAAoC;AACxC,UAAI,QAAQ,cAAc,QAAQ,WAAW,UAAU,IAAI;AACzD,uBAAe,YAAY,QAAQ,YAAY,CAAC,CAAC;AAAA,MACnD;AAEA,YAAM,eAAe,cAAc,UAAU,CAAC,GAAG,OAAO,CAAC,MAAM,EAAE,aAAa,OAAO;AACrF,YAAM,aAAa,YAAY,WAAW;AAC1C,mBAAa,KAAK;AAAA,QAChB,MAAM;AAAA,QACN,QAAQ;AAAA,QACR,QAAQ,EAAE,aAAa,cAAc,OAAO,UAAU,GAAG,aAAa,YAAY,OAAO;AAAA,MAC3F,CAAC;AACD,UAAI,CAAC,YAAY;AACf,gBAAQ,KAAK,wBAAwB,YAAY,MAAM,EAAE;AAAA,MAC3D;AAGA,YAAM,YAAY,aAAa,MAAM,CAAC,MAAM,EAAE,MAAM;AACpD,YAAM,WAAW,YAAY,SAAS;AAEtC,aAAO;AAAA,QACL;AAAA,QACA,SAAS,QAAQ,SAAS,IAAI,UAAU,CAAC,kBAAkB;AAAA,QAC3D,mBAAmB;AAAA,QACnB;AAAA,MACF;AAAA,IACF;AAAA,EACF;AACF;AAEA,SAAS,cACP,WACA,mBACA,WACQ;AACR,MAAI,CAAC,aAAa,UAAU,SAAS,EAAG,QAAO;AAC/C,QAAM,cAAc,IAAI,IAAI,UAAU,IAAI,CAAC,MAAM,EAAE,EAAE,CAAC;AACtD,QAAM,aAAuB,CAAC;AAC9B,aAAW,CAAC,QAAQ,MAAM,KAAK,mBAAmB;AAChD,UAAM,aAAa,OAAO,MAAM,GAAG,EAAE,CAAC,KAAK;AAC3C,QAAI,CAAC,YAAY,IAAI,UAAU,EAAG;AAClC,UAAM,iBAAiB,OAAO,OAAO,MAAM,EAAE,IAAI,CAAC,MAAM,EAAE,SAAS;AACnE,QAAI,eAAe,WAAW,EAAG;AACjC,eAAW,KAAK,eAAe,OAAO,CAAC,GAAG,MAAM,IAAI,GAAG,CAAC,IAAI,eAAe,MAAM;AAAA,EACnF;AACA,MAAI,WAAW,WAAW,EAAG,QAAO;AACpC,SAAO,WAAW,OAAO,CAAC,GAAG,MAAM,IAAI,GAAG,CAAC,IAAI,WAAW;AAC5D;AAEA,SAAS,aACP,WACA,SAC8E;AAC9E,QAAM,WAA0D,CAAC;AACjE,aAAW,CAAC,SAAS,QAAQ,KAAK,WAAW;AAC3C,UAAM,OAAO,YAAY,QAAQ;AACjC,QAAI,SAAS,OAAW;AACxB,eAAW,UAAU,SAAS;AAC5B,YAAM,UAAU,mBAAmB,MAAM,CAAC,GAAG,MAAM;AACnD,UAAI,CAAC,QAAQ,QAAQ;AACnB,iBAAS,KAAK,EAAE,YAAY,OAAO,IAAI,QAAQ,QAAQ,UAAU,wBAAwB,CAAC;AAAA,MAC5F;AAAA,IACF;AAAA,EACF;AACA,SAAO,EAAE,QAAQ,SAAS,WAAW,GAAG,SAAS;AACnD;AAEA,SAAS,YAAY,UAAuC;AAC1D,MAAI,OAAO,aAAa,SAAU,QAAO;AACzC,MAAI,YAAY,OAAO,aAAa,UAAU;AAC5C,UAAM,MAAM;AACZ,QAAI,OAAO,IAAI,SAAS,SAAU,QAAO,IAAI;AAC7C,QAAI,OAAO,IAAI,WAAW,SAAU,QAAO,IAAI;AAC/C,QAAI,OAAO,IAAI,YAAY,SAAU,QAAO,IAAI;AAAA,EAClD;AACA,SAAO;AACT;;;AC5MO,SAAS,YACd,SAC4B;AAC5B,QAAM,iBAAiB,QAAQ,kBAAkB;AACjD,SAAO;AAAA,IACL,MAAM;AAAA,IACN,MAAM,OAAO,KAA6D;AACxE,YAAM,cAAc,IAAI,IAAI,QAAQ,UAAU,IAAI,CAAC,MAAM,EAAE,EAAE,CAAC;AAG9D,YAAM,WAAW,iBAAiB,IAAI,uBAAuB,IAAI,aAAa,WAAW;AACzF,YAAM,YAAY,iBAAiB,IAAI,aAAa,WAAW;AAC/D,YAAM,QAAQ,YAAY;AAC1B,YAAM,SAAS,SAAS;AACxB,aAAO;AAAA,QACL,UAAU,SAAS,SAAS;AAAA,QAC5B,SAAS,SACL,CAAC,kBAAkB,MAAM,QAAQ,CAAC,CAAC,WAAM,cAAc,EAAE,IACzD,CAAC,kBAAkB,MAAM,QAAQ,CAAC,CAAC,MAAM,cAAc,EAAE;AAAA,QAC7D,mBAAmB;AAAA,UACjB,EAAE,MAAM,eAAe,QAAQ,QAAQ,EAAE,UAAU,WAAW,OAAO,eAAe,EAAE;AAAA,QACxF;AAAA,QACA;AAAA,MACF;AAAA,IACF;AAAA,EACF;AACF;AAEA,SAAS,iBACP,mBACA,aACQ;AACR,QAAM,aAAuB,CAAC;AAC9B,aAAW,CAAC,QAAQ,MAAM,KAAK,mBAAmB;AAChD,UAAM,aAAa,OAAO,MAAM,GAAG,EAAE,CAAC,KAAK;AAC3C,QAAI,CAAC,YAAY,IAAI,UAAU,EAAG;AAClC,UAAM,OAAO,OAAO,OAAO,MAAM,EAAE,IAAI,CAAC,MAAM,EAAE,SAAS;AACzD,QAAI,KAAK,SAAS,EAAG,YAAW,KAAK,KAAK,OAAO,CAAC,GAAG,MAAM,IAAI,GAAG,CAAC,IAAI,KAAK,MAAM;AAAA,EACpF;AACA,SAAO,WAAW,WAAW,IAAI,IAAI,WAAW,OAAO,CAAC,GAAG,MAAM,IAAI,GAAG,CAAC,IAAI,WAAW;AAC1F;;;AC1BA,SAAS,kBAAkB;AAC3B,SAAS,YAAY,WAAW,cAAc,iBAAAA,sBAAqB;AACnE,SAAS,QAAAC,aAAY;AAmBd,IAAM,4BAAN,cAAwC,MAAM;AAAA,EACnD,YACkB,MAChB,SACA;AACA,UAAM,OAAO;AAHG;AAIhB,SAAK,OAAO;AAAA,EACd;AAAA,EALkB;AAMpB;AAQO,IAAM,yBAAN,MAA6D;AAAA,EAIlE,YAA6B,SAAwC;AAAxC;AAC3B,QAAI,CAAC,WAAW,QAAQ,IAAI,EAAG,WAAU,QAAQ,MAAM,EAAE,WAAW,KAAK,CAAC;AAC1E,SAAK,MAAM,QAAQ,OAAO,KAAK;AAAA,EACjC;AAAA,EAH6B;AAAA,EAHZ;AAAA,EACA,aAAa,oBAAI,IAA4B;AAAA,EAO9D,MAAM,QAAQ,OAA4C;AACxD,SAAK,iBAAiB,KAAK;AAC3B,SAAK,gBAAgB,KAAK;AAC1B,UAAM,SAAS,KAAK,SAAS,KAAK;AAClC,UAAM,OAAO,KAAK,cAAc,MAAM,MAAM;AAC5C,UAAM,OAAO,GAAG,KAAK,UAAU,MAAM,CAAC;AAAA;AAItC,eAAW,MAAM,IAAI;AAAA,EACvB;AAAA,EAEA,MAAM,OAAO,MAAmE;AAC9E,QAAI,CAAC,KAAK,OAAO;AACf,YAAM,IAAI;AAAA,QACR;AAAA,QACA;AAAA,MACF;AAAA,IACF;AACA,QAAI,CAAC,KAAK,gBAAgB;AACxB,YAAM,IAAI;AAAA,QACR;AAAA,QACA;AAAA,MACF;AAAA,IACF;AAEA,UAAM,MAA+B,CAAC;AACtC,eAAW,UAAU,aAAa;AAGhC,UAAI,KAAK,UAAU,WAAW,WAAW,oBAAoB;AAC3D,cAAM,WAAW,qBAAqB,KAAK,QAAQ,QAAQ,kBAAkB;AAC7E,YAAI,CAAC,SAAU;AAAA,MACjB;AACA,YAAM,OAAO,KAAK,cAAc,MAAM;AACtC,UAAI,CAAC,WAAW,IAAI,EAAG;AACvB,YAAM,QAAQ,aAAa,MAAM,MAAM,EAAE,MAAM,IAAI,EAAE,OAAO,OAAO;AACnE,iBAAW,QAAQ,OAAO;AACxB,YAAI;AACJ,YAAI;AACF,mBAAS,KAAK,MAAM,IAAI;AAAA,QAC1B,QAAQ;AACN;AAAA,QACF;AACA,YAAI,CAAC,cAAc,QAAQ,MAAM,MAAM,EAAG;AAC1C,YAAI,KAAK,MAAM;AAAA,MACjB;AAAA,IACF;AAGA,QAAI,KAAK,CAAC,GAAG,MAAM;AACjB,UAAI,EAAE,eAAe,EAAE,WAAY,QAAO,EAAE,WAAW,cAAc,EAAE,UAAU;AACjF,aAAO,EAAE,WAAW,cAAc,EAAE,UAAU;AAAA,IAChD,CAAC;AAED,WAAO,IAAI,MAAM,GAAG,KAAK,KAAK;AAAA,EAChC;AAAA,EAEA,MAAM,OAAmF;AACvF,UAAM,WAAmC,CAAC;AAC1C,QAAI,QAAQ;AACZ,eAAW,UAAU,aAAa;AAChC,YAAM,OAAO,KAAK,cAAc,MAAM;AACtC,UAAI,CAAC,WAAW,IAAI,GAAG;AACrB,iBAAS,MAAM,IAAI;AACnB;AAAA,MACF;AACA,YAAM,QAAQ,aAAa,MAAM,MAAM,EAAE,MAAM,IAAI,EAAE,OAAO,OAAO,EAAE;AACrE,eAAS,MAAM,IAAI;AACnB,eAAS;AAAA,IACX;AAIA,WAAO,EAAE,OAAO,OAAO,MAAM,OAAO,SAAS;AAAA,EAC/C;AAAA,EAEQ,iBAAiB,OAAmC;AAC1D,QAAI,CAAC,MAAM,QAAQ;AACjB,YAAM,IAAI;AAAA,QACR;AAAA,QACA;AAAA,MACF;AAAA,IACF;AACA,QAAI,CAAC,MAAM,qBAAqB,MAAM,kBAAkB,WAAW,GAAG;AACpE,YAAM,IAAI;AAAA,QACR;AAAA,QACA;AAAA,MACF;AAAA,IACF;AACA,QAAI,CAAC,MAAM,YAAY;AACrB,YAAM,IAAI;AAAA,QACR;AAAA,QACA;AAAA,MACF;AAAA,IACF;AACA,QAAI,CAAC,MAAM,iBAAiB;AAC1B,YAAM,IAAI;AAAA,QACR;AAAA,QACA;AAAA,MACF;AAAA,IACF;AACA,QAAI,CAAC,YAAY,SAAS,MAAM,MAAM,GAAG;AACvC,YAAM,IAAI;AAAA,QACR;AAAA,QACA,+CAA+C,YAAY,KAAK,IAAI,CAAC;AAAA,MACvE;AAAA,IACF;AAAA,EACF;AAAA,EAEQ,gBAAgB,OAAmC;AACzD,UAAM,MAAM,KAAK,QAAQ;AACzB,QAAI,CAAC,OAAO,CAAC,MAAM,gBAAiB;AACpC,UAAM,MAAM,KAAK,IAAI;AACrB,UAAM,WAAW;AACjB,QAAI,QAAQ,KAAK,WAAW,IAAI,MAAM,eAAe;AACrD,QAAI,CAAC,SAAS,MAAM,MAAM,iBAAiB,UAAU;AACnD,cAAQ,EAAE,QAAQ,MAAM,iBAAiB,eAAe,KAAK,OAAO,EAAE;AACtE,WAAK,WAAW,IAAI,MAAM,iBAAiB,KAAK;AAAA,IAClD;AACA,QAAI,MAAM,SAAS,KAAK;AACtB,YAAM,IAAI;AAAA,QACR;AAAA,QACA,gCAAgC,MAAM,eAAe,aAAa,GAAG;AAAA,MACvE;AAAA,IACF;AACA,UAAM,SAAS;AAAA,EACjB;AAAA,EAEQ,SAAS,OAAoD;AACnE,UAAM,aAAa;AAAA,MACjB,KAAK,UAAU;AAAA,QACb,IAAI,MAAM,SAAS;AAAA,QACnB,KAAK,MAAM;AAAA,QACX,IAAI,MAAM;AAAA,QACV,KAAK,MAAM;AAAA,MACb,CAAC;AAAA,IACH;AAGA,WAAO;AAAA,MACL,GAAG;AAAA,MACH;AAAA,MACA,OAAO;AAAA,IACT;AAAA,EACF;AAAA,EAEQ,cAAc,QAAwB;AAC5C,WAAOA,MAAK,KAAK,QAAQ,MAAM,GAAG,MAAM,QAAQ;AAAA,EAClD;AACF;AAEA,IAAM,cAAgD;AAAA,EACpD;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AACF;AAEA,SAAS,qBACP,QACA,QACS;AACT,MAAI,CAAC,OAAQ,QAAO;AACpB,MAAI,MAAM,QAAQ,MAAM,EAAG,QAAO,OAAO,SAAS,MAAM;AACxD,SAAO,WAAW;AACpB;AAEA,SAAS,cACP,QACA,MACA,QACS;AAET,MAAI,KAAK,UAAU,WAAW,OAAO,cAAc,KAAK,eAAgB,QAAO;AAC/E,MAAI,KAAK,UAAU,UAAU,OAAO,aAAa,KAAK,eAAgB,QAAO;AAE7E,QAAM,IAAI,KAAK;AACf,MAAI,CAAC,EAAG,QAAO;AACf,MAAI,EAAE,QAAQ,OAAO,SAAS,SAAS,EAAE,KAAM,QAAO;AACtD,MAAI,EAAE,QAAQ;AACZ,UAAM,UAAU,MAAM,QAAQ,EAAE,MAAM,IAAI,EAAE,SAAS,CAAC,EAAE,MAAM;AAC9D,QAAI,CAAC,QAAQ,SAAS,MAAe,EAAG,QAAO;AAAA,EACjD;AACA,MAAI,EAAE,iBAAiB,UAAa,EAAE,iBAAiB,QAAW;AAChE,UAAM,aAAa,OAAO,OAAO,OAAO,WAAW,EAAE,IAAI,CAAC,MAAM,EAAE,SAAS;AAC3E,UAAM,MAAM,WAAW,WAAW,IAAI,IAAI,KAAK,IAAI,GAAG,UAAU;AAChE,QAAI,EAAE,iBAAiB,UAAa,MAAM,EAAE,aAAc,QAAO;AACjE,QAAI,EAAE,iBAAiB,UAAa,MAAM,EAAE,aAAc,QAAO;AAAA,EACnE;AACA,SAAO;AACT;AAEA,SAAS,OAAO,OAAuB;AACrC,SAAO,WAAW,QAAQ,EAAE,OAAO,KAAK,EAAE,OAAO,KAAK,EAAE,MAAM,GAAG,EAAE;AACrE;AAEA,SAAS,WAAW,MAAc,MAAoB;AACpD,MAAI,WAAW,IAAI,GAAG;AACpB,UAAM,WAAW,aAAa,MAAM,MAAM;AAC1C,IAAAD,eAAc,MAAM,WAAW,IAAI;AAAA,EACrC,OAAO;AACL,IAAAA,eAAc,MAAM,IAAI;AAAA,EAC1B;AACF;;;ACtQA,eAAsB,QACpB,MAC+C;AAC/C,SAAO,YAAY,IAAI;AACzB;;;ACFA,SAAS,cAAAE,mBAAkB;AAkD3B,eAAsB,gBACpB,MACsD;AACtD,QAAM,cAAc,KAAK,eAAe;AAGxC,QAAM,mBAAmB,MAAM,YAAkC;AAAA,IAC/D,GAAG;AAAA,IACH,UAAU,CAAC,UAAU,QAAQ,KAAK,oBAAoB,KAAK,iBAAiB,UAAU,GAAG;AAAA,IACzF,QAAQ,GAAG,KAAK,MAAM;AAAA,EACxB,CAAC;AAED,QAAM,cAA0E,CAAC;AACjF,QAAM,UAA8B,CAAC;AACrC,MAAI,kBAAoC,CAAC,KAAK,eAAe;AAC7D,MAAI,gBAAgB,KAAK;AACzB,MAAI,oBAAoB,YAAY,KAAK,eAAe;AACxD,MAAI,kBAAkBC,eAAc,gBAAgB;AAEpD,WAAS,MAAM,GAAG,MAAM,KAAK,gBAAgB,OAAO;AAElD,QAAI,KAAK,OAAO,SAAS,EAAE,QAAQ,CAAC,EAAE,KAAM;AAI5C,UAAM,aAAa,MAAM,KAAK,OAAO,QAAQ;AAAA,MAC3C,gBAAgB,gBAAgB,CAAC,KAAK,KAAK;AAAA,MAC3C;AAAA,MACA,UAAU,CAAC;AAAA,MACX,gBAAgB,KAAK;AAAA,MACrB,YAAY;AAAA,MACZ,QAAQ,IAAI,gBAAgB,EAAE;AAAA,MAC9B,QAAQ,KAAK;AAAA,MACb,SAAS,KAAK,gBAAgB,KAAK,iBAAiB,QAAQ,KAAK,eAAe;AAAA,MAChF,qBAAqB,KAAK;AAAA,IAC5B,CAAC;AAGD,UAAM,iBAKD,CAAC;AACN,aAAS,IAAI,GAAG,IAAI,WAAW,QAAQ,KAAK;AAC1C,YAAM,UAAU,WAAW,CAAC;AAC5B,YAAM,OAAO,YAAY,OAAO;AAChC,YAAM,WAAW,MAAM,YAAkC;AAAA,QACvD,GAAG;AAAA,QACH,UAAU,CAAC,UAAU,QAAQ,KAAK,oBAAoB,SAAS,UAAU,GAAG;AAAA,QAC5E,QAAQ,GAAG,KAAK,MAAM,QAAQ,GAAG,cAAc,CAAC;AAAA,MAClD,CAAC;AACD,YAAM,YAAYA,eAAc,QAAQ;AACxC,qBAAe,KAAK,EAAE,aAAa,MAAM,SAAS,UAAU,UAAU,CAAC;AAAA,IACzE;AAGA,mBAAe,KAAK,CAAC,GAAG,MAAM,EAAE,YAAY,EAAE,SAAS;AACvD,UAAM,WAAW,eAAe,MAAM,GAAG,WAAW;AACpD,sBAAkB,SAAS,IAAI,CAAC,MAAM,EAAE,OAAO;AAC/C,UAAM,MAAM,eAAe,CAAC;AAC5B,QAAI,OAAO,IAAI,YAAY,iBAAiB;AAC1C,sBAAgB,IAAI;AACpB,0BAAoB,IAAI;AACxB,wBAAkB,IAAI;AAAA,IACxB;AAEA,UAAM,SAA2B;AAAA,MAC/B,iBAAiB;AAAA,MACjB,YAAY,eAAe,IAAI,CAAC,MAAM;AACpC,cAAM,YAAY,mBAAmB,EAAE,QAAQ;AAC/C,eAAO;AAAA,UACL,aAAa,EAAE;AAAA,UACf,WAAW,EAAE;AAAA,UACb,MAAM,CAAC,EAAE,WAAW,EAAE,SAAS;AAAA,UAC/B,YAAY,UAAU;AAAA,UACtB,WAAW,UAAU;AAAA,QACvB;AAAA,MACF,CAAC;AAAA,MACD,UAAU,SAAS,IAAI,CAAC,MAAM,EAAE,WAAW;AAAA,IAC7C;AACA,YAAQ,KAAK,MAAM;AACnB,gBAAY,KAAK;AAAA,MACf;AAAA,MACA,UAAU,eAAe,IAAI,CAAC,OAAO;AAAA,QACnC,aAAa,EAAE;AAAA,QACf,SAAS,EAAE;AAAA,QACX,UAAU,EAAE;AAAA,MACd,EAAE;AAAA,IACJ,CAAC;AAAA,EACH;AAEA,SAAO;AAAA,IACL;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,EACF;AACF;AAEO,SAAS,YAAY,SAAiC;AAG3D,QAAM,WACJ,OAAO,YAAY,WACf,UACA,KAAK,UAAU;AAAA,IACb,MAAM,QAAQ;AAAA,IACd,aAAa,QAAQ;AAAA,IACrB,SAAS,QAAQ,WAAW;AAAA,EAC9B,CAAC;AACP,SAAOC,YAAW,QAAQ,EAAE,OAAO,QAAQ,EAAE,OAAO,KAAK,EAAE,MAAM,GAAG,EAAE;AACxE;AAEA,SAASD,eACP,UACQ;AACR,QAAM,aAAuB,CAAC;AAC9B,aAAW,QAAQ,SAAS,OAAO;AACjC,UAAM,iBAAiB,OAAO,OAAO,KAAK,WAAW,EAAE,IAAI,CAAC,MAAM,EAAE,SAAS;AAC7E,QAAI,eAAe,SAAS,GAAG;AAC7B,iBAAW,KAAK,eAAe,OAAO,CAAC,GAAG,MAAM,IAAI,GAAG,CAAC,IAAI,eAAe,MAAM;AAAA,IACnF;AAAA,EACF;AACA,SAAO,WAAW,WAAW,IAAI,IAAI,WAAW,OAAO,CAAC,GAAG,MAAM,IAAI,GAAG,CAAC,IAAI,WAAW;AAC1F;AAIA,SAAS,mBACP,UAIA;AACA,QAAM,UAAkC,CAAC;AACzC,QAAM,YAAoC,CAAC;AAC3C,QAAM,aAAa,oBAAI,IAAsB;AAC7C,aAAW,QAAQ,SAAS,OAAO;AACjC,UAAM,cAAc,OAAO,OAAO,KAAK,WAAW;AAClD,QAAI,YAAY,WAAW,EAAG;AAC9B,UAAM,gBAAgB,YAAY,OAAO,CAAC,GAAG,MAAM,IAAI,EAAE,WAAW,CAAC,IAAI,YAAY;AACrF,UAAM,MAAM,WAAW,IAAI,KAAK,UAAU,KAAK,CAAC;AAChD,QAAI,KAAK,aAAa;AACtB,eAAW,IAAI,KAAK,YAAY,GAAG;AACnC,eAAW,SAAS,aAAa;AAC/B,iBAAW,CAAC,KAAK,KAAK,KAAK,OAAO,QAAQ,MAAM,UAAU,GAAG;AAC3D,gBAAQ,GAAG,KAAK,QAAQ,GAAG,KAAK,KAAK;AACrC,kBAAU,GAAG,KAAK,UAAU,GAAG,KAAK,KAAK;AAAA,MAC3C;AAAA,IACF;AAAA,EACF;AACA,QAAM,aAAqC,CAAC;AAC5C,aAAW,OAAO,OAAO,KAAK,OAAO,GAAG;AACtC,UAAM,QAAQ,UAAU,GAAG,KAAK;AAChC,eAAW,GAAG,IAAI,QAAQ,KAAK,QAAQ,GAAG,KAAK,KAAK,QAAQ;AAAA,EAC9D;AACA,QAAM,YAAY,CAAC,GAAG,WAAW,QAAQ,CAAC,EAAE,IAAI,CAAC,CAAC,YAAY,KAAK,OAAO;AAAA,IACxE;AAAA,IACA,WAAW,MAAM,OAAO,CAAC,GAAG,MAAM,IAAI,GAAG,CAAC,IAAI,MAAM;AAAA,EACtD,EAAE;AACF,SAAO,EAAE,YAAY,UAAU;AACjC;;;ACxKA,eAAsB,mBACpB,MACyD;AAGzD,MAAK,KAAa,kBAAkB,UAAU;AAC5C,UAAM,IAAI;AAAA,MACR;AAAA,IACF;AAAA,EACF;AAKA,MAAI,KAAK,YAAY,SAAS,KAAK,QAAQ;AACzC,UAAM,IAAI;AAAA,MACR;AAAA,IACF;AAAA,EACF;AACA,MAAI,KAAK,kBAAkB,SAAS,CAAC,KAAK,WAAW,CAAC,KAAK,SAAS;AAClE,UAAM,IAAI,MAAM,mEAAmE;AAAA,EACrF;AAGA,QAAM,eAAe,MAAM,gBAAgB,IAAI;AAG/C,QAAM,EAAE,aAAAE,aAAY,IAAI,MAAM,OAAO,6BAAiB;AAEtD,QAAM,oBAAoB,MAAMA,aAAkC;AAAA,IAChE,GAAG;AAAA,IACH,WAAW,KAAK;AAAA,IAChB,UAAU,CAAC,UAAU,QAAQ,KAAK,oBAAoB,KAAK,iBAAiB,UAAU,GAAG;AAAA,IACzF,QAAQ,GAAG,KAAK,MAAM;AAAA,EACxB,CAAC;AAED,QAAM,kBAAkB,MAAMA,aAAkC;AAAA,IAC9D,GAAG;AAAA,IACH,WAAW,KAAK;AAAA,IAChB,UAAU,CAAC,UAAU,QACnB,KAAK,oBAAoB,aAAa,eAAe,UAAU,GAAG;AAAA,IACpE,QAAQ,GAAG,KAAK,MAAM;AAAA,EACxB,CAAC;AAUD,QAAM,qBAAqB,oBAAI,IAAuB;AACtD,QAAM,oBAAoB,oBAAI,IAAuB;AACrD,QAAM,cAAwB,oBAAI,IAAI;AACtC,QAAM,sBAAgC,oBAAI,IAAI;AAC9C,aAAW,QAAQ,gBAAgB,OAAO;AACxC,uBAAmB,IAAI,KAAK,QAAQ,KAAK,QAAQ;AACjD,gBAAY,IAAI,KAAK,QAAQ,KAAK,WAAW;AAAA,EAC/C;AACA,aAAW,QAAQ,kBAAkB,OAAO;AAC1C,sBAAkB,IAAI,KAAK,QAAQ,KAAK,QAAQ;AAChD,wBAAoB,IAAI,KAAK,QAAQ,KAAK,WAAW;AAAA,EACvD;AAEA,QAAM,aAAa,MAAM,KAAK,KAAK,OAAO;AAAA,IACxC;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,IACA,WAAW,KAAK;AAAA,IAChB,MAAM;AAAA,MACJ,WAAW,gBAAgB,WAAW;AAAA,MACtC,UAAU,kBAAkB,WAAW;AAAA,IACzC;AAAA,IACA,QAAQ,IAAI,gBAAgB,EAAE;AAAA,EAChC,CAAC;AAGD,MAAI;AACJ,MAAI,KAAK,kBAAkB,QAAQ,WAAW,aAAa,QAAQ;AACjE,UAAM,SAAS,KAAK,sBAAsB;AAC1C,UAAM,eAAe,OAAO,aAAa,eAAe,KAAK,eAAe;AAC5E,eAAW,WAAW;AAAA,MACpB,QAAQ;AAAA,MACR,MAAM;AAAA,MACN;AAAA,MACA,SAAS,KAAK;AAAA,MACd,QAAQ,KAAK;AAAA,IACf,CAAC;AAAA,EACH;AAEA,SAAO;AAAA,IACL,GAAG;AAAA,IACH;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,EACF;AACF;AAEA,SAAS,kBAAkB,eAA+B,iBAAyC;AAGjG,MAAI,OAAO,kBAAkB,YAAY,OAAO,oBAAoB,UAAU;AAC5E,UAAM,MAAM,CAAC,MACX,OAAO,MAAM,WACT,qBACA,YAAY,EAAE,WAAW,GAAG,EAAE,UAAU,SAAS,EAAE,OAAO,KAAK,EAAE,GAAG,EAAE,UAAU;AAAA,EAAK,EAAE,OAAO,KAAK,EAAE;AAC3G,WAAO;AAAA,EAAiB,IAAI,eAAe,CAAC;AAAA;AAAA,EAAiB,IAAI,aAAa,CAAC;AAAA,EACjF;AACA,QAAM,QAAkB,CAAC;AACzB,QAAM,KAAK,cAAc;AACzB,QAAM,KAAK,YAAY;AACvB,aAAW,KAAK,gBAAgB,MAAM,IAAI,EAAG,OAAM,KAAK,KAAK,CAAC,EAAE;AAChE,aAAW,KAAK,cAAc,MAAM,IAAI,EAAG,OAAM,KAAK,KAAK,CAAC,EAAE;AAC9D,SAAO,MAAM,KAAK,IAAI;AACxB;;;ACvKA,SAAS,oBAAoB;AAC7B,SAAS,cAAAC,mBAAkB;AAC3B,SAAS,UAAU,YAAY,QAAAC,aAAY;AAuBpC,IAAM,uBAAN,cAAmC,MAAM;AAAA,EAC9C,YACE,SACS,OACT;AACA,UAAM,OAAO;AAFJ;AAGT,SAAK,OAAO;AAAA,EACd;AAAA,EAJW;AAKb;AAaA,SAAS,WAAW,MAAgB,KAAqB;AACvD,MAAI;AACF,WAAO,aAAa,OAAO,MAAM,EAAE,KAAK,UAAU,OAAO,CAAC,EAAE,KAAK;AAAA,EACnE,SAAS,KAAK;AACZ,UAAM,SACJ,OAAO,OAAO,QAAQ,YAAY,YAAY,MAC1C,OAAQ,IAA4B,MAAM,IAC1C;AACN,UAAM,IAAI,qBAAqB,OAAO,KAAK,KAAK,GAAG,CAAC,YAAY,UAAU,OAAO,GAAG,CAAC,IAAI,GAAG;AAAA,EAC9F;AACF;AAGA,SAAS,KAAK,OAAuB;AACnC,SACE,MACG,YAAY,EACZ,QAAQ,eAAe,GAAG,EAC1B,QAAQ,YAAY,EAAE,EACtB,MAAM,GAAG,EAAE,KAAK;AAEvB;AAEO,SAAS,mBAAmB,MAAkD;AACnF,QAAM,MAAM,KAAK,OAAO;AACxB,QAAM,cAAc,KAAK,eAAeA,MAAK,KAAK,UAAU,YAAY;AACxE,QAAM,eAAe,KAAK,gBAAgB;AAE1C,SAAO;AAAA,IACL,MAAM,OAAO,EAAE,SAAS,MAAM,GAAG;AAC/B,YAAM,KAAK,GAAG,KAAK,KAAK,CAAC,IAAI,KAAK,IAAI,EAAE,SAAS,EAAE,CAAC,IAAI,KAAK,OAAO,EAAE,SAAS,EAAE,EAAE,MAAM,GAAG,CAAC,CAAC;AAC9F,YAAM,SAAS,GAAG,YAAY,IAAI,EAAE;AACpC,YAAM,OAAOA,MAAK,aAAa,EAAE;AACjC,UAAI,CAAC,YAAY,OAAO,MAAM,QAAQ,MAAM,OAAO,GAAG,KAAK,QAAQ;AACnE,aAAO,EAAE,MAAM,QAAQ,QAAQ;AAAA,IACjC;AAAA,IAEA,MAAM,SAAS,UAAU,SAAS;AAGhC,YAAM,SAAS,IAAI,CAAC,UAAU,aAAa,GAAG,SAAS,IAAI;AAC3D,UAAI,OAAO,SAAS,GAAG;AACrB,YAAI,CAAC,OAAO,IAAI,GAAG,SAAS,IAAI;AAChC,YAAI,CAAC,UAAU,MAAM,OAAO,GAAG,SAAS,IAAI;AAAA,MAC9C;AACA,aAAO;AAAA,QACL,MAAM;AAAA,QACN,aAAa,SAAS;AAAA,QACtB,SAAS,SAAS;AAAA,QAClB;AAAA,MACF;AAAA,IACF;AAAA,IAEA,MAAM,QAAQ,UAAU;AAGtB,UAAI,CAAC,YAAY,UAAU,WAAW,SAAS,IAAI,GAAG,KAAK,QAAQ;AACnE,UAAI,CAAC,UAAU,MAAM,SAAS,MAAM,GAAG,KAAK,QAAQ;AAAA,IACtD;AAAA,EACF;AACF;AAKO,SAAS,oBAAoB,SAAsB,aAA8B;AACtF,MAAI,WAAW,QAAQ,WAAW,KAAKD,YAAW,QAAQ,WAAW,EAAG,QAAO,QAAQ;AACvF,MAAI,YAAa,QAAOC,MAAK,aAAa,SAAS,QAAQ,WAAW,CAAC;AACvE,SAAO,QAAQ;AACjB;","names":["writeFileSync","join","createHash","meanComposite","createHash","runCampaign","existsSync","join"]}
1
+ {"version":3,"sources":["../../src/campaign/auto-pr.ts","../../src/campaign/drivers/evolutionary.ts","../../src/campaign/drivers/gepa.ts","../../src/campaign/gates/compose.ts","../../src/campaign/gates/default-production-gate.ts","../../src/campaign/gates/heldout-gate.ts","../../src/campaign/labeled-store/fs-adapter.ts","../../src/campaign/presets/run-eval.ts","../../src/campaign/presets/run-optimization.ts","../../src/campaign/presets/run-improvement-loop.ts","../../src/campaign/worktree/index.ts"],"sourcesContent":["/**\n * @experimental\n *\n * `openAutoPr` — thin shell-out helper for the `runImprovementLoop` preset's\n * `autoOnPromote: 'pr'` mode. Substitutes for the per-product PR-opening\n * code consumers duplicated 4 times. The PR body includes the campaign's\n * manifest hash, gate verdict, and scorecard summary so reviewers can see\n * exactly what was promoted + why.\n *\n * NOT a deploy mechanism — this only OPENS a PR. The human reviews + merges.\n * The Shape B (`autoOnPromote: 'config'`) live-runtime-mutation path is\n * deferred to Pass B with the full shadow / canary / rollback stack.\n */\n\nimport { execSync } from 'node:child_process'\nimport { writeFileSync } from 'node:fs'\nimport { tmpdir } from 'node:os'\nimport { join } from 'node:path'\nimport type { CampaignResult, GateResult, Scenario } from './types'\n\nexport interface OpenAutoPrOptions<TArtifact, TScenario extends Scenario> {\n /** Campaign result to attach to the PR. */\n result: CampaignResult<TArtifact, TScenario>\n /** Gate verdict explaining the promotion. Substrate refuses to open a PR\n * when `gate.decision !== 'ship'` — fails loud. */\n gate: GateResult\n /** Promoted surface diff — typically the new system prompt addendum or\n * full profile diff. Substrate writes it as the PR body. */\n promotedDiff: string\n /** GH owner/repo target (e.g., `tangle-network/gtm-agent`). */\n ghOwner: string\n ghRepo: string\n /** Branch name for the PR. Default `auto/<manifestHash[:12]>`. */\n branch?: string\n /** PR title. Default includes manifest hash. */\n title?: string\n /** Whether to actually open the PR or just dry-run. Default reads\n * `GH_AUTO_PR_TOKEN` env — present = open, absent = dry-run. */\n dryRun?: boolean\n /** Test seam — substitute `gh pr create` invocation. */\n ghExec?: (args: string[]) => { stdout: string; stderr: string; status: number }\n}\n\nexport interface OpenAutoPrResult {\n opened: boolean\n prUrl?: string\n dryRun: boolean\n reason: string\n}\n\nexport function openAutoPr<TArtifact, TScenario extends Scenario>(\n options: OpenAutoPrOptions<TArtifact, TScenario>,\n): OpenAutoPrResult {\n if (options.gate.decision !== 'ship') {\n return {\n opened: false,\n dryRun: false,\n reason: `gate verdict was \"${options.gate.decision}\" — refusing to open PR`,\n }\n }\n\n const dryRun = options.dryRun ?? !process.env.GH_AUTO_PR_TOKEN\n const branch = options.branch ?? `auto/${options.result.manifestHash.slice(0, 12)}`\n const title =\n options.title ?? `auto: campaign ${options.result.manifestHash.slice(0, 8)} promoted by gate`\n\n const body = renderPrBody(options.result, options.gate, options.promotedDiff)\n const bodyPath = join(tmpdir(), `auto-pr-body-${Date.now()}.md`)\n writeFileSync(bodyPath, body)\n\n if (dryRun) {\n return {\n opened: false,\n dryRun: true,\n reason: `dry-run (GH_AUTO_PR_TOKEN not set). Would create PR on ${options.ghOwner}/${options.ghRepo} branch ${branch}. Body at ${bodyPath}.`,\n }\n }\n\n const ghExec = options.ghExec ?? defaultGhExec\n const result = ghExec([\n 'pr',\n 'create',\n '--repo',\n `${options.ghOwner}/${options.ghRepo}`,\n '--head',\n branch,\n '--title',\n title,\n '--body-file',\n bodyPath,\n ])\n if (result.status !== 0) {\n return {\n opened: false,\n dryRun: false,\n reason: `gh pr create failed (exit ${result.status}): ${result.stderr.slice(0, 400)}`,\n }\n }\n const prUrl = result.stdout.trim()\n return { opened: true, prUrl, dryRun: false, reason: 'PR opened' }\n}\n\nfunction renderPrBody<TArtifact, TScenario extends Scenario>(\n result: CampaignResult<TArtifact, TScenario>,\n gate: GateResult,\n diff: string,\n): string {\n const lines: string[] = []\n lines.push(`## Automated promotion by \\`runImprovementLoop\\``)\n lines.push('')\n lines.push(`**Manifest**: \\`${result.manifestHash}\\``)\n lines.push(`**Seed**: ${result.seed}`)\n lines.push(`**Duration**: ${Math.round(result.durationMs / 1000)}s`)\n lines.push(\n `**Cells**: executed ${result.aggregates.cellsExecuted}, cached ${result.aggregates.cellsCached}, skipped ${result.aggregates.cellsSkipped}, failed ${result.aggregates.cellsFailed}`,\n )\n lines.push(`**Total spend**: $${result.aggregates.totalCostUsd.toFixed(2)}`)\n lines.push('')\n lines.push(`### Gate verdict: \\`${gate.decision}\\``)\n lines.push('')\n for (const reason of gate.reasons) lines.push(`- ${reason}`)\n if (gate.delta !== undefined) lines.push(`- delta: ${gate.delta.toFixed(3)}`)\n lines.push('')\n lines.push('### Contributing gates')\n lines.push('')\n lines.push('| gate | passed | detail |')\n lines.push('|---|---|---|')\n for (const c of gate.contributingGates) {\n const detail =\n typeof c.detail === 'object'\n ? JSON.stringify(c.detail).slice(0, 80)\n : String(c.detail).slice(0, 80)\n lines.push(`| ${c.name} | ${c.passed ? '✓' : '✗'} | ${detail} |`)\n }\n lines.push('')\n lines.push('### Promoted surface')\n lines.push('')\n lines.push('```diff')\n lines.push(diff.slice(0, 8000))\n lines.push('```')\n lines.push('')\n lines.push('### By-judge aggregates')\n lines.push('')\n lines.push('| judge | mean | ci95 | n |')\n lines.push('|---|---|---|---|')\n for (const [name, agg] of Object.entries(result.aggregates.byJudge)) {\n lines.push(\n `| ${name} | ${agg.mean.toFixed(3)} | [${agg.ci95[0].toFixed(3)}, ${agg.ci95[1].toFixed(3)}] | ${agg.n} |`,\n )\n }\n return lines.join('\\n')\n}\n\nfunction defaultGhExec(args: string[]): { stdout: string; stderr: string; status: number } {\n try {\n const stdout = execSync(`gh ${args.map(quoteArg).join(' ')}`, {\n env: { ...process.env, GH_TOKEN: process.env.GH_AUTO_PR_TOKEN ?? process.env.GH_TOKEN ?? '' },\n stdio: ['ignore', 'pipe', 'pipe'],\n }).toString('utf8')\n return { stdout, stderr: '', status: 0 }\n } catch (err) {\n const e = err as { status?: number; stderr?: Buffer; stdout?: Buffer }\n return {\n stdout: e.stdout?.toString('utf8') ?? '',\n stderr: e.stderr?.toString('utf8') ?? '',\n status: e.status ?? 1,\n }\n }\n}\n\nfunction quoteArg(arg: string): string {\n if (/^[a-zA-Z0-9_/\\-:.@]+$/.test(arg)) return arg\n return `\"${arg.replace(/\"/g, '\\\\\"')}\"`\n}\n","/**\n * @experimental\n *\n * `evolutionaryDriver` — adapts a stateless `Mutator` (population mutation:\n * GEPA / AxGEPA / reflective-mutation) into an `ImprovementDriver`. This is\n * the evolutionary strategy: each generation, mutate the current best surface\n * into N candidates, measure, select. No generation memory beyond the current\n * surface; the loop body handles ranking + promotion.\n *\n * The reflective alternative is agent-runtime's `improvementDriver` with a\n * `reflectiveGenerator` / `agenticGenerator`: it reasons over the report +\n * trace findings to propose targeted edits rather than blind mutations. Both\n * conform to `ImprovementDriver`; the improvement loop is identical regardless\n * of which drives it.\n */\n\nimport type { ImprovementDriver, Mutator } from '../types'\n\nexport interface EvolutionaryDriverOptions<TFindings = unknown> {\n mutator: Mutator<TFindings>\n /** External findings fed to the mutator each generation. Default: []. */\n findings?: TFindings[]\n}\n\nexport function evolutionaryDriver<TFindings = unknown>(\n opts: EvolutionaryDriverOptions<TFindings>,\n): ImprovementDriver<TFindings> {\n return {\n kind: `evolutionary:${opts.mutator.kind}`,\n async propose({ currentSurface, findings, populationSize, signal }) {\n return opts.mutator.mutate({\n findings: findings.length > 0 ? findings : (opts.findings ?? []),\n currentSurface,\n populationSize,\n signal,\n })\n },\n }\n}\n","/**\n * @experimental\n *\n * `gepaDriver` — a reflective `ImprovementDriver` for prompt-tier surfaces.\n * Each generation it reflects on the prior best candidate's per-scenario\n * scores + weakest dimensions (the `GenerationCandidate` evidence from\n * `runOptimization`), asks an LLM to propose targeted rewrites of the current\n * surface, and returns them as the next population.\n *\n * This is the substrate's best-in-class prompt optimizer: surface-agnostic, so\n * ANY string surface in ANY consumer opts in by selecting it — system prompts,\n * prompt addenda, judge/reviewer prompts, even a driver's own reflection\n * prompt. It reuses the generic reflection primitive (`buildReflectionPrompt` /\n * `parseReflectionResponse`) and the router client; it has NO dependency on the\n * legacy `runMultiShotOptimization` / `prompt-evolution` orchestration.\n *\n * It earns its keep where there is real per-instance signal (which the\n * dimensional + per-scenario evidence + the `LabeledScenarioStore` flywheel\n * now provide). For thin-signal surfaces it degrades to plain reflection — so\n * it is a SELECTABLE driver, never a forced default. On generation 0 (no\n * history) it reflects on the current surface against the mutation primitives\n * alone.\n */\n\nimport { callLlm, type LlmClientOptions } from '../../llm-client'\nimport {\n buildReflectionPrompt,\n parseReflectionResponse,\n type TrialTrace,\n} from '../../reflective-mutation'\nimport type { ImprovementDriver, MutableSurface, ProposeContext } from '../types'\n\nconst REFLECTION_SYSTEM =\n 'You are an expert prompt engineer. Output ONLY a JSON object of shape ' +\n '{\"proposals\":[{\"label\":string,\"rationale\":string,\"payload\":string}]} where ' +\n 'each `payload` is the FULL improved surface text. No prose outside the JSON.'\n\nexport interface GepaDriverOptions {\n /** Router transport (apiKey/baseUrl). */\n llm: LlmClientOptions\n /** Model that performs the reflection. */\n model: string\n /** What is being optimized — appears in the reflection prompt for orientation. */\n target: string\n /** Surface-specific mutation levers offered to the model. */\n mutationPrimitives?: string[]\n /** Top/bottom scenarios surfaced as evidence each generation. Default 3. */\n evidenceK?: number\n /** Reflection sampling temperature. Default 0.7. */\n temperature?: number\n /** Reflection max tokens. Default 6000. */\n maxTokens?: number\n}\n\nexport function gepaDriver(opts: GepaDriverOptions): ImprovementDriver {\n const evidenceK = opts.evidenceK ?? 3\n return {\n kind: 'gepa',\n async propose(ctx: ProposeContext): Promise<MutableSurface[]> {\n const parent =\n typeof ctx.currentSurface === 'string'\n ? ctx.currentSurface\n : JSON.stringify(ctx.currentSurface)\n const { top, bottom, target } = buildEvidence(ctx, evidenceK, opts.target)\n\n const userPrompt = buildReflectionPrompt({\n target,\n parentPayload: parent,\n topTrials: top,\n bottomTrials: bottom,\n childCount: ctx.populationSize,\n mutationPrimitives: opts.mutationPrimitives,\n })\n\n const result = await callLlm(\n {\n model: opts.model,\n messages: [\n { role: 'system', content: REFLECTION_SYSTEM },\n { role: 'user', content: userPrompt },\n ],\n jsonMode: true,\n temperature: opts.temperature ?? 0.7,\n maxTokens: opts.maxTokens ?? 6000,\n },\n opts.llm,\n )\n\n const proposals = parseReflectionResponse(result.content, ctx.populationSize)\n const out: MutableSurface[] = []\n for (const proposal of proposals) {\n const text = typeof proposal.payload === 'string' ? proposal.payload.trim() : ''\n if (text && text !== parent && !out.includes(text)) out.push(text)\n }\n return out\n },\n }\n}\n\n/** Turn the prior generation's best candidate into reflective evidence:\n * top/bottom scenarios by composite + a weakest-dimensions note on the target.\n * Empty on generation 0 — the model reflects on the surface alone. */\nfunction buildEvidence(\n ctx: ProposeContext,\n evidenceK: number,\n baseTarget: string,\n): { top: TrialTrace[]; bottom: TrialTrace[]; target: string } {\n const last = ctx.history.at(-1)\n if (!last || last.candidates.length === 0) {\n return { top: [], bottom: [], target: baseTarget }\n }\n const best = [...last.candidates].sort((a, b) => b.composite - a.composite)[0]\n if (!best) return { top: [], bottom: [], target: baseTarget }\n\n const byScore = [...best.scenarios].sort((a, b) => b.composite - a.composite)\n const toTrace = (s: { scenarioId: string; composite: number }): TrialTrace => ({\n id: s.scenarioId,\n score: s.composite,\n })\n const top = byScore.slice(0, evidenceK).map(toTrace)\n const bottom = byScore.slice(-evidenceK).reverse().map(toTrace)\n\n const weakest = Object.entries(best.dimensions)\n .sort((a, b) => a[1] - b[1])\n .slice(0, 3)\n .map(([dim, value]) => `${dim} (${value.toFixed(2)})`)\n const target =\n weakest.length > 0 ? `${baseTarget} — weakest dimensions: ${weakest.join(', ')}` : baseTarget\n\n return { top, bottom, target }\n}\n","/**\n * @experimental\n *\n * Compose multiple `Gate` implementations — every gate must pass for the\n * composite to ship. Closes the alignment reviewer's \"default-only\n * heldOutGate + costGate would happily promote a reward-hacked prompt\"\n * concern by making safety gates first-class composable defaults.\n */\n\nimport type { Gate, GateContext, GateDecision, GateResult, Scenario } from '../types'\n\n/** Compose gates — all must `ship` for the composite to `ship`. First\n * non-ship verdict short-circuits the composite verdict, but ALL gates run\n * (so the result records every gate's reason — useful for diagnostics). */\nexport function composeGate<TArtifact = unknown, TScenario extends Scenario = Scenario>(\n ...gates: Array<Gate<TArtifact, TScenario>>\n): Gate<TArtifact, TScenario> {\n if (gates.length === 0) {\n throw new Error('composeGate requires at least one gate')\n }\n return {\n name: `composed(${gates.map((g) => g.name).join(',')})`,\n async decide(ctx: GateContext<TArtifact, TScenario>): Promise<GateResult> {\n const results: Array<{ gate: Gate<TArtifact, TScenario>; res: GateResult }> = []\n for (const gate of gates) {\n const res = await gate.decide(ctx)\n results.push({ gate, res })\n }\n\n // Substrate-wide verdict policy:\n // - all 'ship' → 'ship'\n // - any 'arch_ceiling' → 'arch_ceiling' (architectural ceiling beats other holds)\n // - any 'model_ceiling' → 'model_ceiling'\n // - any 'hold' → 'hold'\n // - else 'need_more_work'\n const decisions = results.map((r) => r.res.decision)\n const overall: GateDecision = decisions.every((d) => d === 'ship')\n ? 'ship'\n : decisions.includes('arch_ceiling')\n ? 'arch_ceiling'\n : decisions.includes('model_ceiling')\n ? 'model_ceiling'\n : decisions.includes('hold')\n ? 'hold'\n : 'need_more_work'\n\n const contributing = results.flatMap((r) =>\n r.res.contributingGates.length > 0\n ? r.res.contributingGates\n : [{ name: r.gate.name, passed: r.res.decision === 'ship', detail: r.res }],\n )\n\n const reasons = results.flatMap((r) =>\n r.res.reasons.map((reason) => `[${r.gate.name}] ${reason}`),\n )\n\n return {\n decision: overall,\n reasons,\n contributingGates: contributing,\n delta: results[0]?.res.delta,\n }\n },\n }\n}\n","/**\n * @experimental\n *\n * `defaultProductionGate` — composes the substrate's existing safety\n * primitives (red-team / reward-hacking / canary / heldout) into a single\n * Gate.decide shape. Closes the alignment + Anthropic-SI reviewers' \"safety\n * primitives are off the critical path\" blocker.\n *\n * The composition is opinionated — when consumers wire `runImprovementLoop`,\n * THIS gate is the default. Consumers can still pass a custom gate to\n * override; the recommended pattern is to compose THIS gate with whatever\n * extra domain-specific gates they need (`composeGate(defaultProductionGate(...), customGate)`).\n */\n\nimport type { RunRecord } from '@tangle-network/agent-runtime'\nimport type { CanaryReport } from '../../canary'\nimport { runCanaries } from '../../canary'\nimport type { RedTeamCase } from '../../red-team'\nimport { scoreRedTeamOutput } from '../../red-team'\nimport type { RewardHackingReport } from '../../rl/reward-hacking'\nimport { detectRewardHacking } from '../../rl/reward-hacking'\nimport type { Gate, GateContext, GateResult, Scenario } from '../types'\n\nexport interface DefaultProductionGateOptions {\n /** Required: scenarios held out from training; substrate compares\n * candidate-on-holdout vs baseline-on-holdout. */\n holdoutScenarios: Scenario[]\n /** Minimum mean-composite improvement required to ship. Default 0.5. */\n deltaThreshold?: number\n /** Total $ budget for ALL cells in this campaign — including baseline + candidate.\n * Composite verdict refuses to ship when spend exceeded budget. */\n budgetUsd?: number\n /** Red-team cases to probe candidate outputs against. When omitted the\n * substrate uses `DEFAULT_RED_TEAM_CORPUS`. Provide a domain-specific\n * battery for tighter coverage. */\n redTeamBattery?: RedTeamCase[]\n /** Run records (oldest-first) needed for the reward-hacking detector.\n * Substrate populates from prior production-loop generations. */\n recentRuns?: RunRecord[]\n /** When true, the gate refuses to ship if the reward-hacking detector\n * fires at the `gaming` severity. Default true. */\n blockOnRewardHackingGaming?: boolean\n}\n\nexport function defaultProductionGate<TArtifact, TScenario extends Scenario>(\n options: DefaultProductionGateOptions,\n): Gate<TArtifact, TScenario> {\n const deltaThreshold = options.deltaThreshold ?? 0.5\n const blockOnGaming = options.blockOnRewardHackingGaming ?? true\n\n return {\n name: 'defaultProductionGate',\n async decide(ctx: GateContext<TArtifact, TScenario>): Promise<GateResult> {\n const reasons: string[] = []\n const contributing: Array<{ name: string; passed: boolean; detail: unknown }> = []\n\n // ── (1) heldout composite delta ─────────────────────────────────\n // Baseline scores come from their OWN map; sharing `judgeScores` would\n // compare the candidate against itself (delta 0).\n const baselineComposite = meanComposite(\n ctx.baselineArtifacts,\n ctx.baselineJudgeScores ?? ctx.judgeScores,\n options.holdoutScenarios,\n )\n const candidateComposite = meanComposite(\n ctx.candidateArtifacts,\n ctx.judgeScores,\n options.holdoutScenarios,\n )\n const delta = candidateComposite - baselineComposite\n const heldoutPass = delta >= deltaThreshold\n contributing.push({\n name: 'heldout-delta',\n passed: heldoutPass,\n detail: { baselineComposite, candidateComposite, delta, deltaThreshold },\n })\n if (!heldoutPass) {\n reasons.push(`heldout delta ${delta.toFixed(3)} < threshold ${deltaThreshold}`)\n }\n\n // ── (2) budget gate ─────────────────────────────────────────────\n const budgetPass =\n options.budgetUsd === undefined ||\n ctx.cost.candidate + ctx.cost.baseline <= options.budgetUsd\n contributing.push({\n name: 'budget',\n passed: budgetPass,\n detail: {\n candidateUsd: ctx.cost.candidate,\n baselineUsd: ctx.cost.baseline,\n budgetUsd: options.budgetUsd,\n },\n })\n if (!budgetPass) {\n reasons.push(\n `spend ${(ctx.cost.candidate + ctx.cost.baseline).toFixed(2)} > budget ${options.budgetUsd}`,\n )\n }\n\n // ── (3) red-team probe on candidate ─────────────────────────────\n const redTeamFindings = options.redTeamBattery\n ? probeRedTeam(ctx.candidateArtifacts, options.redTeamBattery)\n : { passed: true, findings: [] }\n contributing.push({\n name: 'red-team',\n passed: redTeamFindings.passed,\n detail: {\n failures: redTeamFindings.findings.length,\n sample: redTeamFindings.findings.slice(0, 3),\n },\n })\n if (!redTeamFindings.passed) {\n reasons.push(`red-team probe failed (${redTeamFindings.findings.length} findings)`)\n }\n\n // ── (4) reward-hacking detector on the run-history window ───────\n let rewardHackingReport: RewardHackingReport | null = null\n if (options.recentRuns && options.recentRuns.length >= 10) {\n rewardHackingReport = detectRewardHacking({ runs: options.recentRuns })\n }\n // reward-hacking severity is numeric (0..1). \"gaming\" threshold per\n // detectRewardHacking defaults = 0.6. Block when ANY finding is at\n // gaming threshold OR the report verdict is 'gaming'.\n const gamingThreshold = 0.6\n const gamingFindings = (rewardHackingReport?.findings ?? []).filter(\n (f) => f.severity >= gamingThreshold,\n )\n const rewardHackingPass =\n !rewardHackingReport ||\n !blockOnGaming ||\n (gamingFindings.length === 0 && rewardHackingReport.verdict !== 'gaming')\n contributing.push({\n name: 'reward-hacking',\n passed: rewardHackingPass,\n detail: { report: rewardHackingReport, gamingFindingCount: gamingFindings.length },\n })\n if (!rewardHackingPass) {\n reasons.push(\n `reward-hacking detector flagged ${gamingFindings.length} gaming-severity findings (verdict=${rewardHackingReport!.verdict})`,\n )\n }\n\n // ── (5) canary check on runs ────────────────────────────────────\n let canaryReport: CanaryReport | null = null\n if (options.recentRuns && options.recentRuns.length >= 10) {\n canaryReport = runCanaries(options.recentRuns, {})\n }\n // CanarySeverity is 'info' | 'warn' | 'error' — block on 'error'.\n const errorAlerts = (canaryReport?.alerts ?? []).filter((a) => a.severity === 'error')\n const canaryPass = errorAlerts.length === 0\n contributing.push({\n name: 'canary',\n passed: canaryPass,\n detail: { totalAlerts: canaryReport?.alerts.length ?? 0, errorAlerts: errorAlerts.length },\n })\n if (!canaryPass) {\n reasons.push(`canary error alerts: ${errorAlerts.length}`)\n }\n\n // ── Verdict ─────────────────────────────────────────────────────\n const allPassed = contributing.every((c) => c.passed)\n const decision = allPassed ? 'ship' : 'hold'\n\n return {\n decision,\n reasons: reasons.length > 0 ? reasons : ['all gates passed'],\n contributingGates: contributing,\n delta,\n }\n },\n }\n}\n\nfunction meanComposite<TArtifact, TScenario extends Scenario>(\n artifacts: Map<string, TArtifact> | undefined,\n judgeScoresByCell: Map<string, Record<string, { composite: number }>>,\n scenarios: TScenario[],\n): number {\n if (!artifacts || artifacts.size === 0) return 0\n const scenarioIds = new Set(scenarios.map((s) => s.id))\n const composites: number[] = []\n for (const [cellId, scores] of judgeScoresByCell) {\n const scenarioId = cellId.split(':')[0] ?? ''\n if (!scenarioIds.has(scenarioId)) continue\n const cellComposites = Object.values(scores).map((s) => s.composite)\n if (cellComposites.length === 0) continue\n composites.push(cellComposites.reduce((a, b) => a + b, 0) / cellComposites.length)\n }\n if (composites.length === 0) return 0\n return composites.reduce((a, b) => a + b, 0) / composites.length\n}\n\nfunction probeRedTeam<TArtifact>(\n artifacts: Map<string, TArtifact>,\n battery: RedTeamCase[],\n): { passed: boolean; findings: Array<{ scenarioId: string; reason: string }> } {\n const findings: Array<{ scenarioId: string; reason: string }> = []\n for (const [_cellId, artifact] of artifacts) {\n const text = extractText(artifact)\n if (text === undefined) continue\n for (const rtCase of battery) {\n const finding = scoreRedTeamOutput(text, [], rtCase)\n if (!finding.passed) {\n findings.push({ scenarioId: rtCase.id, reason: finding.reason ?? 'red-team probe failed' })\n }\n }\n }\n return { passed: findings.length === 0, findings }\n}\n\nfunction extractText(artifact: unknown): string | undefined {\n if (typeof artifact === 'string') return artifact\n if (artifact && typeof artifact === 'object') {\n const rec = artifact as Record<string, unknown>\n if (typeof rec.text === 'string') return rec.text\n if (typeof rec.output === 'string') return rec.output\n if (typeof rec.content === 'string') return rec.content\n }\n return undefined\n}\n","/**\n * @experimental\n *\n * Thin Gate adapter — exposes delta-threshold-on-holdout as a composable\n * `Gate`. Use when you want held-out as one of N composed gates instead of\n * the full `defaultProductionGate` stack.\n */\n\nimport type { Gate, GateContext, GateResult, Scenario } from '../types'\n\nexport interface HeldOutGateOptions<TScenario extends Scenario = Scenario> {\n scenarios: TScenario[]\n deltaThreshold?: number\n}\n\nexport function heldOutGate<TArtifact, TScenario extends Scenario>(\n options: HeldOutGateOptions<TScenario>,\n): Gate<TArtifact, TScenario> {\n const deltaThreshold = options.deltaThreshold ?? 0.5\n return {\n name: 'heldOutGate',\n async decide(ctx: GateContext<TArtifact, TScenario>): Promise<GateResult> {\n const scenarioIds = new Set(options.scenarios.map((s) => s.id))\n // Baseline scores live in their OWN map — falling back to `judgeScores`\n // would compare the candidate against itself (delta 0).\n const baseline = meanForScenarios(ctx.baselineJudgeScores ?? ctx.judgeScores, scenarioIds)\n const candidate = meanForScenarios(ctx.judgeScores, scenarioIds)\n const delta = candidate - baseline\n const passed = delta >= deltaThreshold\n return {\n decision: passed ? 'ship' : 'hold',\n reasons: passed\n ? [`held-out delta ${delta.toFixed(3)} ≥ ${deltaThreshold}`]\n : [`held-out delta ${delta.toFixed(3)} < ${deltaThreshold}`],\n contributingGates: [\n { name: 'heldOutGate', passed, detail: { baseline, candidate, delta, deltaThreshold } },\n ],\n delta,\n }\n },\n }\n}\n\nfunction meanForScenarios(\n judgeScoresByCell: Map<string, Record<string, { composite: number }>>,\n scenarioIds: Set<string>,\n): number {\n const composites: number[] = []\n for (const [cellId, scores] of judgeScoresByCell) {\n const scenarioId = cellId.split(':')[0] ?? ''\n if (!scenarioIds.has(scenarioId)) continue\n const vals = Object.values(scores).map((s) => s.composite)\n if (vals.length > 0) composites.push(vals.reduce((a, b) => a + b, 0) / vals.length)\n }\n return composites.length === 0 ? 0 : composites.reduce((a, b) => a + b, 0) / composites.length\n}\n","/**\n * @experimental\n *\n * Filesystem `LabeledScenarioStore` adapter. The default capture sink for\n * traces + eval artifacts. Production deployments typically swap for a\n * Turso/SQLite adapter (same interface).\n *\n * Records land as one JSONL file per source under `<root>/<source>.jsonl`.\n * Each line is a `LabeledScenarioRecord`. Append-only — no in-place edits.\n *\n * Safety properties enforced at write-time:\n *\n * - **Provenance required**: writes without `source`, `sourceVersionHash`,\n * `capturedAt`, `redactionStatus` are rejected. Closes the alignment\n * reviewer's data-poisoning gap.\n * - **Per-source rate limits**: optional `rateLimitBucket` + `maxWritesPerMinute`\n * stops a single tenant/source from flooding the store.\n *\n * Safety properties enforced at sample-time:\n *\n * - **Required split + capturedBefore**: substrate refuses to sample without\n * an explicit `split` ('train' | 'test') AND a temporal cutoff. Eliminates\n * accidental train/test contamination.\n * - **Default training-source filter**: when the store is sampled with\n * `split: 'train'`, production-trace records are EXCLUDED unless the\n * caller passes `filter.source: 'production-trace'` explicitly. Closes\n * the contamination-by-default gap flagged by the senior eval engineer.\n */\n\nimport { createHash } from 'node:crypto'\nimport { existsSync, mkdirSync, readFileSync, writeFileSync } from 'node:fs'\nimport { join } from 'node:path'\nimport type {\n LabeledScenarioRecord,\n LabeledScenarioSampleArgs,\n LabeledScenarioSource,\n LabeledScenarioStore,\n LabeledScenarioWrite,\n} from '../types'\n\nexport interface FsLabeledScenarioStoreOptions {\n /** Root directory for JSONL files. Created if missing. */\n root: string\n /** Per-source rate limit. When set, writes exceeding the cap are rejected\n * with a typed error. Default: no limit. */\n maxWritesPerMinutePerBucket?: number\n /** Test seam — override `Date.now()` for deterministic tests. */\n now?: () => number\n}\n\nexport class LabeledScenarioStoreError extends Error {\n constructor(\n public readonly code: string,\n message: string,\n ) {\n super(message)\n this.name = 'LabeledScenarioStoreError'\n }\n}\n\ninterface RateLimitState {\n bucket: string\n windowStartMs: number\n count: number\n}\n\nexport class FsLabeledScenarioStore implements LabeledScenarioStore {\n private readonly now: () => number\n private readonly rateLimits = new Map<string, RateLimitState>()\n\n constructor(private readonly options: FsLabeledScenarioStoreOptions) {\n if (!existsSync(options.root)) mkdirSync(options.root, { recursive: true })\n this.now = options.now ?? Date.now\n }\n\n async observe(write: LabeledScenarioWrite): Promise<void> {\n this.assertProvenance(write)\n this.assertRateLimit(write)\n const record = this.toRecord(write)\n const path = this.pathForSource(write.source)\n const line = `${JSON.stringify(record)}\\n`\n // Append atomically. For high-throughput a writev-friendly buffered\n // implementation lands in the Turso adapter; FS adapter is for tests +\n // local dev + small workloads.\n appendLine(path, line)\n }\n\n async sample(args: LabeledScenarioSampleArgs): Promise<LabeledScenarioRecord[]> {\n if (!args.split) {\n throw new LabeledScenarioStoreError(\n 'split_required',\n 'sample() requires an explicit `split` (train | test) — substrate refuses ambiguous reads',\n )\n }\n if (!args.capturedBefore) {\n throw new LabeledScenarioStoreError(\n 'capturedBefore_required',\n 'sample() requires an explicit `capturedBefore` timestamp for temporal-split discipline',\n )\n }\n\n const all: LabeledScenarioRecord[] = []\n for (const source of ALL_SOURCES) {\n // Default training-source filter: when sampling train, EXCLUDE\n // production-trace records unless the caller asks for them.\n if (args.split === 'train' && source === 'production-trace') {\n const explicit = sourceFilterContains(args.filter?.source, 'production-trace')\n if (!explicit) continue\n }\n const path = this.pathForSource(source)\n if (!existsSync(path)) continue\n const lines = readFileSync(path, 'utf8').split('\\n').filter(Boolean)\n for (const line of lines) {\n let record: LabeledScenarioRecord\n try {\n record = JSON.parse(line) as LabeledScenarioRecord\n } catch {\n continue\n }\n if (!matchesFilter(record, args, source)) continue\n all.push(record)\n }\n }\n\n // Deterministic order: by capturedAt ascending, then recordHash.\n all.sort((a, b) => {\n if (a.capturedAt !== b.capturedAt) return a.capturedAt.localeCompare(b.capturedAt)\n return a.recordHash.localeCompare(b.recordHash)\n })\n\n return all.slice(0, args.count)\n }\n\n async size(): Promise<{ train: number; test: number; bySource: Record<string, number> }> {\n const bySource: Record<string, number> = {}\n let total = 0\n for (const source of ALL_SOURCES) {\n const path = this.pathForSource(source)\n if (!existsSync(path)) {\n bySource[source] = 0\n continue\n }\n const count = readFileSync(path, 'utf8').split('\\n').filter(Boolean).length\n bySource[source] = count\n total += count\n }\n // FS adapter doesn't track split assignments per-record (split is\n // computed at sample-time based on `capturedBefore`). For size(), we\n // report `train`+`test` as the same total — split is a sampling concept.\n return { train: total, test: total, bySource }\n }\n\n private assertProvenance(write: LabeledScenarioWrite): void {\n if (!write.source) {\n throw new LabeledScenarioStoreError(\n 'missing_source',\n 'LabeledScenarioWrite requires `source`',\n )\n }\n if (!write.sourceVersionHash || write.sourceVersionHash.length === 0) {\n throw new LabeledScenarioStoreError(\n 'missing_source_version',\n 'LabeledScenarioWrite requires `sourceVersionHash` (git sha or substrate version)',\n )\n }\n if (!write.capturedAt) {\n throw new LabeledScenarioStoreError(\n 'missing_captured_at',\n 'LabeledScenarioWrite requires `capturedAt` ISO timestamp',\n )\n }\n if (!write.redactionStatus) {\n throw new LabeledScenarioStoreError(\n 'missing_redaction_status',\n 'LabeledScenarioWrite requires explicit `redactionStatus` — raw / redacted-pii / redacted-secrets / fully-redacted',\n )\n }\n if (!ALL_SOURCES.includes(write.source)) {\n throw new LabeledScenarioStoreError(\n 'unknown_source',\n `LabeledScenarioWrite.source must be one of: ${ALL_SOURCES.join(', ')}`,\n )\n }\n }\n\n private assertRateLimit(write: LabeledScenarioWrite): void {\n const cap = this.options.maxWritesPerMinutePerBucket\n if (!cap || !write.rateLimitBucket) return\n const now = this.now()\n const windowMs = 60_000\n let state = this.rateLimits.get(write.rateLimitBucket)\n if (!state || now - state.windowStartMs >= windowMs) {\n state = { bucket: write.rateLimitBucket, windowStartMs: now, count: 0 }\n this.rateLimits.set(write.rateLimitBucket, state)\n }\n if (state.count >= cap) {\n throw new LabeledScenarioStoreError(\n 'rate_limit_exceeded',\n `LabeledScenarioStore: bucket ${write.rateLimitBucket} exceeded ${cap} writes/min`,\n )\n }\n state.count += 1\n }\n\n private toRecord(write: LabeledScenarioWrite): LabeledScenarioRecord {\n const recordHash = sha256(\n JSON.stringify({\n id: write.scenario.id,\n src: write.source,\n at: write.capturedAt,\n ver: write.sourceVersionHash,\n }),\n )\n // FS adapter assigns split at sample-time, but we cache a hint here\n // based on capturedAt vs the world's \"now\" — sampler overrides this.\n return {\n ...write,\n recordHash,\n split: 'train',\n }\n }\n\n private pathForSource(source: string): string {\n return join(this.options.root, `${source}.jsonl`)\n }\n}\n\nconst ALL_SOURCES: LabeledScenarioWrite['source'][] = [\n 'production-trace',\n 'eval-run',\n 'manual',\n 'red-team',\n 'synthetic',\n]\n\nfunction sourceFilterContains(\n filter: LabeledScenarioSource | LabeledScenarioSource[] | undefined,\n needle: LabeledScenarioSource,\n): boolean {\n if (!filter) return false\n if (Array.isArray(filter)) return filter.includes(needle)\n return filter === needle\n}\n\nfunction matchesFilter(\n record: LabeledScenarioRecord,\n args: LabeledScenarioSampleArgs,\n source: string,\n): boolean {\n // Temporal cutoff — train must be capturedAt < capturedBefore.\n if (args.split === 'train' && record.capturedAt >= args.capturedBefore) return false\n if (args.split === 'test' && record.capturedAt < args.capturedBefore) return false\n\n const f = args.filter\n if (!f) return true\n if (f.kind && record.scenario.kind !== f.kind) return false\n if (f.source) {\n const sources = Array.isArray(f.source) ? f.source : [f.source]\n if (!sources.includes(source as never)) return false\n }\n if (f.minComposite !== undefined || f.maxComposite !== undefined) {\n const composites = Object.values(record.judgeScores).map((s) => s.composite)\n const max = composites.length === 0 ? 0 : Math.max(...composites)\n if (f.minComposite !== undefined && max < f.minComposite) return false\n if (f.maxComposite !== undefined && max > f.maxComposite) return false\n }\n return true\n}\n\nfunction sha256(input: string): string {\n return createHash('sha256').update(input).digest('hex').slice(0, 16)\n}\n\nfunction appendLine(path: string, line: string): void {\n if (existsSync(path)) {\n const existing = readFileSync(path, 'utf8')\n writeFileSync(path, existing + line)\n } else {\n writeFileSync(path, line)\n }\n}\n","/**\n * @experimental\n *\n * `runEval` — the simplest preset over `runCampaign`. No optimizer, no\n * gate, no auto-PR. Just: run scenarios through dispatch, score with\n * judges, return CampaignResult.\n *\n * The 80% case for consumers who want a scorecard, not an improvement loop.\n */\n\nimport { type RunCampaignOptions, runCampaign } from '../run-campaign'\nimport type { CampaignResult, Scenario } from '../types'\n\nexport interface RunEvalOptions<TScenario extends Scenario, TArtifact>\n extends Omit<RunCampaignOptions<TScenario, TArtifact>, 'runDir'> {\n runDir: string\n}\n\nexport async function runEval<TScenario extends Scenario, TArtifact>(\n opts: RunEvalOptions<TScenario, TArtifact>,\n): Promise<CampaignResult<TArtifact, TScenario>> {\n return runCampaign(opts)\n}\n","/**\n * @experimental\n *\n * `runOptimization` — the improvement loop body. Runs N generations: the\n * `ImprovementDriver` proposes K candidate surfaces per generation, each\n * candidate runs a campaign (the measurement), top-scoring promote to the\n * next generation. Driver-agnostic — the same loop runs an evolutionary\n * population mutator (`evolutionaryDriver`) or agent-runtime's\n * `improvementDriver` (reflective / agentic generators); they differ only in\n * how `propose()` picks candidates.\n *\n * This is `runLoop`'s shape (plan → measure → decide) specialized to surface\n * improvement: `driver.propose` = plan, `runCampaign` = the measurement (which\n * runs the worker behind `dispatch`), the mean-composite ranking = the\n * validator, `driver.decide` = the stop check.\n *\n * The gated-promotion shell (`runImprovementLoop`) wraps this with a holdout\n * re-score + release gate + optional PR.\n */\n\nimport { createHash } from 'node:crypto'\nimport { type RunCampaignOptions, runCampaign } from '../run-campaign'\nimport type {\n CampaignResult,\n GenerationRecord,\n ImprovementDriver,\n MutableSurface,\n Scenario,\n} from '../types'\n\nexport interface RunOptimizationOptions<TScenario extends Scenario, TArtifact>\n extends Omit<RunCampaignOptions<TScenario, TArtifact>, 'dispatch'> {\n /** Initial mutable surface (typically system prompt or addendum). */\n baselineSurface: MutableSurface\n /** Dispatcher that takes the CURRENT surface + scenario → artifact. */\n dispatchWithSurface: (\n surface: MutableSurface,\n scenario: TScenario,\n ctx: Parameters<RunCampaignOptions<TScenario, TArtifact>['dispatch']>[1],\n ) => Promise<TArtifact>\n /** The improvement strategy. Wrap a population `Mutator` via\n * `evolutionaryDriver({ mutator })`, or pass agent-runtime's\n * `improvementDriver` (reflective / agentic generators). */\n driver: ImprovementDriver\n populationSize: number\n maxGenerations: number\n /** How many top-scoring candidates carry to the next generation. Default 2. */\n promoteTopK?: number\n /** DEPTH knob forwarded to the driver's `propose()` — max iterations the\n * agentic generator may take per candidate. */\n maxImprovementShots?: number\n /** Phase-2 research report forwarded to `propose()` (analyst findings +\n * diff). Opaque here; the driver types it. */\n report?: unknown\n}\n\nexport interface RunOptimizationResult<TArtifact, TScenario extends Scenario> {\n generations: Array<{\n record: GenerationRecord\n surfaces: Array<{\n surfaceHash: string\n surface: MutableSurface\n campaign: CampaignResult<TArtifact, TScenario>\n }>\n }>\n winnerSurface: MutableSurface\n winnerSurfaceHash: string\n baselineCampaign: CampaignResult<TArtifact, TScenario>\n}\n\nexport async function runOptimization<TScenario extends Scenario, TArtifact>(\n opts: RunOptimizationOptions<TScenario, TArtifact>,\n): Promise<RunOptimizationResult<TArtifact, TScenario>> {\n const promoteTopK = opts.promoteTopK ?? 2\n\n // Baseline run\n const baselineCampaign = await runCampaign<TScenario, TArtifact>({\n ...opts,\n dispatch: (scenario, ctx) => opts.dispatchWithSurface(opts.baselineSurface, scenario, ctx),\n runDir: `${opts.runDir}/baseline`,\n })\n\n const generations: RunOptimizationResult<TArtifact, TScenario>['generations'] = []\n const history: GenerationRecord[] = []\n let currentSurfaces: MutableSurface[] = [opts.baselineSurface]\n let winnerSurface = opts.baselineSurface\n let winnerSurfaceHash = surfaceHash(opts.baselineSurface)\n let winnerComposite = meanComposite(baselineCampaign)\n\n for (let gen = 0; gen < opts.maxGenerations; gen++) {\n // Decide: the driver may stop early based on accumulated history.\n if (opts.driver.decide?.({ history }).stop) break\n\n // Plan: the driver proposes N candidates from the current best surface,\n // the accumulated generation history, and any external findings.\n const candidates = await opts.driver.propose({\n currentSurface: currentSurfaces[0] ?? opts.baselineSurface,\n history,\n findings: [],\n populationSize: opts.populationSize,\n generation: gen,\n signal: new AbortController().signal,\n report: opts.report,\n dataset: opts.labeledStore && opts.labeledStore !== 'off' ? opts.labeledStore : undefined,\n maxImprovementShots: opts.maxImprovementShots,\n })\n\n // Run each candidate as its own campaign.\n const surfaceResults: Array<{\n surfaceHash: string\n surface: MutableSurface\n campaign: CampaignResult<TArtifact, TScenario>\n composite: number\n }> = []\n for (let i = 0; i < candidates.length; i++) {\n const surface = candidates[i] as MutableSurface\n const hash = surfaceHash(surface)\n const campaign = await runCampaign<TScenario, TArtifact>({\n ...opts,\n dispatch: (scenario, ctx) => opts.dispatchWithSurface(surface, scenario, ctx),\n runDir: `${opts.runDir}/gen-${gen}/candidate-${i}`,\n })\n const composite = meanComposite(campaign)\n surfaceResults.push({ surfaceHash: hash, surface, campaign, composite })\n }\n\n // Rank, promote top-K.\n surfaceResults.sort((a, b) => b.composite - a.composite)\n const promoted = surfaceResults.slice(0, promoteTopK)\n currentSurfaces = promoted.map((p) => p.surface)\n const top = surfaceResults[0]\n if (top && top.composite > winnerComposite) {\n winnerSurface = top.surface\n winnerSurfaceHash = top.surfaceHash\n winnerComposite = top.composite\n }\n\n const record: GenerationRecord = {\n generationIndex: gen,\n candidates: surfaceResults.map((s) => {\n const breakdown = candidateBreakdown(s.campaign)\n return {\n surfaceHash: s.surfaceHash,\n composite: s.composite,\n ci95: [s.composite, s.composite] as [number, number],\n dimensions: breakdown.dimensions,\n scenarios: breakdown.scenarios,\n }\n }),\n promoted: promoted.map((p) => p.surfaceHash),\n }\n history.push(record)\n generations.push({\n record,\n surfaces: surfaceResults.map((s) => ({\n surfaceHash: s.surfaceHash,\n surface: s.surface,\n campaign: s.campaign,\n })),\n })\n }\n\n return {\n generations,\n winnerSurface,\n winnerSurfaceHash,\n baselineCampaign,\n }\n}\n\nexport function surfaceHash(surface: MutableSurface): string {\n // Prompt/tool surfaces (string) hash by content; code surfaces hash by the\n // worktree + base ref pair (the content lives in git, not in the string).\n const material =\n typeof surface === 'string'\n ? surface\n : JSON.stringify({\n kind: surface.kind,\n worktreeRef: surface.worktreeRef,\n baseRef: surface.baseRef ?? null,\n })\n return createHash('sha256').update(material).digest('hex').slice(0, 16)\n}\n\nfunction meanComposite<TArtifact, TScenario extends Scenario>(\n campaign: CampaignResult<TArtifact, TScenario>,\n): number {\n const composites: number[] = []\n for (const cell of campaign.cells) {\n const cellComposites = Object.values(cell.judgeScores).map((s) => s.composite)\n if (cellComposites.length > 0) {\n composites.push(cellComposites.reduce((a, b) => a + b, 0) / cellComposites.length)\n }\n }\n return composites.length === 0 ? 0 : composites.reduce((a, b) => a + b, 0) / composites.length\n}\n\n/** Per-candidate evidence a reflective driver grounds its next proposal on:\n * mean score per judge dimension + per-scenario composite. */\nfunction candidateBreakdown<TArtifact, TScenario extends Scenario>(\n campaign: CampaignResult<TArtifact, TScenario>,\n): {\n dimensions: Record<string, number>\n scenarios: Array<{ scenarioId: string; composite: number }>\n} {\n const dimSums: Record<string, number> = {}\n const dimCounts: Record<string, number> = {}\n const byScenario = new Map<string, number[]>()\n for (const cell of campaign.cells) {\n const judgeScores = Object.values(cell.judgeScores)\n if (judgeScores.length === 0) continue\n const cellComposite = judgeScores.reduce((a, s) => a + s.composite, 0) / judgeScores.length\n const arr = byScenario.get(cell.scenarioId) ?? []\n arr.push(cellComposite)\n byScenario.set(cell.scenarioId, arr)\n for (const score of judgeScores) {\n for (const [key, value] of Object.entries(score.dimensions)) {\n dimSums[key] = (dimSums[key] ?? 0) + value\n dimCounts[key] = (dimCounts[key] ?? 0) + 1\n }\n }\n }\n const dimensions: Record<string, number> = {}\n for (const key of Object.keys(dimSums)) {\n const count = dimCounts[key] ?? 0\n dimensions[key] = count > 0 ? (dimSums[key] ?? 0) / count : 0\n }\n const scenarios = [...byScenario.entries()].map(([scenarioId, comps]) => ({\n scenarioId,\n composite: comps.reduce((a, b) => a + b, 0) / comps.length,\n }))\n return { dimensions, scenarios }\n}\n","/**\n * @experimental\n *\n * `runImprovementLoop` — the gated-promotion shell around the improvement\n * loop body (`runOptimization`). Drives candidate surfaces via the\n * `ImprovementDriver`, re-scores the winner against the baseline on a\n * holdout set, runs the release gate, and optionally opens a PR.\n *\n * Role vocabulary (see docs/design/loop-taxonomy.md):\n * - DRIVER = the `ImprovementDriver` (evolutionary GEPA mutator OR\n * reflective analyst). Proposes candidate SURFACES — the\n * worker's system prompt / tool config — NOT conversation\n * turns.\n * - MEASUREMENT= `runCampaign`. Scores one surface by running the worker\n * (via `dispatch`) over scenarios and judging the output.\n * - WORKER = the agent harness in the sandbox, invoked behind the\n * topology-opaque `dispatch` seam — never referenced here.\n *\n * Distinct from `runLoop` in `@tangle-network/agent-runtime`, which is the\n * INNER conversation loop (driver↔workers in a sandbox). `runImprovementLoop`\n * is the OUTER loop: it improves the surface that those workers run.\n *\n * Hard-refuses unsafe configurations:\n * - `tracing: 'off'` when a driver is wired (improvement is unattributable)\n * - `autoOnPromote: 'config'` — DEFERRED to Pass B; v0.40 only ships\n * `'pr'` and `'none'`.\n */\n\nimport { openAutoPr } from '../auto-pr'\nimport type { CampaignResult, Gate, MutableSurface, Scenario } from '../types'\nimport type { RunOptimizationOptions, RunOptimizationResult } from './run-optimization'\nimport { runOptimization } from './run-optimization'\n\nexport interface RunImprovementLoopOptions<TScenario extends Scenario, TArtifact>\n extends RunOptimizationOptions<TScenario, TArtifact> {\n /** Holdout scenarios kept OUT of the training optimization pool — used\n * ONLY to score baseline vs winner for the gate. */\n holdoutScenarios: TScenario[]\n /** Promotion gate. Substrate strongly recommends `defaultProductionGate`\n * for production wiring (composes red-team / reward-hacking / canary /\n * heldout). */\n gate: Gate<TArtifact, TScenario>\n /** What to do when the gate ships:\n * - `'pr'`: open a PR via `openAutoPr`\n * - `'none'`: just report — caller decides what to do with the winner\n * v0.40 does NOT support `'config'` (live-runtime self-mutation) —\n * deferred to Pass B behind safety stack. */\n autoOnPromote: 'pr' | 'none'\n /** GH owner / repo for the auto-PR. Required when autoOnPromote === 'pr'. */\n ghOwner?: string\n ghRepo?: string\n /** Optional render override — substrate writes a diff-shaped surface; pass\n * a function to format the promoted surface differently. */\n renderPromotedDiff?: (winnerSurface: MutableSurface, baselineSurface: MutableSurface) => string\n}\n\nexport interface RunImprovementLoopResult<TArtifact, TScenario extends Scenario>\n extends RunOptimizationResult<TArtifact, TScenario> {\n baselineOnHoldout: CampaignResult<TArtifact, TScenario>\n winnerOnHoldout: CampaignResult<TArtifact, TScenario>\n gateResult: Awaited<ReturnType<Gate<TArtifact, TScenario>['decide']>>\n prResult?: ReturnType<typeof openAutoPr>\n}\n\nexport async function runImprovementLoop<TScenario extends Scenario, TArtifact>(\n opts: RunImprovementLoopOptions<TScenario, TArtifact>,\n): Promise<RunImprovementLoopResult<TArtifact, TScenario>> {\n // ── Safety pre-flight ─────────────────────────────────────────────\n // biome-ignore lint/suspicious/noExplicitAny: Pass A reserved field for Pass B Shape B\n if ((opts as any).autoOnPromote === 'config') {\n throw new Error(\n \"runImprovementLoop: autoOnPromote='config' is deferred to Pass B (requires shadow deploy + rollback + ensemble judges). Use 'pr' or 'none' in v0.40.\",\n )\n }\n // Refuse tracing=off whenever a driver is wired. An improvement loop\n // without traces is unattributable — its candidate surfaces cannot be\n // cited back to the spans that motivated them, and the dataset flywheel\n // (LabeledScenarioStore) that GEPA optimizes against goes unfed.\n if (opts.tracing === 'off' && opts.driver) {\n throw new Error(\n \"runImprovementLoop: tracing='off' is forbidden when a driver is wired. The improvement loop without traces is unattributable; candidate surfaces cannot be cited back to spans and the optimization dataset goes unfed.\",\n )\n }\n if (opts.autoOnPromote === 'pr' && (!opts.ghOwner || !opts.ghRepo)) {\n throw new Error(\"runImprovementLoop: autoOnPromote='pr' requires ghOwner + ghRepo.\")\n }\n\n // ── (1) optimization loop produces a winner ────────────────────────\n const optimization = await runOptimization(opts)\n\n // ── (2) baseline + winner re-scored on the holdout set ─────────────\n const { runCampaign } = await import('../run-campaign')\n\n const baselineOnHoldout = await runCampaign<TScenario, TArtifact>({\n ...opts,\n scenarios: opts.holdoutScenarios,\n dispatch: (scenario, ctx) => opts.dispatchWithSurface(opts.baselineSurface, scenario, ctx),\n runDir: `${opts.runDir}/holdout-baseline`,\n })\n\n const winnerOnHoldout = await runCampaign<TScenario, TArtifact>({\n ...opts,\n scenarios: opts.holdoutScenarios,\n dispatch: (scenario, ctx) =>\n opts.dispatchWithSurface(optimization.winnerSurface, scenario, ctx),\n runDir: `${opts.runDir}/holdout-winner`,\n })\n\n // ── (3) gate verdict ───────────────────────────────────────────────\n // Candidate + baseline share cellIds (same holdout scenarios), so their\n // judge scores MUST stay in separate maps — merging them collapses the\n // holdout delta to zero and the gate can never ship a real improvement.\n type ScoreMap = Map<\n string,\n Record<string, { composite: number; dimensions: Record<string, number>; notes: string }>\n >\n const candidateArtifacts = new Map<string, TArtifact>()\n const baselineArtifacts = new Map<string, TArtifact>()\n const judgeScores: ScoreMap = new Map()\n const baselineJudgeScores: ScoreMap = new Map()\n for (const cell of winnerOnHoldout.cells) {\n candidateArtifacts.set(cell.cellId, cell.artifact)\n judgeScores.set(cell.cellId, cell.judgeScores)\n }\n for (const cell of baselineOnHoldout.cells) {\n baselineArtifacts.set(cell.cellId, cell.artifact)\n baselineJudgeScores.set(cell.cellId, cell.judgeScores)\n }\n\n const gateResult = await opts.gate.decide({\n candidateArtifacts,\n baselineArtifacts,\n judgeScores,\n baselineJudgeScores,\n scenarios: opts.holdoutScenarios,\n cost: {\n candidate: winnerOnHoldout.aggregates.totalCostUsd,\n baseline: baselineOnHoldout.aggregates.totalCostUsd,\n },\n signal: new AbortController().signal,\n })\n\n // ── (4) auto-PR when gate ships ────────────────────────────────────\n let prResult: ReturnType<typeof openAutoPr> | undefined\n if (opts.autoOnPromote === 'pr' && gateResult.decision === 'ship') {\n const render = opts.renderPromotedDiff ?? defaultRenderDiff\n const promotedDiff = render(optimization.winnerSurface, opts.baselineSurface)\n prResult = openAutoPr({\n result: winnerOnHoldout,\n gate: gateResult,\n promotedDiff,\n ghOwner: opts.ghOwner!,\n ghRepo: opts.ghRepo!,\n })\n }\n\n return {\n ...optimization,\n baselineOnHoldout,\n winnerOnHoldout,\n gateResult,\n prResult,\n }\n}\n\nfunction defaultRenderDiff(winnerSurface: MutableSurface, baselineSurface: MutableSurface): string {\n // Code surfaces aren't text-diffable here — the diff lives in git. Render\n // the worktree/base refs + summary so the PR body points at the change.\n if (typeof winnerSurface !== 'string' || typeof baselineSurface !== 'string') {\n const fmt = (s: MutableSurface): string =>\n typeof s === 'string'\n ? '(prompt surface)'\n : `worktree=${s.worktreeRef}${s.baseRef ? ` base=${s.baseRef}` : ''}${s.summary ? `\\n${s.summary}` : ''}`\n return `--- baseline\\n${fmt(baselineSurface)}\\n+++ winner\\n${fmt(winnerSurface)}`\n }\n const lines: string[] = []\n lines.push('--- baseline')\n lines.push('+++ winner')\n for (const l of baselineSurface.split('\\n')) lines.push(`- ${l}`)\n for (const l of winnerSurface.split('\\n')) lines.push(`+ ${l}`)\n return lines.join('\\n')\n}\n","/**\n * @experimental\n *\n * VCS-pluggable worktree adapter. One improvement = one worktree, PR-like\n * (multiple commits allowed). A code-tier driver's `propose()` creates a\n * worktree, an agent commits the change into it, and `finalize()` returns a\n * `CodeSurface{ worktreeRef }` the measurement checks out to run the worker\n * against the changed code. On promotion the worktree becomes the PR branch.\n *\n * The interface is VCS-agnostic so a future `jj` ([jj-vcs](https://github.com/jj-vcs/jj))\n * adapter can slot in without touching driver code. Only the git adapter\n * ships today. See `docs/design/self-improvement-engine.md`.\n */\n\nimport { execFileSync } from 'node:child_process'\nimport { existsSync } from 'node:fs'\nimport { basename, isAbsolute, join } from 'node:path'\nimport type { CodeSurface } from '../types'\n\nexport interface Worktree {\n /** Absolute path to the checked-out worktree directory. */\n path: string\n /** The branch the worktree is on (becomes the PR branch on promotion). */\n branch: string\n /** The ref the worktree was forked from. */\n baseRef: string\n}\n\nexport interface WorktreeAdapter {\n /** Create an isolated worktree on a fresh branch off `baseRef`. */\n create(opts: { baseRef: string; label: string }): Promise<Worktree>\n /** Commit any pending changes in the worktree, then return a CodeSurface\n * pointing at it. The agent has already written its change into\n * `worktree.path` by the time this is called. */\n finalize(worktree: Worktree, summary: string): Promise<CodeSurface>\n /** Remove the worktree (and its branch) — called for losing candidates. */\n discard(worktree: Worktree): Promise<void>\n}\n\nexport class WorktreeAdapterError extends Error {\n constructor(\n message: string,\n readonly cause?: unknown,\n ) {\n super(message)\n this.name = 'WorktreeAdapterError'\n }\n}\n\nexport interface GitWorktreeAdapterOptions {\n /** Repo root the worktrees fork from. */\n repoRoot: string\n /** Directory worktrees are created under. Default: `<repoRoot>/.worktrees`. */\n worktreeDir?: string\n /** Branch-name prefix. Default: `improve`. */\n branchPrefix?: string\n /** Test seam — defaults to a real `git` runner. */\n git?: (args: string[], cwd: string) => string\n}\n\nfunction defaultGit(args: string[], cwd: string): string {\n try {\n return execFileSync('git', args, { cwd, encoding: 'utf8' }).trim()\n } catch (err) {\n const stderr =\n err && typeof err === 'object' && 'stderr' in err\n ? String((err as { stderr: unknown }).stderr)\n : ''\n throw new WorktreeAdapterError(`git ${args.join(' ')} failed: ${stderr || String(err)}`, err)\n }\n}\n\n/** Slugify a label into a branch-safe segment. */\nfunction slug(label: string): string {\n return (\n label\n .toLowerCase()\n .replace(/[^a-z0-9]+/g, '-')\n .replace(/^-+|-+$/g, '')\n .slice(0, 48) || 'candidate'\n )\n}\n\nexport function gitWorktreeAdapter(opts: GitWorktreeAdapterOptions): WorktreeAdapter {\n const git = opts.git ?? defaultGit\n const worktreeDir = opts.worktreeDir ?? join(opts.repoRoot, '.worktrees')\n const branchPrefix = opts.branchPrefix ?? 'improve'\n\n return {\n async create({ baseRef, label }) {\n const id = `${slug(label)}-${Date.now().toString(36)}-${Math.random().toString(36).slice(2, 6)}`\n const branch = `${branchPrefix}/${id}`\n const path = join(worktreeDir, id)\n git(['worktree', 'add', '-b', branch, path, baseRef], opts.repoRoot)\n return { path, branch, baseRef }\n },\n\n async finalize(worktree, summary) {\n // Stage + commit any pending changes the agent left in the worktree.\n // A no-op commit is refused by git, so only commit when the tree is dirty.\n const status = git(['status', '--porcelain'], worktree.path)\n if (status.length > 0) {\n git(['add', '-A'], worktree.path)\n git(['commit', '-m', summary], worktree.path)\n }\n return {\n kind: 'code',\n worktreeRef: worktree.path,\n baseRef: worktree.baseRef,\n summary,\n }\n },\n\n async discard(worktree) {\n // Remove the worktree, then delete its branch. Force-remove because the\n // worktree may hold uncommitted experiment state we're discarding.\n git(['worktree', 'remove', '--force', worktree.path], opts.repoRoot)\n git(['branch', '-D', worktree.branch], opts.repoRoot)\n },\n }\n}\n\n/** Resolve a `CodeSurface`'s worktreeRef to a directory the measurement can\n * run the worker in. A path ref is returned as-is; anything else is treated\n * as a ref under the adapter's worktree dir. */\nexport function resolveWorktreePath(surface: CodeSurface, worktreeDir?: string): string {\n if (isAbsolute(surface.worktreeRef) && existsSync(surface.worktreeRef)) return surface.worktreeRef\n if (worktreeDir) return join(worktreeDir, basename(surface.worktreeRef))\n return surface.worktreeRef\n}\n"],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;AAcA,SAAS,gBAAgB;AACzB,SAAS,qBAAqB;AAC9B,SAAS,cAAc;AACvB,SAAS,YAAY;AAiCd,SAAS,WACd,SACkB;AAClB,MAAI,QAAQ,KAAK,aAAa,QAAQ;AACpC,WAAO;AAAA,MACL,QAAQ;AAAA,MACR,QAAQ;AAAA,MACR,QAAQ,qBAAqB,QAAQ,KAAK,QAAQ;AAAA,IACpD;AAAA,EACF;AAEA,QAAM,SAAS,QAAQ,UAAU,CAAC,QAAQ,IAAI;AAC9C,QAAM,SAAS,QAAQ,UAAU,QAAQ,QAAQ,OAAO,aAAa,MAAM,GAAG,EAAE,CAAC;AACjF,QAAM,QACJ,QAAQ,SAAS,kBAAkB,QAAQ,OAAO,aAAa,MAAM,GAAG,CAAC,CAAC;AAE5E,QAAM,OAAO,aAAa,QAAQ,QAAQ,QAAQ,MAAM,QAAQ,YAAY;AAC5E,QAAM,WAAW,KAAK,OAAO,GAAG,gBAAgB,KAAK,IAAI,CAAC,KAAK;AAC/D,gBAAc,UAAU,IAAI;AAE5B,MAAI,QAAQ;AACV,WAAO;AAAA,MACL,QAAQ;AAAA,MACR,QAAQ;AAAA,MACR,QAAQ,0DAA0D,QAAQ,OAAO,IAAI,QAAQ,MAAM,WAAW,MAAM,aAAa,QAAQ;AAAA,IAC3I;AAAA,EACF;AAEA,QAAM,SAAS,QAAQ,UAAU;AACjC,QAAM,SAAS,OAAO;AAAA,IACpB;AAAA,IACA;AAAA,IACA;AAAA,IACA,GAAG,QAAQ,OAAO,IAAI,QAAQ,MAAM;AAAA,IACpC;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,EACF,CAAC;AACD,MAAI,OAAO,WAAW,GAAG;AACvB,WAAO;AAAA,MACL,QAAQ;AAAA,MACR,QAAQ;AAAA,MACR,QAAQ,6BAA6B,OAAO,MAAM,MAAM,OAAO,OAAO,MAAM,GAAG,GAAG,CAAC;AAAA,IACrF;AAAA,EACF;AACA,QAAM,QAAQ,OAAO,OAAO,KAAK;AACjC,SAAO,EAAE,QAAQ,MAAM,OAAO,QAAQ,OAAO,QAAQ,YAAY;AACnE;AAEA,SAAS,aACP,QACA,MACA,MACQ;AACR,QAAM,QAAkB,CAAC;AACzB,QAAM,KAAK,kDAAkD;AAC7D,QAAM,KAAK,EAAE;AACb,QAAM,KAAK,mBAAmB,OAAO,YAAY,IAAI;AACrD,QAAM,KAAK,aAAa,OAAO,IAAI,EAAE;AACrC,QAAM,KAAK,iBAAiB,KAAK,MAAM,OAAO,aAAa,GAAI,CAAC,GAAG;AACnE,QAAM;AAAA,IACJ,uBAAuB,OAAO,WAAW,aAAa,YAAY,OAAO,WAAW,WAAW,aAAa,OAAO,WAAW,YAAY,YAAY,OAAO,WAAW,WAAW;AAAA,EACrL;AACA,QAAM,KAAK,qBAAqB,OAAO,WAAW,aAAa,QAAQ,CAAC,CAAC,EAAE;AAC3E,QAAM,KAAK,EAAE;AACb,QAAM,KAAK,uBAAuB,KAAK,QAAQ,IAAI;AACnD,QAAM,KAAK,EAAE;AACb,aAAW,UAAU,KAAK,QAAS,OAAM,KAAK,KAAK,MAAM,EAAE;AAC3D,MAAI,KAAK,UAAU,OAAW,OAAM,KAAK,YAAY,KAAK,MAAM,QAAQ,CAAC,CAAC,EAAE;AAC5E,QAAM,KAAK,EAAE;AACb,QAAM,KAAK,wBAAwB;AACnC,QAAM,KAAK,EAAE;AACb,QAAM,KAAK,4BAA4B;AACvC,QAAM,KAAK,eAAe;AAC1B,aAAW,KAAK,KAAK,mBAAmB;AACtC,UAAM,SACJ,OAAO,EAAE,WAAW,WAChB,KAAK,UAAU,EAAE,MAAM,EAAE,MAAM,GAAG,EAAE,IACpC,OAAO,EAAE,MAAM,EAAE,MAAM,GAAG,EAAE;AAClC,UAAM,KAAK,KAAK,EAAE,IAAI,MAAM,EAAE,SAAS,WAAM,QAAG,MAAM,MAAM,IAAI;AAAA,EAClE;AACA,QAAM,KAAK,EAAE;AACb,QAAM,KAAK,sBAAsB;AACjC,QAAM,KAAK,EAAE;AACb,QAAM,KAAK,SAAS;AACpB,QAAM,KAAK,KAAK,MAAM,GAAG,GAAI,CAAC;AAC9B,QAAM,KAAK,KAAK;AAChB,QAAM,KAAK,EAAE;AACb,QAAM,KAAK,yBAAyB;AACpC,QAAM,KAAK,EAAE;AACb,QAAM,KAAK,6BAA6B;AACxC,QAAM,KAAK,mBAAmB;AAC9B,aAAW,CAAC,MAAM,GAAG,KAAK,OAAO,QAAQ,OAAO,WAAW,OAAO,GAAG;AACnE,UAAM;AAAA,MACJ,KAAK,IAAI,MAAM,IAAI,KAAK,QAAQ,CAAC,CAAC,OAAO,IAAI,KAAK,CAAC,EAAE,QAAQ,CAAC,CAAC,KAAK,IAAI,KAAK,CAAC,EAAE,QAAQ,CAAC,CAAC,OAAO,IAAI,CAAC;AAAA,IACxG;AAAA,EACF;AACA,SAAO,MAAM,KAAK,IAAI;AACxB;AAEA,SAAS,cAAc,MAAoE;AACzF,MAAI;AACF,UAAM,SAAS,SAAS,MAAM,KAAK,IAAI,QAAQ,EAAE,KAAK,GAAG,CAAC,IAAI;AAAA,MAC5D,KAAK,EAAE,GAAG,QAAQ,KAAK,UAAU,QAAQ,IAAI,oBAAoB,QAAQ,IAAI,YAAY,GAAG;AAAA,MAC5F,OAAO,CAAC,UAAU,QAAQ,MAAM;AAAA,IAClC,CAAC,EAAE,SAAS,MAAM;AAClB,WAAO,EAAE,QAAQ,QAAQ,IAAI,QAAQ,EAAE;AAAA,EACzC,SAAS,KAAK;AACZ,UAAM,IAAI;AACV,WAAO;AAAA,MACL,QAAQ,EAAE,QAAQ,SAAS,MAAM,KAAK;AAAA,MACtC,QAAQ,EAAE,QAAQ,SAAS,MAAM,KAAK;AAAA,MACtC,QAAQ,EAAE,UAAU;AAAA,IACtB;AAAA,EACF;AACF;AAEA,SAAS,SAAS,KAAqB;AACrC,MAAI,wBAAwB,KAAK,GAAG,EAAG,QAAO;AAC9C,SAAO,IAAI,IAAI,QAAQ,MAAM,KAAK,CAAC;AACrC;;;ACrJO,SAAS,mBACd,MAC8B;AAC9B,SAAO;AAAA,IACL,MAAM,gBAAgB,KAAK,QAAQ,IAAI;AAAA,IACvC,MAAM,QAAQ,EAAE,gBAAgB,UAAU,gBAAgB,OAAO,GAAG;AAClE,aAAO,KAAK,QAAQ,OAAO;AAAA,QACzB,UAAU,SAAS,SAAS,IAAI,WAAY,KAAK,YAAY,CAAC;AAAA,QAC9D;AAAA,QACA;AAAA,QACA;AAAA,MACF,CAAC;AAAA,IACH;AAAA,EACF;AACF;;;ACNA,IAAM,oBACJ;AAqBK,SAAS,WAAW,MAA4C;AACrE,QAAM,YAAY,KAAK,aAAa;AACpC,SAAO;AAAA,IACL,MAAM;AAAA,IACN,MAAM,QAAQ,KAAgD;AAC5D,YAAM,SACJ,OAAO,IAAI,mBAAmB,WAC1B,IAAI,iBACJ,KAAK,UAAU,IAAI,cAAc;AACvC,YAAM,EAAE,KAAK,QAAQ,OAAO,IAAI,cAAc,KAAK,WAAW,KAAK,MAAM;AAEzE,YAAM,aAAa,sBAAsB;AAAA,QACvC;AAAA,QACA,eAAe;AAAA,QACf,WAAW;AAAA,QACX,cAAc;AAAA,QACd,YAAY,IAAI;AAAA,QAChB,oBAAoB,KAAK;AAAA,MAC3B,CAAC;AAED,YAAM,SAAS,MAAM;AAAA,QACnB;AAAA,UACE,OAAO,KAAK;AAAA,UACZ,UAAU;AAAA,YACR,EAAE,MAAM,UAAU,SAAS,kBAAkB;AAAA,YAC7C,EAAE,MAAM,QAAQ,SAAS,WAAW;AAAA,UACtC;AAAA,UACA,UAAU;AAAA,UACV,aAAa,KAAK,eAAe;AAAA,UACjC,WAAW,KAAK,aAAa;AAAA,QAC/B;AAAA,QACA,KAAK;AAAA,MACP;AAEA,YAAM,YAAY,wBAAwB,OAAO,SAAS,IAAI,cAAc;AAC5E,YAAM,MAAwB,CAAC;AAC/B,iBAAW,YAAY,WAAW;AAChC,cAAM,OAAO,OAAO,SAAS,YAAY,WAAW,SAAS,QAAQ,KAAK,IAAI;AAC9E,YAAI,QAAQ,SAAS,UAAU,CAAC,IAAI,SAAS,IAAI,EAAG,KAAI,KAAK,IAAI;AAAA,MACnE;AACA,aAAO;AAAA,IACT;AAAA,EACF;AACF;AAKA,SAAS,cACP,KACA,WACA,YAC6D;AAC7D,QAAM,OAAO,IAAI,QAAQ,GAAG,EAAE;AAC9B,MAAI,CAAC,QAAQ,KAAK,WAAW,WAAW,GAAG;AACzC,WAAO,EAAE,KAAK,CAAC,GAAG,QAAQ,CAAC,GAAG,QAAQ,WAAW;AAAA,EACnD;AACA,QAAM,OAAO,CAAC,GAAG,KAAK,UAAU,EAAE,KAAK,CAAC,GAAG,MAAM,EAAE,YAAY,EAAE,SAAS,EAAE,CAAC;AAC7E,MAAI,CAAC,KAAM,QAAO,EAAE,KAAK,CAAC,GAAG,QAAQ,CAAC,GAAG,QAAQ,WAAW;AAE5D,QAAM,UAAU,CAAC,GAAG,KAAK,SAAS,EAAE,KAAK,CAAC,GAAG,MAAM,EAAE,YAAY,EAAE,SAAS;AAC5E,QAAM,UAAU,CAAC,OAA8D;AAAA,IAC7E,IAAI,EAAE;AAAA,IACN,OAAO,EAAE;AAAA,EACX;AACA,QAAM,MAAM,QAAQ,MAAM,GAAG,SAAS,EAAE,IAAI,OAAO;AACnD,QAAM,SAAS,QAAQ,MAAM,CAAC,SAAS,EAAE,QAAQ,EAAE,IAAI,OAAO;AAE9D,QAAM,UAAU,OAAO,QAAQ,KAAK,UAAU,EAC3C,KAAK,CAAC,GAAG,MAAM,EAAE,CAAC,IAAI,EAAE,CAAC,CAAC,EAC1B,MAAM,GAAG,CAAC,EACV,IAAI,CAAC,CAAC,KAAK,KAAK,MAAM,GAAG,GAAG,KAAK,MAAM,QAAQ,CAAC,CAAC,GAAG;AACvD,QAAM,SACJ,QAAQ,SAAS,IAAI,GAAG,UAAU,+BAA0B,QAAQ,KAAK,IAAI,CAAC,KAAK;AAErF,SAAO,EAAE,KAAK,QAAQ,OAAO;AAC/B;;;ACpHO,SAAS,eACX,OACyB;AAC5B,MAAI,MAAM,WAAW,GAAG;AACtB,UAAM,IAAI,MAAM,wCAAwC;AAAA,EAC1D;AACA,SAAO;AAAA,IACL,MAAM,YAAY,MAAM,IAAI,CAAC,MAAM,EAAE,IAAI,EAAE,KAAK,GAAG,CAAC;AAAA,IACpD,MAAM,OAAO,KAA6D;AACxE,YAAM,UAAwE,CAAC;AAC/E,iBAAW,QAAQ,OAAO;AACxB,cAAM,MAAM,MAAM,KAAK,OAAO,GAAG;AACjC,gBAAQ,KAAK,EAAE,MAAM,IAAI,CAAC;AAAA,MAC5B;AAQA,YAAM,YAAY,QAAQ,IAAI,CAAC,MAAM,EAAE,IAAI,QAAQ;AACnD,YAAM,UAAwB,UAAU,MAAM,CAAC,MAAM,MAAM,MAAM,IAC7D,SACA,UAAU,SAAS,cAAc,IAC/B,iBACA,UAAU,SAAS,eAAe,IAChC,kBACA,UAAU,SAAS,MAAM,IACvB,SACA;AAEV,YAAM,eAAe,QAAQ;AAAA,QAAQ,CAAC,MACpC,EAAE,IAAI,kBAAkB,SAAS,IAC7B,EAAE,IAAI,oBACN,CAAC,EAAE,MAAM,EAAE,KAAK,MAAM,QAAQ,EAAE,IAAI,aAAa,QAAQ,QAAQ,EAAE,IAAI,CAAC;AAAA,MAC9E;AAEA,YAAM,UAAU,QAAQ;AAAA,QAAQ,CAAC,MAC/B,EAAE,IAAI,QAAQ,IAAI,CAAC,WAAW,IAAI,EAAE,KAAK,IAAI,KAAK,MAAM,EAAE;AAAA,MAC5D;AAEA,aAAO;AAAA,QACL,UAAU;AAAA,QACV;AAAA,QACA,mBAAmB;AAAA,QACnB,OAAO,QAAQ,CAAC,GAAG,IAAI;AAAA,MACzB;AAAA,IACF;AAAA,EACF;AACF;;;ACpBO,SAAS,sBACd,SAC4B;AAC5B,QAAM,iBAAiB,QAAQ,kBAAkB;AACjD,QAAM,gBAAgB,QAAQ,8BAA8B;AAE5D,SAAO;AAAA,IACL,MAAM;AAAA,IACN,MAAM,OAAO,KAA6D;AACxE,YAAM,UAAoB,CAAC;AAC3B,YAAM,eAA0E,CAAC;AAKjF,YAAM,oBAAoB;AAAA,QACxB,IAAI;AAAA,QACJ,IAAI,uBAAuB,IAAI;AAAA,QAC/B,QAAQ;AAAA,MACV;AACA,YAAM,qBAAqB;AAAA,QACzB,IAAI;AAAA,QACJ,IAAI;AAAA,QACJ,QAAQ;AAAA,MACV;AACA,YAAM,QAAQ,qBAAqB;AACnC,YAAM,cAAc,SAAS;AAC7B,mBAAa,KAAK;AAAA,QAChB,MAAM;AAAA,QACN,QAAQ;AAAA,QACR,QAAQ,EAAE,mBAAmB,oBAAoB,OAAO,eAAe;AAAA,MACzE,CAAC;AACD,UAAI,CAAC,aAAa;AAChB,gBAAQ,KAAK,iBAAiB,MAAM,QAAQ,CAAC,CAAC,gBAAgB,cAAc,EAAE;AAAA,MAChF;AAGA,YAAM,aACJ,QAAQ,cAAc,UACtB,IAAI,KAAK,YAAY,IAAI,KAAK,YAAY,QAAQ;AACpD,mBAAa,KAAK;AAAA,QAChB,MAAM;AAAA,QACN,QAAQ;AAAA,QACR,QAAQ;AAAA,UACN,cAAc,IAAI,KAAK;AAAA,UACvB,aAAa,IAAI,KAAK;AAAA,UACtB,WAAW,QAAQ;AAAA,QACrB;AAAA,MACF,CAAC;AACD,UAAI,CAAC,YAAY;AACf,gBAAQ;AAAA,UACN,UAAU,IAAI,KAAK,YAAY,IAAI,KAAK,UAAU,QAAQ,CAAC,CAAC,aAAa,QAAQ,SAAS;AAAA,QAC5F;AAAA,MACF;AAGA,YAAM,kBAAkB,QAAQ,iBAC5B,aAAa,IAAI,oBAAoB,QAAQ,cAAc,IAC3D,EAAE,QAAQ,MAAM,UAAU,CAAC,EAAE;AACjC,mBAAa,KAAK;AAAA,QAChB,MAAM;AAAA,QACN,QAAQ,gBAAgB;AAAA,QACxB,QAAQ;AAAA,UACN,UAAU,gBAAgB,SAAS;AAAA,UACnC,QAAQ,gBAAgB,SAAS,MAAM,GAAG,CAAC;AAAA,QAC7C;AAAA,MACF,CAAC;AACD,UAAI,CAAC,gBAAgB,QAAQ;AAC3B,gBAAQ,KAAK,0BAA0B,gBAAgB,SAAS,MAAM,YAAY;AAAA,MACpF;AAGA,UAAI,sBAAkD;AACtD,UAAI,QAAQ,cAAc,QAAQ,WAAW,UAAU,IAAI;AACzD,8BAAsB,oBAAoB,EAAE,MAAM,QAAQ,WAAW,CAAC;AAAA,MACxE;AAIA,YAAM,kBAAkB;AACxB,YAAM,kBAAkB,qBAAqB,YAAY,CAAC,GAAG;AAAA,QAC3D,CAAC,MAAM,EAAE,YAAY;AAAA,MACvB;AACA,YAAM,oBACJ,CAAC,uBACD,CAAC,iBACA,eAAe,WAAW,KAAK,oBAAoB,YAAY;AAClE,mBAAa,KAAK;AAAA,QAChB,MAAM;AAAA,QACN,QAAQ;AAAA,QACR,QAAQ,EAAE,QAAQ,qBAAqB,oBAAoB,eAAe,OAAO;AAAA,MACnF,CAAC;AACD,UAAI,CAAC,mBAAmB;AACtB,gBAAQ;AAAA,UACN,mCAAmC,eAAe,MAAM,sCAAsC,oBAAqB,OAAO;AAAA,QAC5H;AAAA,MACF;AAGA,UAAI,eAAoC;AACxC,UAAI,QAAQ,cAAc,QAAQ,WAAW,UAAU,IAAI;AACzD,uBAAe,YAAY,QAAQ,YAAY,CAAC,CAAC;AAAA,MACnD;AAEA,YAAM,eAAe,cAAc,UAAU,CAAC,GAAG,OAAO,CAAC,MAAM,EAAE,aAAa,OAAO;AACrF,YAAM,aAAa,YAAY,WAAW;AAC1C,mBAAa,KAAK;AAAA,QAChB,MAAM;AAAA,QACN,QAAQ;AAAA,QACR,QAAQ,EAAE,aAAa,cAAc,OAAO,UAAU,GAAG,aAAa,YAAY,OAAO;AAAA,MAC3F,CAAC;AACD,UAAI,CAAC,YAAY;AACf,gBAAQ,KAAK,wBAAwB,YAAY,MAAM,EAAE;AAAA,MAC3D;AAGA,YAAM,YAAY,aAAa,MAAM,CAAC,MAAM,EAAE,MAAM;AACpD,YAAM,WAAW,YAAY,SAAS;AAEtC,aAAO;AAAA,QACL;AAAA,QACA,SAAS,QAAQ,SAAS,IAAI,UAAU,CAAC,kBAAkB;AAAA,QAC3D,mBAAmB;AAAA,QACnB;AAAA,MACF;AAAA,IACF;AAAA,EACF;AACF;AAEA,SAAS,cACP,WACA,mBACA,WACQ;AACR,MAAI,CAAC,aAAa,UAAU,SAAS,EAAG,QAAO;AAC/C,QAAM,cAAc,IAAI,IAAI,UAAU,IAAI,CAAC,MAAM,EAAE,EAAE,CAAC;AACtD,QAAM,aAAuB,CAAC;AAC9B,aAAW,CAAC,QAAQ,MAAM,KAAK,mBAAmB;AAChD,UAAM,aAAa,OAAO,MAAM,GAAG,EAAE,CAAC,KAAK;AAC3C,QAAI,CAAC,YAAY,IAAI,UAAU,EAAG;AAClC,UAAM,iBAAiB,OAAO,OAAO,MAAM,EAAE,IAAI,CAAC,MAAM,EAAE,SAAS;AACnE,QAAI,eAAe,WAAW,EAAG;AACjC,eAAW,KAAK,eAAe,OAAO,CAAC,GAAG,MAAM,IAAI,GAAG,CAAC,IAAI,eAAe,MAAM;AAAA,EACnF;AACA,MAAI,WAAW,WAAW,EAAG,QAAO;AACpC,SAAO,WAAW,OAAO,CAAC,GAAG,MAAM,IAAI,GAAG,CAAC,IAAI,WAAW;AAC5D;AAEA,SAAS,aACP,WACA,SAC8E;AAC9E,QAAM,WAA0D,CAAC;AACjE,aAAW,CAAC,SAAS,QAAQ,KAAK,WAAW;AAC3C,UAAM,OAAO,YAAY,QAAQ;AACjC,QAAI,SAAS,OAAW;AACxB,eAAW,UAAU,SAAS;AAC5B,YAAM,UAAU,mBAAmB,MAAM,CAAC,GAAG,MAAM;AACnD,UAAI,CAAC,QAAQ,QAAQ;AACnB,iBAAS,KAAK,EAAE,YAAY,OAAO,IAAI,QAAQ,QAAQ,UAAU,wBAAwB,CAAC;AAAA,MAC5F;AAAA,IACF;AAAA,EACF;AACA,SAAO,EAAE,QAAQ,SAAS,WAAW,GAAG,SAAS;AACnD;AAEA,SAAS,YAAY,UAAuC;AAC1D,MAAI,OAAO,aAAa,SAAU,QAAO;AACzC,MAAI,YAAY,OAAO,aAAa,UAAU;AAC5C,UAAM,MAAM;AACZ,QAAI,OAAO,IAAI,SAAS,SAAU,QAAO,IAAI;AAC7C,QAAI,OAAO,IAAI,WAAW,SAAU,QAAO,IAAI;AAC/C,QAAI,OAAO,IAAI,YAAY,SAAU,QAAO,IAAI;AAAA,EAClD;AACA,SAAO;AACT;;;AC5MO,SAAS,YACd,SAC4B;AAC5B,QAAM,iBAAiB,QAAQ,kBAAkB;AACjD,SAAO;AAAA,IACL,MAAM;AAAA,IACN,MAAM,OAAO,KAA6D;AACxE,YAAM,cAAc,IAAI,IAAI,QAAQ,UAAU,IAAI,CAAC,MAAM,EAAE,EAAE,CAAC;AAG9D,YAAM,WAAW,iBAAiB,IAAI,uBAAuB,IAAI,aAAa,WAAW;AACzF,YAAM,YAAY,iBAAiB,IAAI,aAAa,WAAW;AAC/D,YAAM,QAAQ,YAAY;AAC1B,YAAM,SAAS,SAAS;AACxB,aAAO;AAAA,QACL,UAAU,SAAS,SAAS;AAAA,QAC5B,SAAS,SACL,CAAC,kBAAkB,MAAM,QAAQ,CAAC,CAAC,WAAM,cAAc,EAAE,IACzD,CAAC,kBAAkB,MAAM,QAAQ,CAAC,CAAC,MAAM,cAAc,EAAE;AAAA,QAC7D,mBAAmB;AAAA,UACjB,EAAE,MAAM,eAAe,QAAQ,QAAQ,EAAE,UAAU,WAAW,OAAO,eAAe,EAAE;AAAA,QACxF;AAAA,QACA;AAAA,MACF;AAAA,IACF;AAAA,EACF;AACF;AAEA,SAAS,iBACP,mBACA,aACQ;AACR,QAAM,aAAuB,CAAC;AAC9B,aAAW,CAAC,QAAQ,MAAM,KAAK,mBAAmB;AAChD,UAAM,aAAa,OAAO,MAAM,GAAG,EAAE,CAAC,KAAK;AAC3C,QAAI,CAAC,YAAY,IAAI,UAAU,EAAG;AAClC,UAAM,OAAO,OAAO,OAAO,MAAM,EAAE,IAAI,CAAC,MAAM,EAAE,SAAS;AACzD,QAAI,KAAK,SAAS,EAAG,YAAW,KAAK,KAAK,OAAO,CAAC,GAAG,MAAM,IAAI,GAAG,CAAC,IAAI,KAAK,MAAM;AAAA,EACpF;AACA,SAAO,WAAW,WAAW,IAAI,IAAI,WAAW,OAAO,CAAC,GAAG,MAAM,IAAI,GAAG,CAAC,IAAI,WAAW;AAC1F;;;AC1BA,SAAS,kBAAkB;AAC3B,SAAS,YAAY,WAAW,cAAc,iBAAAA,sBAAqB;AACnE,SAAS,QAAAC,aAAY;AAmBd,IAAM,4BAAN,cAAwC,MAAM;AAAA,EACnD,YACkB,MAChB,SACA;AACA,UAAM,OAAO;AAHG;AAIhB,SAAK,OAAO;AAAA,EACd;AAAA,EALkB;AAMpB;AAQO,IAAM,yBAAN,MAA6D;AAAA,EAIlE,YAA6B,SAAwC;AAAxC;AAC3B,QAAI,CAAC,WAAW,QAAQ,IAAI,EAAG,WAAU,QAAQ,MAAM,EAAE,WAAW,KAAK,CAAC;AAC1E,SAAK,MAAM,QAAQ,OAAO,KAAK;AAAA,EACjC;AAAA,EAH6B;AAAA,EAHZ;AAAA,EACA,aAAa,oBAAI,IAA4B;AAAA,EAO9D,MAAM,QAAQ,OAA4C;AACxD,SAAK,iBAAiB,KAAK;AAC3B,SAAK,gBAAgB,KAAK;AAC1B,UAAM,SAAS,KAAK,SAAS,KAAK;AAClC,UAAM,OAAO,KAAK,cAAc,MAAM,MAAM;AAC5C,UAAM,OAAO,GAAG,KAAK,UAAU,MAAM,CAAC;AAAA;AAItC,eAAW,MAAM,IAAI;AAAA,EACvB;AAAA,EAEA,MAAM,OAAO,MAAmE;AAC9E,QAAI,CAAC,KAAK,OAAO;AACf,YAAM,IAAI;AAAA,QACR;AAAA,QACA;AAAA,MACF;AAAA,IACF;AACA,QAAI,CAAC,KAAK,gBAAgB;AACxB,YAAM,IAAI;AAAA,QACR;AAAA,QACA;AAAA,MACF;AAAA,IACF;AAEA,UAAM,MAA+B,CAAC;AACtC,eAAW,UAAU,aAAa;AAGhC,UAAI,KAAK,UAAU,WAAW,WAAW,oBAAoB;AAC3D,cAAM,WAAW,qBAAqB,KAAK,QAAQ,QAAQ,kBAAkB;AAC7E,YAAI,CAAC,SAAU;AAAA,MACjB;AACA,YAAM,OAAO,KAAK,cAAc,MAAM;AACtC,UAAI,CAAC,WAAW,IAAI,EAAG;AACvB,YAAM,QAAQ,aAAa,MAAM,MAAM,EAAE,MAAM,IAAI,EAAE,OAAO,OAAO;AACnE,iBAAW,QAAQ,OAAO;AACxB,YAAI;AACJ,YAAI;AACF,mBAAS,KAAK,MAAM,IAAI;AAAA,QAC1B,QAAQ;AACN;AAAA,QACF;AACA,YAAI,CAAC,cAAc,QAAQ,MAAM,MAAM,EAAG;AAC1C,YAAI,KAAK,MAAM;AAAA,MACjB;AAAA,IACF;AAGA,QAAI,KAAK,CAAC,GAAG,MAAM;AACjB,UAAI,EAAE,eAAe,EAAE,WAAY,QAAO,EAAE,WAAW,cAAc,EAAE,UAAU;AACjF,aAAO,EAAE,WAAW,cAAc,EAAE,UAAU;AAAA,IAChD,CAAC;AAED,WAAO,IAAI,MAAM,GAAG,KAAK,KAAK;AAAA,EAChC;AAAA,EAEA,MAAM,OAAmF;AACvF,UAAM,WAAmC,CAAC;AAC1C,QAAI,QAAQ;AACZ,eAAW,UAAU,aAAa;AAChC,YAAM,OAAO,KAAK,cAAc,MAAM;AACtC,UAAI,CAAC,WAAW,IAAI,GAAG;AACrB,iBAAS,MAAM,IAAI;AACnB;AAAA,MACF;AACA,YAAM,QAAQ,aAAa,MAAM,MAAM,EAAE,MAAM,IAAI,EAAE,OAAO,OAAO,EAAE;AACrE,eAAS,MAAM,IAAI;AACnB,eAAS;AAAA,IACX;AAIA,WAAO,EAAE,OAAO,OAAO,MAAM,OAAO,SAAS;AAAA,EAC/C;AAAA,EAEQ,iBAAiB,OAAmC;AAC1D,QAAI,CAAC,MAAM,QAAQ;AACjB,YAAM,IAAI;AAAA,QACR;AAAA,QACA;AAAA,MACF;AAAA,IACF;AACA,QAAI,CAAC,MAAM,qBAAqB,MAAM,kBAAkB,WAAW,GAAG;AACpE,YAAM,IAAI;AAAA,QACR;AAAA,QACA;AAAA,MACF;AAAA,IACF;AACA,QAAI,CAAC,MAAM,YAAY;AACrB,YAAM,IAAI;AAAA,QACR;AAAA,QACA;AAAA,MACF;AAAA,IACF;AACA,QAAI,CAAC,MAAM,iBAAiB;AAC1B,YAAM,IAAI;AAAA,QACR;AAAA,QACA;AAAA,MACF;AAAA,IACF;AACA,QAAI,CAAC,YAAY,SAAS,MAAM,MAAM,GAAG;AACvC,YAAM,IAAI;AAAA,QACR;AAAA,QACA,+CAA+C,YAAY,KAAK,IAAI,CAAC;AAAA,MACvE;AAAA,IACF;AAAA,EACF;AAAA,EAEQ,gBAAgB,OAAmC;AACzD,UAAM,MAAM,KAAK,QAAQ;AACzB,QAAI,CAAC,OAAO,CAAC,MAAM,gBAAiB;AACpC,UAAM,MAAM,KAAK,IAAI;AACrB,UAAM,WAAW;AACjB,QAAI,QAAQ,KAAK,WAAW,IAAI,MAAM,eAAe;AACrD,QAAI,CAAC,SAAS,MAAM,MAAM,iBAAiB,UAAU;AACnD,cAAQ,EAAE,QAAQ,MAAM,iBAAiB,eAAe,KAAK,OAAO,EAAE;AACtE,WAAK,WAAW,IAAI,MAAM,iBAAiB,KAAK;AAAA,IAClD;AACA,QAAI,MAAM,SAAS,KAAK;AACtB,YAAM,IAAI;AAAA,QACR;AAAA,QACA,gCAAgC,MAAM,eAAe,aAAa,GAAG;AAAA,MACvE;AAAA,IACF;AACA,UAAM,SAAS;AAAA,EACjB;AAAA,EAEQ,SAAS,OAAoD;AACnE,UAAM,aAAa;AAAA,MACjB,KAAK,UAAU;AAAA,QACb,IAAI,MAAM,SAAS;AAAA,QACnB,KAAK,MAAM;AAAA,QACX,IAAI,MAAM;AAAA,QACV,KAAK,MAAM;AAAA,MACb,CAAC;AAAA,IACH;AAGA,WAAO;AAAA,MACL,GAAG;AAAA,MACH;AAAA,MACA,OAAO;AAAA,IACT;AAAA,EACF;AAAA,EAEQ,cAAc,QAAwB;AAC5C,WAAOA,MAAK,KAAK,QAAQ,MAAM,GAAG,MAAM,QAAQ;AAAA,EAClD;AACF;AAEA,IAAM,cAAgD;AAAA,EACpD;AAAA,EACA;AAAA,EACA;AAAA,EACA;AAAA,EACA;AACF;AAEA,SAAS,qBACP,QACA,QACS;AACT,MAAI,CAAC,OAAQ,QAAO;AACpB,MAAI,MAAM,QAAQ,MAAM,EAAG,QAAO,OAAO,SAAS,MAAM;AACxD,SAAO,WAAW;AACpB;AAEA,SAAS,cACP,QACA,MACA,QACS;AAET,MAAI,KAAK,UAAU,WAAW,OAAO,cAAc,KAAK,eAAgB,QAAO;AAC/E,MAAI,KAAK,UAAU,UAAU,OAAO,aAAa,KAAK,eAAgB,QAAO;AAE7E,QAAM,IAAI,KAAK;AACf,MAAI,CAAC,EAAG,QAAO;AACf,MAAI,EAAE,QAAQ,OAAO,SAAS,SAAS,EAAE,KAAM,QAAO;AACtD,MAAI,EAAE,QAAQ;AACZ,UAAM,UAAU,MAAM,QAAQ,EAAE,MAAM,IAAI,EAAE,SAAS,CAAC,EAAE,MAAM;AAC9D,QAAI,CAAC,QAAQ,SAAS,MAAe,EAAG,QAAO;AAAA,EACjD;AACA,MAAI,EAAE,iBAAiB,UAAa,EAAE,iBAAiB,QAAW;AAChE,UAAM,aAAa,OAAO,OAAO,OAAO,WAAW,EAAE,IAAI,CAAC,MAAM,EAAE,SAAS;AAC3E,UAAM,MAAM,WAAW,WAAW,IAAI,IAAI,KAAK,IAAI,GAAG,UAAU;AAChE,QAAI,EAAE,iBAAiB,UAAa,MAAM,EAAE,aAAc,QAAO;AACjE,QAAI,EAAE,iBAAiB,UAAa,MAAM,EAAE,aAAc,QAAO;AAAA,EACnE;AACA,SAAO;AACT;AAEA,SAAS,OAAO,OAAuB;AACrC,SAAO,WAAW,QAAQ,EAAE,OAAO,KAAK,EAAE,OAAO,KAAK,EAAE,MAAM,GAAG,EAAE;AACrE;AAEA,SAAS,WAAW,MAAc,MAAoB;AACpD,MAAI,WAAW,IAAI,GAAG;AACpB,UAAM,WAAW,aAAa,MAAM,MAAM;AAC1C,IAAAD,eAAc,MAAM,WAAW,IAAI;AAAA,EACrC,OAAO;AACL,IAAAA,eAAc,MAAM,IAAI;AAAA,EAC1B;AACF;;;ACtQA,eAAsB,QACpB,MAC+C;AAC/C,SAAO,YAAY,IAAI;AACzB;;;ACFA,SAAS,cAAAE,mBAAkB;AAkD3B,eAAsB,gBACpB,MACsD;AACtD,QAAM,cAAc,KAAK,eAAe;AAGxC,QAAM,mBAAmB,MAAM,YAAkC;AAAA,IAC/D,GAAG;AAAA,IACH,UAAU,CAAC,UAAU,QAAQ,KAAK,oBAAoB,KAAK,iBAAiB,UAAU,GAAG;AAAA,IACzF,QAAQ,GAAG,KAAK,MAAM;AAAA,EACxB,CAAC;AAED,QAAM,cAA0E,CAAC;AACjF,QAAM,UAA8B,CAAC;AACrC,MAAI,kBAAoC,CAAC,KAAK,eAAe;AAC7D,MAAI,gBAAgB,KAAK;AACzB,MAAI,oBAAoB,YAAY,KAAK,eAAe;AACxD,MAAI,kBAAkBC,eAAc,gBAAgB;AAEpD,WAAS,MAAM,GAAG,MAAM,KAAK,gBAAgB,OAAO;AAElD,QAAI,KAAK,OAAO,SAAS,EAAE,QAAQ,CAAC,EAAE,KAAM;AAI5C,UAAM,aAAa,MAAM,KAAK,OAAO,QAAQ;AAAA,MAC3C,gBAAgB,gBAAgB,CAAC,KAAK,KAAK;AAAA,MAC3C;AAAA,MACA,UAAU,CAAC;AAAA,MACX,gBAAgB,KAAK;AAAA,MACrB,YAAY;AAAA,MACZ,QAAQ,IAAI,gBAAgB,EAAE;AAAA,MAC9B,QAAQ,KAAK;AAAA,MACb,SAAS,KAAK,gBAAgB,KAAK,iBAAiB,QAAQ,KAAK,eAAe;AAAA,MAChF,qBAAqB,KAAK;AAAA,IAC5B,CAAC;AAGD,UAAM,iBAKD,CAAC;AACN,aAAS,IAAI,GAAG,IAAI,WAAW,QAAQ,KAAK;AAC1C,YAAM,UAAU,WAAW,CAAC;AAC5B,YAAM,OAAO,YAAY,OAAO;AAChC,YAAM,WAAW,MAAM,YAAkC;AAAA,QACvD,GAAG;AAAA,QACH,UAAU,CAAC,UAAU,QAAQ,KAAK,oBAAoB,SAAS,UAAU,GAAG;AAAA,QAC5E,QAAQ,GAAG,KAAK,MAAM,QAAQ,GAAG,cAAc,CAAC;AAAA,MAClD,CAAC;AACD,YAAM,YAAYA,eAAc,QAAQ;AACxC,qBAAe,KAAK,EAAE,aAAa,MAAM,SAAS,UAAU,UAAU,CAAC;AAAA,IACzE;AAGA,mBAAe,KAAK,CAAC,GAAG,MAAM,EAAE,YAAY,EAAE,SAAS;AACvD,UAAM,WAAW,eAAe,MAAM,GAAG,WAAW;AACpD,sBAAkB,SAAS,IAAI,CAAC,MAAM,EAAE,OAAO;AAC/C,UAAM,MAAM,eAAe,CAAC;AAC5B,QAAI,OAAO,IAAI,YAAY,iBAAiB;AAC1C,sBAAgB,IAAI;AACpB,0BAAoB,IAAI;AACxB,wBAAkB,IAAI;AAAA,IACxB;AAEA,UAAM,SAA2B;AAAA,MAC/B,iBAAiB;AAAA,MACjB,YAAY,eAAe,IAAI,CAAC,MAAM;AACpC,cAAM,YAAY,mBAAmB,EAAE,QAAQ;AAC/C,eAAO;AAAA,UACL,aAAa,EAAE;AAAA,UACf,WAAW,EAAE;AAAA,UACb,MAAM,CAAC,EAAE,WAAW,EAAE,SAAS;AAAA,UAC/B,YAAY,UAAU;AAAA,UACtB,WAAW,UAAU;AAAA,QACvB;AAAA,MACF,CAAC;AAAA,MACD,UAAU,SAAS,IAAI,CAAC,MAAM,EAAE,WAAW;AAAA,IAC7C;AACA,YAAQ,KAAK,MAAM;AACnB,gBAAY,KAAK;AAAA,MACf;AAAA,MACA,UAAU,eAAe,IAAI,CAAC,OAAO;AAAA,QACnC,aAAa,EAAE;AAAA,QACf,SAAS,EAAE;AAAA,QACX,UAAU,EAAE;AAAA,MACd,EAAE;AAAA,IACJ,CAAC;AAAA,EACH;AAEA,SAAO;AAAA,IACL;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,EACF;AACF;AAEO,SAAS,YAAY,SAAiC;AAG3D,QAAM,WACJ,OAAO,YAAY,WACf,UACA,KAAK,UAAU;AAAA,IACb,MAAM,QAAQ;AAAA,IACd,aAAa,QAAQ;AAAA,IACrB,SAAS,QAAQ,WAAW;AAAA,EAC9B,CAAC;AACP,SAAOC,YAAW,QAAQ,EAAE,OAAO,QAAQ,EAAE,OAAO,KAAK,EAAE,MAAM,GAAG,EAAE;AACxE;AAEA,SAASD,eACP,UACQ;AACR,QAAM,aAAuB,CAAC;AAC9B,aAAW,QAAQ,SAAS,OAAO;AACjC,UAAM,iBAAiB,OAAO,OAAO,KAAK,WAAW,EAAE,IAAI,CAAC,MAAM,EAAE,SAAS;AAC7E,QAAI,eAAe,SAAS,GAAG;AAC7B,iBAAW,KAAK,eAAe,OAAO,CAAC,GAAG,MAAM,IAAI,GAAG,CAAC,IAAI,eAAe,MAAM;AAAA,IACnF;AAAA,EACF;AACA,SAAO,WAAW,WAAW,IAAI,IAAI,WAAW,OAAO,CAAC,GAAG,MAAM,IAAI,GAAG,CAAC,IAAI,WAAW;AAC1F;AAIA,SAAS,mBACP,UAIA;AACA,QAAM,UAAkC,CAAC;AACzC,QAAM,YAAoC,CAAC;AAC3C,QAAM,aAAa,oBAAI,IAAsB;AAC7C,aAAW,QAAQ,SAAS,OAAO;AACjC,UAAM,cAAc,OAAO,OAAO,KAAK,WAAW;AAClD,QAAI,YAAY,WAAW,EAAG;AAC9B,UAAM,gBAAgB,YAAY,OAAO,CAAC,GAAG,MAAM,IAAI,EAAE,WAAW,CAAC,IAAI,YAAY;AACrF,UAAM,MAAM,WAAW,IAAI,KAAK,UAAU,KAAK,CAAC;AAChD,QAAI,KAAK,aAAa;AACtB,eAAW,IAAI,KAAK,YAAY,GAAG;AACnC,eAAW,SAAS,aAAa;AAC/B,iBAAW,CAAC,KAAK,KAAK,KAAK,OAAO,QAAQ,MAAM,UAAU,GAAG;AAC3D,gBAAQ,GAAG,KAAK,QAAQ,GAAG,KAAK,KAAK;AACrC,kBAAU,GAAG,KAAK,UAAU,GAAG,KAAK,KAAK;AAAA,MAC3C;AAAA,IACF;AAAA,EACF;AACA,QAAM,aAAqC,CAAC;AAC5C,aAAW,OAAO,OAAO,KAAK,OAAO,GAAG;AACtC,UAAM,QAAQ,UAAU,GAAG,KAAK;AAChC,eAAW,GAAG,IAAI,QAAQ,KAAK,QAAQ,GAAG,KAAK,KAAK,QAAQ;AAAA,EAC9D;AACA,QAAM,YAAY,CAAC,GAAG,WAAW,QAAQ,CAAC,EAAE,IAAI,CAAC,CAAC,YAAY,KAAK,OAAO;AAAA,IACxE;AAAA,IACA,WAAW,MAAM,OAAO,CAAC,GAAG,MAAM,IAAI,GAAG,CAAC,IAAI,MAAM;AAAA,EACtD,EAAE;AACF,SAAO,EAAE,YAAY,UAAU;AACjC;;;ACxKA,eAAsB,mBACpB,MACyD;AAGzD,MAAK,KAAa,kBAAkB,UAAU;AAC5C,UAAM,IAAI;AAAA,MACR;AAAA,IACF;AAAA,EACF;AAKA,MAAI,KAAK,YAAY,SAAS,KAAK,QAAQ;AACzC,UAAM,IAAI;AAAA,MACR;AAAA,IACF;AAAA,EACF;AACA,MAAI,KAAK,kBAAkB,SAAS,CAAC,KAAK,WAAW,CAAC,KAAK,SAAS;AAClE,UAAM,IAAI,MAAM,mEAAmE;AAAA,EACrF;AAGA,QAAM,eAAe,MAAM,gBAAgB,IAAI;AAG/C,QAAM,EAAE,aAAAE,aAAY,IAAI,MAAM,OAAO,6BAAiB;AAEtD,QAAM,oBAAoB,MAAMA,aAAkC;AAAA,IAChE,GAAG;AAAA,IACH,WAAW,KAAK;AAAA,IAChB,UAAU,CAAC,UAAU,QAAQ,KAAK,oBAAoB,KAAK,iBAAiB,UAAU,GAAG;AAAA,IACzF,QAAQ,GAAG,KAAK,MAAM;AAAA,EACxB,CAAC;AAED,QAAM,kBAAkB,MAAMA,aAAkC;AAAA,IAC9D,GAAG;AAAA,IACH,WAAW,KAAK;AAAA,IAChB,UAAU,CAAC,UAAU,QACnB,KAAK,oBAAoB,aAAa,eAAe,UAAU,GAAG;AAAA,IACpE,QAAQ,GAAG,KAAK,MAAM;AAAA,EACxB,CAAC;AAUD,QAAM,qBAAqB,oBAAI,IAAuB;AACtD,QAAM,oBAAoB,oBAAI,IAAuB;AACrD,QAAM,cAAwB,oBAAI,IAAI;AACtC,QAAM,sBAAgC,oBAAI,IAAI;AAC9C,aAAW,QAAQ,gBAAgB,OAAO;AACxC,uBAAmB,IAAI,KAAK,QAAQ,KAAK,QAAQ;AACjD,gBAAY,IAAI,KAAK,QAAQ,KAAK,WAAW;AAAA,EAC/C;AACA,aAAW,QAAQ,kBAAkB,OAAO;AAC1C,sBAAkB,IAAI,KAAK,QAAQ,KAAK,QAAQ;AAChD,wBAAoB,IAAI,KAAK,QAAQ,KAAK,WAAW;AAAA,EACvD;AAEA,QAAM,aAAa,MAAM,KAAK,KAAK,OAAO;AAAA,IACxC;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,IACA,WAAW,KAAK;AAAA,IAChB,MAAM;AAAA,MACJ,WAAW,gBAAgB,WAAW;AAAA,MACtC,UAAU,kBAAkB,WAAW;AAAA,IACzC;AAAA,IACA,QAAQ,IAAI,gBAAgB,EAAE;AAAA,EAChC,CAAC;AAGD,MAAI;AACJ,MAAI,KAAK,kBAAkB,QAAQ,WAAW,aAAa,QAAQ;AACjE,UAAM,SAAS,KAAK,sBAAsB;AAC1C,UAAM,eAAe,OAAO,aAAa,eAAe,KAAK,eAAe;AAC5E,eAAW,WAAW;AAAA,MACpB,QAAQ;AAAA,MACR,MAAM;AAAA,MACN;AAAA,MACA,SAAS,KAAK;AAAA,MACd,QAAQ,KAAK;AAAA,IACf,CAAC;AAAA,EACH;AAEA,SAAO;AAAA,IACL,GAAG;AAAA,IACH;AAAA,IACA;AAAA,IACA;AAAA,IACA;AAAA,EACF;AACF;AAEA,SAAS,kBAAkB,eAA+B,iBAAyC;AAGjG,MAAI,OAAO,kBAAkB,YAAY,OAAO,oBAAoB,UAAU;AAC5E,UAAM,MAAM,CAAC,MACX,OAAO,MAAM,WACT,qBACA,YAAY,EAAE,WAAW,GAAG,EAAE,UAAU,SAAS,EAAE,OAAO,KAAK,EAAE,GAAG,EAAE,UAAU;AAAA,EAAK,EAAE,OAAO,KAAK,EAAE;AAC3G,WAAO;AAAA,EAAiB,IAAI,eAAe,CAAC;AAAA;AAAA,EAAiB,IAAI,aAAa,CAAC;AAAA,EACjF;AACA,QAAM,QAAkB,CAAC;AACzB,QAAM,KAAK,cAAc;AACzB,QAAM,KAAK,YAAY;AACvB,aAAW,KAAK,gBAAgB,MAAM,IAAI,EAAG,OAAM,KAAK,KAAK,CAAC,EAAE;AAChE,aAAW,KAAK,cAAc,MAAM,IAAI,EAAG,OAAM,KAAK,KAAK,CAAC,EAAE;AAC9D,SAAO,MAAM,KAAK,IAAI;AACxB;;;ACvKA,SAAS,oBAAoB;AAC7B,SAAS,cAAAC,mBAAkB;AAC3B,SAAS,UAAU,YAAY,QAAAC,aAAY;AAuBpC,IAAM,uBAAN,cAAmC,MAAM;AAAA,EAC9C,YACE,SACS,OACT;AACA,UAAM,OAAO;AAFJ;AAGT,SAAK,OAAO;AAAA,EACd;AAAA,EAJW;AAKb;AAaA,SAAS,WAAW,MAAgB,KAAqB;AACvD,MAAI;AACF,WAAO,aAAa,OAAO,MAAM,EAAE,KAAK,UAAU,OAAO,CAAC,EAAE,KAAK;AAAA,EACnE,SAAS,KAAK;AACZ,UAAM,SACJ,OAAO,OAAO,QAAQ,YAAY,YAAY,MAC1C,OAAQ,IAA4B,MAAM,IAC1C;AACN,UAAM,IAAI,qBAAqB,OAAO,KAAK,KAAK,GAAG,CAAC,YAAY,UAAU,OAAO,GAAG,CAAC,IAAI,GAAG;AAAA,EAC9F;AACF;AAGA,SAAS,KAAK,OAAuB;AACnC,SACE,MACG,YAAY,EACZ,QAAQ,eAAe,GAAG,EAC1B,QAAQ,YAAY,EAAE,EACtB,MAAM,GAAG,EAAE,KAAK;AAEvB;AAEO,SAAS,mBAAmB,MAAkD;AACnF,QAAM,MAAM,KAAK,OAAO;AACxB,QAAM,cAAc,KAAK,eAAeA,MAAK,KAAK,UAAU,YAAY;AACxE,QAAM,eAAe,KAAK,gBAAgB;AAE1C,SAAO;AAAA,IACL,MAAM,OAAO,EAAE,SAAS,MAAM,GAAG;AAC/B,YAAM,KAAK,GAAG,KAAK,KAAK,CAAC,IAAI,KAAK,IAAI,EAAE,SAAS,EAAE,CAAC,IAAI,KAAK,OAAO,EAAE,SAAS,EAAE,EAAE,MAAM,GAAG,CAAC,CAAC;AAC9F,YAAM,SAAS,GAAG,YAAY,IAAI,EAAE;AACpC,YAAM,OAAOA,MAAK,aAAa,EAAE;AACjC,UAAI,CAAC,YAAY,OAAO,MAAM,QAAQ,MAAM,OAAO,GAAG,KAAK,QAAQ;AACnE,aAAO,EAAE,MAAM,QAAQ,QAAQ;AAAA,IACjC;AAAA,IAEA,MAAM,SAAS,UAAU,SAAS;AAGhC,YAAM,SAAS,IAAI,CAAC,UAAU,aAAa,GAAG,SAAS,IAAI;AAC3D,UAAI,OAAO,SAAS,GAAG;AACrB,YAAI,CAAC,OAAO,IAAI,GAAG,SAAS,IAAI;AAChC,YAAI,CAAC,UAAU,MAAM,OAAO,GAAG,SAAS,IAAI;AAAA,MAC9C;AACA,aAAO;AAAA,QACL,MAAM;AAAA,QACN,aAAa,SAAS;AAAA,QACtB,SAAS,SAAS;AAAA,QAClB;AAAA,MACF;AAAA,IACF;AAAA,IAEA,MAAM,QAAQ,UAAU;AAGtB,UAAI,CAAC,YAAY,UAAU,WAAW,SAAS,IAAI,GAAG,KAAK,QAAQ;AACnE,UAAI,CAAC,UAAU,MAAM,SAAS,MAAM,GAAG,KAAK,QAAQ;AAAA,IACtD;AAAA,EACF;AACF;AAKO,SAAS,oBAAoB,SAAsB,aAA8B;AACtF,MAAI,WAAW,QAAQ,WAAW,KAAKD,YAAW,QAAQ,WAAW,EAAG,QAAO,QAAQ;AACvF,MAAI,YAAa,QAAOC,MAAK,aAAa,SAAS,QAAQ,WAAW,CAAC;AACvE,SAAO,QAAQ;AACjB;","names":["writeFileSync","join","createHash","meanComposite","createHash","runCampaign","existsSync","join"]}
@@ -1,6 +1,6 @@
1
1
  import {
2
2
  __export
3
- } from "./chunk-PZ5AY32C.js";
3
+ } from "./chunk-NSBPE2FW.js";
4
4
 
5
5
  // src/benchmarks/index.ts
6
6
  var benchmarks_exports = {};
@@ -220,4 +220,4 @@ export {
220
220
  routing_exports,
221
221
  benchmarks_exports
222
222
  };
223
- //# sourceMappingURL=chunk-6QDKWHLS.js.map
223
+ //# sourceMappingURL=chunk-MHQPVHXU.js.map
@@ -0,0 +1,17 @@
1
+ var __defProp = Object.defineProperty;
2
+ var __require = /* @__PURE__ */ ((x) => typeof require !== "undefined" ? require : typeof Proxy !== "undefined" ? new Proxy(x, {
3
+ get: (a, b) => (typeof require !== "undefined" ? require : a)[b]
4
+ }) : x)(function(x) {
5
+ if (typeof require !== "undefined") return require.apply(this, arguments);
6
+ throw Error('Dynamic require of "' + x + '" is not supported');
7
+ });
8
+ var __export = (target, all) => {
9
+ for (var name in all)
10
+ __defProp(target, name, { get: all[name], enumerable: true });
11
+ };
12
+
13
+ export {
14
+ __require,
15
+ __export
16
+ };
17
+ //# sourceMappingURL=chunk-NSBPE2FW.js.map
@@ -1,11 +1,58 @@
1
1
  import {
2
2
  confidenceInterval
3
3
  } from "./chunk-WP7SY7AI.js";
4
+ import {
5
+ __require
6
+ } from "./chunk-NSBPE2FW.js";
4
7
 
5
8
  // src/campaign/run-campaign.ts
6
9
  import { createHash } from "crypto";
7
- import { existsSync, mkdirSync, readFileSync, writeFileSync } from "fs";
8
10
  import { join } from "path";
11
+
12
+ // src/campaign/storage.ts
13
+ function fsCampaignStorage() {
14
+ const { existsSync, mkdirSync, readFileSync, writeFileSync } = __require("fs");
15
+ return {
16
+ ensureDir(dir) {
17
+ if (!existsSync(dir)) mkdirSync(dir, { recursive: true });
18
+ },
19
+ exists(path) {
20
+ return existsSync(path);
21
+ },
22
+ read(path) {
23
+ try {
24
+ return readFileSync(path, "utf8");
25
+ } catch {
26
+ return void 0;
27
+ }
28
+ },
29
+ write(path, content) {
30
+ writeFileSync(path, content);
31
+ }
32
+ };
33
+ }
34
+ function inMemoryCampaignStorage() {
35
+ const files = /* @__PURE__ */ new Map();
36
+ const dirs = /* @__PURE__ */ new Set();
37
+ return {
38
+ ensureDir(dir) {
39
+ dirs.add(dir);
40
+ },
41
+ exists(path) {
42
+ return files.has(path) || dirs.has(path);
43
+ },
44
+ read(path) {
45
+ const value = files.get(path);
46
+ if (value === void 0) return void 0;
47
+ return typeof value === "string" ? value : new TextDecoder().decode(value);
48
+ },
49
+ write(path, content) {
50
+ files.set(path, content);
51
+ }
52
+ };
53
+ }
54
+
55
+ // src/campaign/run-campaign.ts
9
56
  async function runCampaign(opts) {
10
57
  const seed = opts.seed ?? 42;
11
58
  const reps = opts.reps ?? 1;
@@ -13,7 +60,8 @@ async function runCampaign(opts) {
13
60
  const maxConcurrency = opts.maxConcurrency ?? 2;
14
61
  const now = opts.now ?? (() => /* @__PURE__ */ new Date());
15
62
  const judges = opts.judges ?? [];
16
- if (!existsSync(opts.runDir)) mkdirSync(opts.runDir, { recursive: true });
63
+ const storage = opts.storage ?? fsCampaignStorage();
64
+ storage.ensureDir(opts.runDir);
17
65
  const manifestHash = computeManifestHash({
18
66
  scenarios: opts.scenarios,
19
67
  judges,
@@ -57,7 +105,8 @@ async function runCampaign(opts) {
57
105
  manifestHash,
58
106
  resumable,
59
107
  now,
60
- buildTraceWriter: opts.buildTraceWriter ?? defaultBuildTraceWriter,
108
+ storage,
109
+ buildTraceWriter: opts.buildTraceWriter ?? defaultBuildTraceWriter(storage),
61
110
  signal: abortController.signal
62
111
  });
63
112
  cellsRef.push(result.cell);
@@ -105,16 +154,20 @@ async function runCampaign(opts) {
105
154
  };
106
155
  }
107
156
  async function executeCell(args) {
157
+ const storage = args.storage;
108
158
  const cellDir = join(args.opts.runDir, args.slot.cellId.replace(/[^a-zA-Z0-9_-]/g, "_"));
109
- if (!existsSync(cellDir)) mkdirSync(cellDir, { recursive: true });
159
+ storage.ensureDir(cellDir);
110
160
  const cachePath = join(cellDir, "cached-result.json");
111
- if (args.resumable && existsSync(cachePath)) {
112
- try {
113
- const cached = JSON.parse(readFileSync(cachePath, "utf8"));
114
- if (cached.cellId === args.slot.cellId) {
115
- return { cell: { ...cached, cached: true }, artifactsByPath: {} };
161
+ if (args.resumable) {
162
+ const raw = storage.read(cachePath);
163
+ if (raw !== void 0) {
164
+ try {
165
+ const cached = JSON.parse(raw);
166
+ if (cached.cellId === args.slot.cellId) {
167
+ return { cell: { ...cached, cached: true }, artifactsByPath: {} };
168
+ }
169
+ } catch {
116
170
  }
117
- } catch {
118
171
  }
119
172
  }
120
173
  const startMs = Date.now();
@@ -123,9 +176,8 @@ async function executeCell(args) {
123
176
  const artifacts = {
124
177
  async write(path, content) {
125
178
  const fullPath = join(cellDir, path);
126
- const dir = join(fullPath, "..");
127
- if (!existsSync(dir)) mkdirSync(dir, { recursive: true });
128
- writeFileSync(fullPath, content);
179
+ storage.ensureDir(join(fullPath, ".."));
180
+ storage.write(fullPath, content);
129
181
  artifactsByPath[`${args.slot.cellId}/${path}`] = fullPath;
130
182
  return fullPath;
131
183
  },
@@ -189,35 +241,36 @@ async function executeCell(args) {
189
241
  error: errorMessage
190
242
  };
191
243
  if (!errorMessage && args.resumable) {
192
- writeFileSync(cachePath, JSON.stringify(cell));
244
+ storage.write(cachePath, JSON.stringify(cell));
193
245
  }
194
246
  return { cell, artifactsByPath };
195
247
  }
196
248
  async function runJudgeCell(judge, input) {
197
249
  return judge.score(input);
198
250
  }
199
- function defaultBuildTraceWriter(cellId, dir) {
200
- const spans = [];
201
- return {
202
- span(name, attributes) {
203
- const startMs = Date.now();
204
- const record = { name, cellId, startMs, ...attributes ?? {} };
205
- const finish = {
206
- end(endAttrs) {
207
- record.durationMs = Date.now() - startMs;
208
- if (endAttrs) Object.assign(record, endAttrs);
209
- spans.push(record);
210
- },
211
- setAttribute(key, value) {
212
- record[key] = value;
213
- }
214
- };
215
- return finish;
216
- },
217
- async flush() {
218
- const path = join(dir, "spans.jsonl");
219
- writeFileSync(path, spans.map((s) => JSON.stringify(s)).join("\n"));
220
- }
251
+ function defaultBuildTraceWriter(storage) {
252
+ return (cellId, dir) => {
253
+ const spans = [];
254
+ return {
255
+ span(name, attributes) {
256
+ const startMs = Date.now();
257
+ const record = { name, cellId, startMs, ...attributes ?? {} };
258
+ const finish = {
259
+ end(endAttrs) {
260
+ record.durationMs = Date.now() - startMs;
261
+ if (endAttrs) Object.assign(record, endAttrs);
262
+ spans.push(record);
263
+ },
264
+ setAttribute(key, value) {
265
+ record[key] = value;
266
+ }
267
+ };
268
+ return finish;
269
+ },
270
+ async flush() {
271
+ storage.write(join(dir, "spans.jsonl"), spans.map((s) => JSON.stringify(s)).join("\n"));
272
+ }
273
+ };
221
274
  };
222
275
  }
223
276
  function skippedCell(slot, reason) {
@@ -300,6 +353,8 @@ function aggregate(samples, seed) {
300
353
  }
301
354
 
302
355
  export {
356
+ fsCampaignStorage,
357
+ inMemoryCampaignStorage,
303
358
  runCampaign
304
359
  };
305
- //# sourceMappingURL=chunk-YNMCYUWT.js.map
360
+ //# sourceMappingURL=chunk-RXK7FXLV.js.map
@@ -0,0 +1 @@
1
+ {"version":3,"sources":["../src/campaign/run-campaign.ts","../src/campaign/storage.ts"],"sourcesContent":["/**\n * @experimental\n *\n * `runCampaign` — Pass A substrate primitive. ONE function that orchestrates\n * scenarios → dispatch → artifacts → judges → aggregates, with full\n * reproducibility (seed + manifest hash), cell-level resumability, bootstrap\n * CIs, and the `LabeledScenarioStore` capture flywheel.\n *\n * Improvement loops (optimizer / gate / autoOnPromote) ride on top of this\n * primitive but live in `presets/run-improvement-loop.ts`. This file keeps\n * the core orchestrator minimal — Phase 1 of the Pass A track.\n */\n\nimport { createHash } from 'node:crypto'\nimport { join } from 'node:path'\nimport { confidenceInterval } from '../statistics'\nimport { type CampaignStorage, fsCampaignStorage } from './storage'\nimport type {\n CampaignAggregates,\n CampaignArtifactWriter,\n CampaignCellResult,\n CampaignCostMeter,\n CampaignResult,\n CampaignTraceWriter,\n DispatchContext,\n DispatchFn,\n JudgeAggregate,\n JudgeConfig,\n JudgeScore,\n LabeledScenarioStore,\n Scenario,\n ScenarioAggregate,\n TraceSpan,\n} from './types'\n\nexport interface RunCampaignOptions<TScenario extends Scenario, TArtifact> {\n scenarios: TScenario[]\n dispatch: DispatchFn<TScenario, TArtifact>\n judges?: JudgeConfig<TArtifact, TScenario>[]\n /** Required for reproducibility. Default 42. */\n seed?: number\n /** Per-scenario replicates for CI bands. Default 1; raise to 5+ for\n * bootstrap-tight intervals on critical eval. */\n reps?: number\n /** When true (default), completed cells are cached by\n * (manifestHash, scenarioId, rep, generation). Re-runs skip cached cells. */\n resumable?: boolean\n /** Optional store — when present, every artifact + judge score is captured\n * with the configured `captureSource`. Capture is default ON; pass `'off'`\n * to disable. */\n labeledStore?: LabeledScenarioStore | 'off'\n captureSource?: 'production-trace' | 'eval-run' | 'manual' | 'red-team' | 'synthetic'\n captureSourceVersionHash?: string\n /** Wall-clock cost cap across all cells. Cells beyond ceiling are skipped. */\n costCeiling?: number\n /** Max concurrent cells. Default 2. */\n maxConcurrency?: number\n /** Required: where artifacts + traces land. */\n runDir: string\n /** Tracing posture. Default is the substrate's `FileSystemTraceStore` rooted\n * at `<runDir>/traces/`. `'off'` disables capture entirely — substrate\n * refuses this when the caller wires `autoOnPromote !== 'none'`. */\n tracing?: 'on' | 'off'\n /** Test seam — override the wall clock for deterministic tests. */\n now?: () => Date\n /** Test seam — override per-cell trace writer factory. */\n buildTraceWriter?: (cellId: string, dir: string) => CampaignTraceWriter\n /** Storage backend for run/cell dirs, the resumability cache, artifacts,\n * and trace spans. Default: the Node filesystem (`fsCampaignStorage`).\n * Pass `inMemoryCampaignStorage()` to run in a filesystem-less runtime\n * (Cloudflare Workers, Deno, edge) — the `CampaignResult` is still\n * produced; artifacts/traces just aren't persisted to disk. */\n storage?: CampaignStorage\n}\n\nexport async function runCampaign<TScenario extends Scenario, TArtifact>(\n opts: RunCampaignOptions<TScenario, TArtifact>,\n): Promise<CampaignResult<TArtifact, TScenario>> {\n const seed = opts.seed ?? 42\n const reps = opts.reps ?? 1\n const resumable = opts.resumable ?? true\n const maxConcurrency = opts.maxConcurrency ?? 2\n const now = opts.now ?? (() => new Date())\n const judges = opts.judges ?? []\n const storage = opts.storage ?? fsCampaignStorage()\n\n storage.ensureDir(opts.runDir)\n\n const manifestHash = computeManifestHash({\n scenarios: opts.scenarios,\n judges: judges as unknown as JudgeConfig<unknown>[],\n dispatchRef: opts.dispatch.name || 'anonymous',\n seed,\n reps,\n })\n\n const startedAt = now()\n const cells: CampaignCellResult<TArtifact>[] = []\n const artifactsByPath: Record<string, string> = {}\n\n // Build the cell schedule (scenario × rep).\n const schedule: Array<{ scenario: TScenario; rep: number; cellId: string; cellSeed: number }> = []\n let cellIndex = 0\n for (const scenario of opts.scenarios) {\n for (let rep = 0; rep < reps; rep++) {\n const cellId = `${scenario.id}:${rep}`\n const cellSeed = seed + cellIndex\n schedule.push({ scenario, rep, cellId, cellSeed })\n cellIndex += 1\n }\n }\n\n // Concurrency-limited execution.\n let totalCostUsd = 0\n let costCeilingReached = false\n const abortController = new AbortController()\n // Concurrency lanes that drain the cell schedule. Named \"lanes\" — not\n // \"workers\" — to avoid clashing with the taxonomy's worker (= the agent\n // harness in a sandbox, invoked behind `dispatch`). See loop-taxonomy.md.\n const lanes: Promise<void>[] = []\n let nextIdx = 0\n const cellsRef = cells\n\n for (let i = 0; i < maxConcurrency; i++) {\n lanes.push(\n (async () => {\n while (true) {\n const myIdx = nextIdx++\n if (myIdx >= schedule.length) return\n const slot = schedule[myIdx]!\n if (costCeilingReached) {\n cellsRef.push(skippedCell(slot, 'cost_ceiling_reached'))\n continue\n }\n const result = await executeCell({\n slot,\n opts,\n manifestHash,\n resumable,\n now,\n storage,\n buildTraceWriter: opts.buildTraceWriter ?? defaultBuildTraceWriter(storage),\n signal: abortController.signal,\n })\n cellsRef.push(result.cell)\n totalCostUsd += result.cell.costUsd\n Object.assign(artifactsByPath, result.artifactsByPath)\n if (opts.costCeiling !== undefined && totalCostUsd >= opts.costCeiling) {\n costCeilingReached = true\n }\n // Capture into LabeledScenarioStore unless explicitly disabled.\n if (opts.labeledStore && opts.labeledStore !== 'off' && !result.cell.error) {\n await captureToStore({\n store: opts.labeledStore,\n cell: result.cell,\n scenario: slot.scenario,\n opts,\n now,\n }).catch((err) => {\n // Capture failures are non-fatal — log but don't crash the campaign.\n // (Trace would normally land here.)\n console.warn(\n `[runCampaign] capture failed for ${result.cell.cellId}: ${err instanceof Error ? err.message : String(err)}`,\n )\n })\n }\n }\n })(),\n )\n }\n await Promise.all(lanes)\n\n const endedAt = now()\n cellsRef.sort((a, b) => a.cellId.localeCompare(b.cellId))\n\n const aggregates = computeAggregates(\n cellsRef,\n judges as unknown as JudgeConfig<TArtifact>[],\n seed,\n )\n\n return {\n manifestHash,\n seed,\n startedAt: startedAt.toISOString(),\n endedAt: endedAt.toISOString(),\n durationMs: endedAt.getTime() - startedAt.getTime(),\n cells: cellsRef,\n aggregates,\n runDir: opts.runDir,\n artifactsByPath,\n scenarios: opts.scenarios.map((s) => ({ id: s.id, kind: s.kind })),\n }\n}\n\n// ── Internals ─────────────────────────────────────────────────────────\n\ninterface ExecuteCellArgs<TScenario extends Scenario, TArtifact> {\n slot: { scenario: TScenario; rep: number; cellId: string; cellSeed: number }\n opts: RunCampaignOptions<TScenario, TArtifact>\n manifestHash: string\n resumable: boolean\n now: () => Date\n storage: CampaignStorage\n buildTraceWriter: (cellId: string, dir: string) => CampaignTraceWriter\n signal: AbortSignal\n}\n\nasync function executeCell<TScenario extends Scenario, TArtifact>(\n args: ExecuteCellArgs<TScenario, TArtifact>,\n): Promise<{ cell: CampaignCellResult<TArtifact>; artifactsByPath: Record<string, string> }> {\n const storage = args.storage\n const cellDir = join(args.opts.runDir, args.slot.cellId.replace(/[^a-zA-Z0-9_-]/g, '_'))\n storage.ensureDir(cellDir)\n\n // Resumability: cache key = (manifestHash, scenarioId, rep)\n const cachePath = join(cellDir, 'cached-result.json')\n if (args.resumable) {\n const raw = storage.read(cachePath)\n if (raw !== undefined) {\n try {\n const cached = JSON.parse(raw) as CampaignCellResult<TArtifact>\n if (cached.cellId === args.slot.cellId) {\n return { cell: { ...cached, cached: true }, artifactsByPath: {} }\n }\n } catch {\n // Corrupt cache — fall through to re-run.\n }\n }\n }\n\n const startMs = Date.now()\n const trace = args.buildTraceWriter(args.slot.cellId, cellDir)\n const artifactsByPath: Record<string, string> = {}\n const artifacts: CampaignArtifactWriter = {\n async write(path, content) {\n const fullPath = join(cellDir, path)\n storage.ensureDir(join(fullPath, '..'))\n storage.write(fullPath, content)\n artifactsByPath[`${args.slot.cellId}/${path}`] = fullPath\n return fullPath\n },\n async writeJson(path, value) {\n return artifacts.write(path, JSON.stringify(value, null, 2))\n },\n }\n let costSoFar = 0\n const cost: CampaignCostMeter = {\n observe(amount, source) {\n costSoFar += amount\n trace.span(`cost.${source}`, { amountUsd: amount }).end()\n },\n current() {\n return costSoFar\n },\n }\n\n const ctx: DispatchContext = {\n cellId: args.slot.cellId,\n rep: args.slot.rep,\n seed: args.slot.cellSeed,\n signal: args.signal,\n trace,\n artifacts,\n cost,\n }\n\n let artifact: TArtifact | undefined\n let errorMessage: string | undefined\n try {\n artifact = await args.opts.dispatch(args.slot.scenario, ctx)\n } catch (err) {\n errorMessage = err instanceof Error ? err.message : String(err)\n }\n\n // Run judges (only if we have an artifact). A judge that throws invalidates\n // the cell — recorded as `error`, NOT folded into a fake composite:0 (a fake\n // zero is indistinguishable from a real zero and poisons every aggregate).\n const judgeScores: Record<string, JudgeScore> = {}\n if (artifact !== undefined) {\n for (const judge of args.opts.judges ?? []) {\n if (judge.appliesTo && !judge.appliesTo(args.slot.scenario)) continue\n try {\n judgeScores[judge.name] = await runJudgeCell(judge, {\n artifact,\n scenario: args.slot.scenario,\n signal: args.signal,\n })\n } catch (err) {\n errorMessage = `judge '${judge.name}' failed: ${err instanceof Error ? err.message : String(err)}`\n break\n }\n }\n }\n\n await trace.flush()\n\n const cell: CampaignCellResult<TArtifact> = {\n cellId: args.slot.cellId,\n scenarioId: args.slot.scenario.id,\n rep: args.slot.rep,\n artifact: (artifact ?? null) as TArtifact,\n judgeScores,\n costUsd: costSoFar,\n durationMs: Date.now() - startMs,\n seed: args.slot.cellSeed,\n cached: false,\n error: errorMessage,\n }\n\n if (!errorMessage && args.resumable) {\n storage.write(cachePath, JSON.stringify(cell))\n }\n\n return { cell, artifactsByPath }\n}\n\nasync function runJudgeCell<TArtifact, TScenario extends Scenario>(\n judge: JudgeConfig<TArtifact, TScenario>,\n input: { artifact: TArtifact; scenario: TScenario; signal: AbortSignal },\n): Promise<JudgeScore> {\n return judge.score(input)\n}\n\nfunction defaultBuildTraceWriter(\n storage: CampaignStorage,\n): (cellId: string, dir: string) => CampaignTraceWriter {\n return (cellId, dir) => {\n const spans: Array<Record<string, unknown>> = []\n return {\n span(name, attributes) {\n const startMs = Date.now()\n const record: Record<string, unknown> = { name, cellId, startMs, ...(attributes ?? {}) }\n const finish: TraceSpan = {\n end(endAttrs) {\n record.durationMs = Date.now() - startMs\n if (endAttrs) Object.assign(record, endAttrs)\n spans.push(record)\n },\n setAttribute(key, value) {\n record[key] = value\n },\n }\n return finish\n },\n async flush() {\n storage.write(join(dir, 'spans.jsonl'), spans.map((s) => JSON.stringify(s)).join('\\n'))\n },\n }\n }\n}\n\nfunction skippedCell<TScenario extends Scenario, TArtifact>(\n slot: { scenario: TScenario; rep: number; cellId: string; cellSeed: number },\n reason: string,\n): CampaignCellResult<TArtifact> {\n return {\n cellId: slot.cellId,\n scenarioId: slot.scenario.id,\n rep: slot.rep,\n artifact: null as unknown as TArtifact,\n judgeScores: {},\n costUsd: 0,\n durationMs: 0,\n seed: slot.cellSeed,\n cached: false,\n error: `skipped: ${reason}`,\n }\n}\n\ninterface CaptureArgs<TScenario extends Scenario, TArtifact> {\n store: LabeledScenarioStore\n cell: CampaignCellResult<TArtifact>\n scenario: TScenario\n opts: RunCampaignOptions<TScenario, TArtifact>\n now: () => Date\n}\n\nasync function captureToStore<TScenario extends Scenario, TArtifact>(\n args: CaptureArgs<TScenario, TArtifact>,\n): Promise<void> {\n await args.store.observe({\n scenario: args.scenario,\n artifact: args.cell.artifact,\n judgeScores: args.cell.judgeScores,\n source: args.opts.captureSource ?? 'eval-run',\n sourceVersionHash: args.opts.captureSourceVersionHash ?? 'unknown',\n capturedAt: args.now().toISOString(),\n redactionStatus: 'raw',\n })\n}\n\n// ── Aggregates + manifest hash ────────────────────────────────────────\n\nfunction computeManifestHash(input: {\n scenarios: Scenario[]\n judges: JudgeConfig<unknown>[]\n dispatchRef: string\n seed: number\n reps: number\n}): string {\n const canonical = {\n scenarios: input.scenarios.map((s) => ({ id: s.id, kind: s.kind })),\n judges: input.judges.map((j) => ({ name: j.name, dims: j.dimensions.map((d) => d.key) })),\n dispatch: input.dispatchRef,\n seed: input.seed,\n reps: input.reps,\n }\n return createHash('sha256').update(JSON.stringify(canonical)).digest('hex')\n}\n\nfunction computeAggregates<TArtifact>(\n cells: CampaignCellResult<TArtifact>[],\n judges: JudgeConfig<TArtifact>[],\n seed: number,\n): CampaignAggregates {\n const byJudge: Record<string, JudgeAggregate> = {}\n for (const judge of judges) {\n const scores: number[] = []\n for (const cell of cells) {\n const s = cell.judgeScores[judge.name]\n if (s !== undefined) scores.push(s.composite)\n }\n byJudge[judge.name] = aggregate(scores, seed)\n }\n const byScenario: Record<string, ScenarioAggregate> = {}\n const scenarioGroups = new Map<string, number[]>()\n for (const cell of cells) {\n const composites = Object.values(cell.judgeScores).map((s) => s.composite)\n if (composites.length === 0) continue\n const mean = composites.reduce((a, b) => a + b, 0) / composites.length\n const arr = scenarioGroups.get(cell.scenarioId) ?? []\n arr.push(mean)\n scenarioGroups.set(cell.scenarioId, arr)\n }\n for (const [scenarioId, samples] of scenarioGroups) {\n const ag = aggregate(samples, seed)\n byScenario[scenarioId] = { meanComposite: ag.mean, ci95: ag.ci95, n: ag.n }\n }\n return {\n byJudge,\n byScenario,\n totalCostUsd: cells.reduce((a, c) => a + c.costUsd, 0),\n cellsExecuted: cells.filter((c) => !c.error).length,\n cellsSkipped: cells.filter((c) => c.error?.startsWith('skipped:')).length,\n cellsCached: cells.filter((c) => c.cached).length,\n cellsFailed: cells.filter((c) => c.error && !c.error.startsWith('skipped:')).length,\n }\n}\n\n// Percentile bootstrap CI95 via seeded resampling. Deterministic for a given\n// seed — same campaign re-run produces identical CI bands. Falls back to\n// degenerate intervals at n<=1 (the bootstrap is undefined there).\nfunction aggregate(samples: number[], seed: number): JudgeAggregate {\n const n = samples.length\n if (n === 0) return { mean: 0, stdev: 0, ci95: [0, 0], n: 0 }\n const mean = samples.reduce((a, b) => a + b, 0) / n\n const variance = samples.reduce((a, b) => a + (b - mean) ** 2, 0) / Math.max(1, n - 1)\n const stdev = Math.sqrt(variance)\n const ci = confidenceInterval(samples, 0.95, { seed, resamples: 1000 })\n return { mean, stdev, ci95: [ci.lower, ci.upper], n }\n}\n","/**\n * @experimental\n *\n * `CampaignStorage` — the filesystem seam `runCampaign` writes through\n * (run/cell dirs, the resumability cache, per-cell artifacts, trace spans).\n *\n * The default (`fsCampaignStorage`) is the Node filesystem — identical\n * behavior to the inline `node:fs` calls it replaces, so existing CLI\n * consumers are unaffected. `inMemoryCampaignStorage` keeps everything in a\n * `Map`, so the substrate runs in environments WITHOUT a filesystem\n * (Cloudflare Workers, Deno Deploy, other edge runtimes) — the campaign\n * still produces its `CampaignResult` (cells + aggregates) in memory;\n * artifacts/traces simply aren't persisted to disk.\n *\n * Paths are opaque keys to the in-memory adapter — it does not parse them,\n * so the same `join(...)`-built paths work unchanged across both adapters.\n */\nexport interface CampaignStorage {\n /** Ensure a directory exists (recursive). No-op for in-memory. */\n ensureDir(dir: string): void\n /** Does this path exist (as a written file or an ensured dir)? */\n exists(path: string): boolean\n /** Read a UTF-8 file; `undefined` when missing or unreadable. */\n read(path: string): string | undefined\n /** Write a file (string or bytes). Parent dir is assumed ensured. */\n write(path: string, content: string | Uint8Array): void\n}\n\n/** Node-filesystem storage — the default. Lazily requires `node:fs` so the\n * module imports cleanly in non-Node runtimes (where the caller passes\n * `inMemoryCampaignStorage` instead and never constructs this). */\nexport function fsCampaignStorage(): CampaignStorage {\n const { existsSync, mkdirSync, readFileSync, writeFileSync } =\n require('node:fs') as typeof import('node:fs')\n return {\n ensureDir(dir) {\n if (!existsSync(dir)) mkdirSync(dir, { recursive: true })\n },\n exists(path) {\n return existsSync(path)\n },\n read(path) {\n try {\n return readFileSync(path, 'utf8')\n } catch {\n return undefined\n }\n },\n write(path, content) {\n writeFileSync(path, content as Uint8Array)\n },\n }\n}\n\n/** In-memory storage for filesystem-less runtimes. Artifacts + trace spans\n * live in a `Map` for the duration of the run; the `CampaignResult` is\n * fully populated, but nothing is persisted to disk. */\nexport function inMemoryCampaignStorage(): CampaignStorage {\n const files = new Map<string, string | Uint8Array>()\n const dirs = new Set<string>()\n return {\n ensureDir(dir) {\n dirs.add(dir)\n },\n exists(path) {\n return files.has(path) || dirs.has(path)\n },\n read(path) {\n const value = files.get(path)\n if (value === undefined) return undefined\n return typeof value === 'string' ? value : new TextDecoder().decode(value)\n },\n write(path, content) {\n files.set(path, content)\n },\n }\n}\n"],"mappings":";;;;;;;;AAaA,SAAS,kBAAkB;AAC3B,SAAS,YAAY;;;ACiBd,SAAS,oBAAqC;AACnD,QAAM,EAAE,YAAY,WAAW,cAAc,cAAc,IACzD,UAAQ,IAAS;AACnB,SAAO;AAAA,IACL,UAAU,KAAK;AACb,UAAI,CAAC,WAAW,GAAG,EAAG,WAAU,KAAK,EAAE,WAAW,KAAK,CAAC;AAAA,IAC1D;AAAA,IACA,OAAO,MAAM;AACX,aAAO,WAAW,IAAI;AAAA,IACxB;AAAA,IACA,KAAK,MAAM;AACT,UAAI;AACF,eAAO,aAAa,MAAM,MAAM;AAAA,MAClC,QAAQ;AACN,eAAO;AAAA,MACT;AAAA,IACF;AAAA,IACA,MAAM,MAAM,SAAS;AACnB,oBAAc,MAAM,OAAqB;AAAA,IAC3C;AAAA,EACF;AACF;AAKO,SAAS,0BAA2C;AACzD,QAAM,QAAQ,oBAAI,IAAiC;AACnD,QAAM,OAAO,oBAAI,IAAY;AAC7B,SAAO;AAAA,IACL,UAAU,KAAK;AACb,WAAK,IAAI,GAAG;AAAA,IACd;AAAA,IACA,OAAO,MAAM;AACX,aAAO,MAAM,IAAI,IAAI,KAAK,KAAK,IAAI,IAAI;AAAA,IACzC;AAAA,IACA,KAAK,MAAM;AACT,YAAM,QAAQ,MAAM,IAAI,IAAI;AAC5B,UAAI,UAAU,OAAW,QAAO;AAChC,aAAO,OAAO,UAAU,WAAW,QAAQ,IAAI,YAAY,EAAE,OAAO,KAAK;AAAA,IAC3E;AAAA,IACA,MAAM,MAAM,SAAS;AACnB,YAAM,IAAI,MAAM,OAAO;AAAA,IACzB;AAAA,EACF;AACF;;;ADDA,eAAsB,YACpB,MAC+C;AAC/C,QAAM,OAAO,KAAK,QAAQ;AAC1B,QAAM,OAAO,KAAK,QAAQ;AAC1B,QAAM,YAAY,KAAK,aAAa;AACpC,QAAM,iBAAiB,KAAK,kBAAkB;AAC9C,QAAM,MAAM,KAAK,QAAQ,MAAM,oBAAI,KAAK;AACxC,QAAM,SAAS,KAAK,UAAU,CAAC;AAC/B,QAAM,UAAU,KAAK,WAAW,kBAAkB;AAElD,UAAQ,UAAU,KAAK,MAAM;AAE7B,QAAM,eAAe,oBAAoB;AAAA,IACvC,WAAW,KAAK;AAAA,IAChB;AAAA,IACA,aAAa,KAAK,SAAS,QAAQ;AAAA,IACnC;AAAA,IACA;AAAA,EACF,CAAC;AAED,QAAM,YAAY,IAAI;AACtB,QAAM,QAAyC,CAAC;AAChD,QAAM,kBAA0C,CAAC;AAGjD,QAAM,WAA0F,CAAC;AACjG,MAAI,YAAY;AAChB,aAAW,YAAY,KAAK,WAAW;AACrC,aAAS,MAAM,GAAG,MAAM,MAAM,OAAO;AACnC,YAAM,SAAS,GAAG,SAAS,EAAE,IAAI,GAAG;AACpC,YAAM,WAAW,OAAO;AACxB,eAAS,KAAK,EAAE,UAAU,KAAK,QAAQ,SAAS,CAAC;AACjD,mBAAa;AAAA,IACf;AAAA,EACF;AAGA,MAAI,eAAe;AACnB,MAAI,qBAAqB;AACzB,QAAM,kBAAkB,IAAI,gBAAgB;AAI5C,QAAM,QAAyB,CAAC;AAChC,MAAI,UAAU;AACd,QAAM,WAAW;AAEjB,WAAS,IAAI,GAAG,IAAI,gBAAgB,KAAK;AACvC,UAAM;AAAA,OACH,YAAY;AACX,eAAO,MAAM;AACX,gBAAM,QAAQ;AACd,cAAI,SAAS,SAAS,OAAQ;AAC9B,gBAAM,OAAO,SAAS,KAAK;AAC3B,cAAI,oBAAoB;AACtB,qBAAS,KAAK,YAAY,MAAM,sBAAsB,CAAC;AACvD;AAAA,UACF;AACA,gBAAM,SAAS,MAAM,YAAY;AAAA,YAC/B;AAAA,YACA;AAAA,YACA;AAAA,YACA;AAAA,YACA;AAAA,YACA;AAAA,YACA,kBAAkB,KAAK,oBAAoB,wBAAwB,OAAO;AAAA,YAC1E,QAAQ,gBAAgB;AAAA,UAC1B,CAAC;AACD,mBAAS,KAAK,OAAO,IAAI;AACzB,0BAAgB,OAAO,KAAK;AAC5B,iBAAO,OAAO,iBAAiB,OAAO,eAAe;AACrD,cAAI,KAAK,gBAAgB,UAAa,gBAAgB,KAAK,aAAa;AACtE,iCAAqB;AAAA,UACvB;AAEA,cAAI,KAAK,gBAAgB,KAAK,iBAAiB,SAAS,CAAC,OAAO,KAAK,OAAO;AAC1E,kBAAM,eAAe;AAAA,cACnB,OAAO,KAAK;AAAA,cACZ,MAAM,OAAO;AAAA,cACb,UAAU,KAAK;AAAA,cACf;AAAA,cACA;AAAA,YACF,CAAC,EAAE,MAAM,CAAC,QAAQ;AAGhB,sBAAQ;AAAA,gBACN,oCAAoC,OAAO,KAAK,MAAM,KAAK,eAAe,QAAQ,IAAI,UAAU,OAAO,GAAG,CAAC;AAAA,cAC7G;AAAA,YACF,CAAC;AAAA,UACH;AAAA,QACF;AAAA,MACF,GAAG;AAAA,IACL;AAAA,EACF;AACA,QAAM,QAAQ,IAAI,KAAK;AAEvB,QAAM,UAAU,IAAI;AACpB,WAAS,KAAK,CAAC,GAAG,MAAM,EAAE,OAAO,cAAc,EAAE,MAAM,CAAC;AAExD,QAAM,aAAa;AAAA,IACjB;AAAA,IACA;AAAA,IACA;AAAA,EACF;AAEA,SAAO;AAAA,IACL;AAAA,IACA;AAAA,IACA,WAAW,UAAU,YAAY;AAAA,IACjC,SAAS,QAAQ,YAAY;AAAA,IAC7B,YAAY,QAAQ,QAAQ,IAAI,UAAU,QAAQ;AAAA,IAClD,OAAO;AAAA,IACP;AAAA,IACA,QAAQ,KAAK;AAAA,IACb;AAAA,IACA,WAAW,KAAK,UAAU,IAAI,CAAC,OAAO,EAAE,IAAI,EAAE,IAAI,MAAM,EAAE,KAAK,EAAE;AAAA,EACnE;AACF;AAeA,eAAe,YACb,MAC2F;AAC3F,QAAM,UAAU,KAAK;AACrB,QAAM,UAAU,KAAK,KAAK,KAAK,QAAQ,KAAK,KAAK,OAAO,QAAQ,mBAAmB,GAAG,CAAC;AACvF,UAAQ,UAAU,OAAO;AAGzB,QAAM,YAAY,KAAK,SAAS,oBAAoB;AACpD,MAAI,KAAK,WAAW;AAClB,UAAM,MAAM,QAAQ,KAAK,SAAS;AAClC,QAAI,QAAQ,QAAW;AACrB,UAAI;AACF,cAAM,SAAS,KAAK,MAAM,GAAG;AAC7B,YAAI,OAAO,WAAW,KAAK,KAAK,QAAQ;AACtC,iBAAO,EAAE,MAAM,EAAE,GAAG,QAAQ,QAAQ,KAAK,GAAG,iBAAiB,CAAC,EAAE;AAAA,QAClE;AAAA,MACF,QAAQ;AAAA,MAER;AAAA,IACF;AAAA,EACF;AAEA,QAAM,UAAU,KAAK,IAAI;AACzB,QAAM,QAAQ,KAAK,iBAAiB,KAAK,KAAK,QAAQ,OAAO;AAC7D,QAAM,kBAA0C,CAAC;AACjD,QAAM,YAAoC;AAAA,IACxC,MAAM,MAAM,MAAM,SAAS;AACzB,YAAM,WAAW,KAAK,SAAS,IAAI;AACnC,cAAQ,UAAU,KAAK,UAAU,IAAI,CAAC;AACtC,cAAQ,MAAM,UAAU,OAAO;AAC/B,sBAAgB,GAAG,KAAK,KAAK,MAAM,IAAI,IAAI,EAAE,IAAI;AACjD,aAAO;AAAA,IACT;AAAA,IACA,MAAM,UAAU,MAAM,OAAO;AAC3B,aAAO,UAAU,MAAM,MAAM,KAAK,UAAU,OAAO,MAAM,CAAC,CAAC;AAAA,IAC7D;AAAA,EACF;AACA,MAAI,YAAY;AAChB,QAAM,OAA0B;AAAA,IAC9B,QAAQ,QAAQ,QAAQ;AACtB,mBAAa;AACb,YAAM,KAAK,QAAQ,MAAM,IAAI,EAAE,WAAW,OAAO,CAAC,EAAE,IAAI;AAAA,IAC1D;AAAA,IACA,UAAU;AACR,aAAO;AAAA,IACT;AAAA,EACF;AAEA,QAAM,MAAuB;AAAA,IAC3B,QAAQ,KAAK,KAAK;AAAA,IAClB,KAAK,KAAK,KAAK;AAAA,IACf,MAAM,KAAK,KAAK;AAAA,IAChB,QAAQ,KAAK;AAAA,IACb;AAAA,IACA;AAAA,IACA;AAAA,EACF;AAEA,MAAI;AACJ,MAAI;AACJ,MAAI;AACF,eAAW,MAAM,KAAK,KAAK,SAAS,KAAK,KAAK,UAAU,GAAG;AAAA,EAC7D,SAAS,KAAK;AACZ,mBAAe,eAAe,QAAQ,IAAI,UAAU,OAAO,GAAG;AAAA,EAChE;AAKA,QAAM,cAA0C,CAAC;AACjD,MAAI,aAAa,QAAW;AAC1B,eAAW,SAAS,KAAK,KAAK,UAAU,CAAC,GAAG;AAC1C,UAAI,MAAM,aAAa,CAAC,MAAM,UAAU,KAAK,KAAK,QAAQ,EAAG;AAC7D,UAAI;AACF,oBAAY,MAAM,IAAI,IAAI,MAAM,aAAa,OAAO;AAAA,UAClD;AAAA,UACA,UAAU,KAAK,KAAK;AAAA,UACpB,QAAQ,KAAK;AAAA,QACf,CAAC;AAAA,MACH,SAAS,KAAK;AACZ,uBAAe,UAAU,MAAM,IAAI,aAAa,eAAe,QAAQ,IAAI,UAAU,OAAO,GAAG,CAAC;AAChG;AAAA,MACF;AAAA,IACF;AAAA,EACF;AAEA,QAAM,MAAM,MAAM;AAElB,QAAM,OAAsC;AAAA,IAC1C,QAAQ,KAAK,KAAK;AAAA,IAClB,YAAY,KAAK,KAAK,SAAS;AAAA,IAC/B,KAAK,KAAK,KAAK;AAAA,IACf,UAAW,YAAY;AAAA,IACvB;AAAA,IACA,SAAS;AAAA,IACT,YAAY,KAAK,IAAI,IAAI;AAAA,IACzB,MAAM,KAAK,KAAK;AAAA,IAChB,QAAQ;AAAA,IACR,OAAO;AAAA,EACT;AAEA,MAAI,CAAC,gBAAgB,KAAK,WAAW;AACnC,YAAQ,MAAM,WAAW,KAAK,UAAU,IAAI,CAAC;AAAA,EAC/C;AAEA,SAAO,EAAE,MAAM,gBAAgB;AACjC;AAEA,eAAe,aACb,OACA,OACqB;AACrB,SAAO,MAAM,MAAM,KAAK;AAC1B;AAEA,SAAS,wBACP,SACsD;AACtD,SAAO,CAAC,QAAQ,QAAQ;AACtB,UAAM,QAAwC,CAAC;AAC/C,WAAO;AAAA,MACL,KAAK,MAAM,YAAY;AACrB,cAAM,UAAU,KAAK,IAAI;AACzB,cAAM,SAAkC,EAAE,MAAM,QAAQ,SAAS,GAAI,cAAc,CAAC,EAAG;AACvF,cAAM,SAAoB;AAAA,UACxB,IAAI,UAAU;AACZ,mBAAO,aAAa,KAAK,IAAI,IAAI;AACjC,gBAAI,SAAU,QAAO,OAAO,QAAQ,QAAQ;AAC5C,kBAAM,KAAK,MAAM;AAAA,UACnB;AAAA,UACA,aAAa,KAAK,OAAO;AACvB,mBAAO,GAAG,IAAI;AAAA,UAChB;AAAA,QACF;AACA,eAAO;AAAA,MACT;AAAA,MACA,MAAM,QAAQ;AACZ,gBAAQ,MAAM,KAAK,KAAK,aAAa,GAAG,MAAM,IAAI,CAAC,MAAM,KAAK,UAAU,CAAC,CAAC,EAAE,KAAK,IAAI,CAAC;AAAA,MACxF;AAAA,IACF;AAAA,EACF;AACF;AAEA,SAAS,YACP,MACA,QAC+B;AAC/B,SAAO;AAAA,IACL,QAAQ,KAAK;AAAA,IACb,YAAY,KAAK,SAAS;AAAA,IAC1B,KAAK,KAAK;AAAA,IACV,UAAU;AAAA,IACV,aAAa,CAAC;AAAA,IACd,SAAS;AAAA,IACT,YAAY;AAAA,IACZ,MAAM,KAAK;AAAA,IACX,QAAQ;AAAA,IACR,OAAO,YAAY,MAAM;AAAA,EAC3B;AACF;AAUA,eAAe,eACb,MACe;AACf,QAAM,KAAK,MAAM,QAAQ;AAAA,IACvB,UAAU,KAAK;AAAA,IACf,UAAU,KAAK,KAAK;AAAA,IACpB,aAAa,KAAK,KAAK;AAAA,IACvB,QAAQ,KAAK,KAAK,iBAAiB;AAAA,IACnC,mBAAmB,KAAK,KAAK,4BAA4B;AAAA,IACzD,YAAY,KAAK,IAAI,EAAE,YAAY;AAAA,IACnC,iBAAiB;AAAA,EACnB,CAAC;AACH;AAIA,SAAS,oBAAoB,OAMlB;AACT,QAAM,YAAY;AAAA,IAChB,WAAW,MAAM,UAAU,IAAI,CAAC,OAAO,EAAE,IAAI,EAAE,IAAI,MAAM,EAAE,KAAK,EAAE;AAAA,IAClE,QAAQ,MAAM,OAAO,IAAI,CAAC,OAAO,EAAE,MAAM,EAAE,MAAM,MAAM,EAAE,WAAW,IAAI,CAAC,MAAM,EAAE,GAAG,EAAE,EAAE;AAAA,IACxF,UAAU,MAAM;AAAA,IAChB,MAAM,MAAM;AAAA,IACZ,MAAM,MAAM;AAAA,EACd;AACA,SAAO,WAAW,QAAQ,EAAE,OAAO,KAAK,UAAU,SAAS,CAAC,EAAE,OAAO,KAAK;AAC5E;AAEA,SAAS,kBACP,OACA,QACA,MACoB;AACpB,QAAM,UAA0C,CAAC;AACjD,aAAW,SAAS,QAAQ;AAC1B,UAAM,SAAmB,CAAC;AAC1B,eAAW,QAAQ,OAAO;AACxB,YAAM,IAAI,KAAK,YAAY,MAAM,IAAI;AACrC,UAAI,MAAM,OAAW,QAAO,KAAK,EAAE,SAAS;AAAA,IAC9C;AACA,YAAQ,MAAM,IAAI,IAAI,UAAU,QAAQ,IAAI;AAAA,EAC9C;AACA,QAAM,aAAgD,CAAC;AACvD,QAAM,iBAAiB,oBAAI,IAAsB;AACjD,aAAW,QAAQ,OAAO;AACxB,UAAM,aAAa,OAAO,OAAO,KAAK,WAAW,EAAE,IAAI,CAAC,MAAM,EAAE,SAAS;AACzE,QAAI,WAAW,WAAW,EAAG;AAC7B,UAAM,OAAO,WAAW,OAAO,CAAC,GAAG,MAAM,IAAI,GAAG,CAAC,IAAI,WAAW;AAChE,UAAM,MAAM,eAAe,IAAI,KAAK,UAAU,KAAK,CAAC;AACpD,QAAI,KAAK,IAAI;AACb,mBAAe,IAAI,KAAK,YAAY,GAAG;AAAA,EACzC;AACA,aAAW,CAAC,YAAY,OAAO,KAAK,gBAAgB;AAClD,UAAM,KAAK,UAAU,SAAS,IAAI;AAClC,eAAW,UAAU,IAAI,EAAE,eAAe,GAAG,MAAM,MAAM,GAAG,MAAM,GAAG,GAAG,EAAE;AAAA,EAC5E;AACA,SAAO;AAAA,IACL;AAAA,IACA;AAAA,IACA,cAAc,MAAM,OAAO,CAAC,GAAG,MAAM,IAAI,EAAE,SAAS,CAAC;AAAA,IACrD,eAAe,MAAM,OAAO,CAAC,MAAM,CAAC,EAAE,KAAK,EAAE;AAAA,IAC7C,cAAc,MAAM,OAAO,CAAC,MAAM,EAAE,OAAO,WAAW,UAAU,CAAC,EAAE;AAAA,IACnE,aAAa,MAAM,OAAO,CAAC,MAAM,EAAE,MAAM,EAAE;AAAA,IAC3C,aAAa,MAAM,OAAO,CAAC,MAAM,EAAE,SAAS,CAAC,EAAE,MAAM,WAAW,UAAU,CAAC,EAAE;AAAA,EAC/E;AACF;AAKA,SAAS,UAAU,SAAmB,MAA8B;AAClE,QAAM,IAAI,QAAQ;AAClB,MAAI,MAAM,EAAG,QAAO,EAAE,MAAM,GAAG,OAAO,GAAG,MAAM,CAAC,GAAG,CAAC,GAAG,GAAG,EAAE;AAC5D,QAAM,OAAO,QAAQ,OAAO,CAAC,GAAG,MAAM,IAAI,GAAG,CAAC,IAAI;AAClD,QAAM,WAAW,QAAQ,OAAO,CAAC,GAAG,MAAM,KAAK,IAAI,SAAS,GAAG,CAAC,IAAI,KAAK,IAAI,GAAG,IAAI,CAAC;AACrF,QAAM,QAAQ,KAAK,KAAK,QAAQ;AAChC,QAAM,KAAK,mBAAmB,SAAS,MAAM,EAAE,MAAM,WAAW,IAAK,CAAC;AACtE,SAAO,EAAE,MAAM,OAAO,MAAM,CAAC,GAAG,OAAO,GAAG,KAAK,GAAG,EAAE;AACtD;","names":[]}
package/dist/cli.js CHANGED
@@ -9,7 +9,7 @@ import {
9
9
  import "./chunk-VXNVVBZO.js";
10
10
  import "./chunk-PC4UYEBM.js";
11
11
  import "./chunk-QYJT52YW.js";
12
- import "./chunk-PZ5AY32C.js";
12
+ import "./chunk-NSBPE2FW.js";
13
13
 
14
14
  // src/cli.ts
15
15
  import { writeFileSync } from "fs";
package/dist/control.js CHANGED
@@ -17,7 +17,7 @@ import "./chunk-BWZEGTES.js";
17
17
  import "./chunk-TVVP3ZZQ.js";
18
18
  import "./chunk-VSMTAMNK.js";
19
19
  import "./chunk-QYJT52YW.js";
20
- import "./chunk-PZ5AY32C.js";
20
+ import "./chunk-NSBPE2FW.js";
21
21
  export {
22
22
  allCriticalPassed,
23
23
  controlRunToRunRecord,
@@ -6,7 +6,7 @@ import {
6
6
  soc2Report,
7
7
  summarize
8
8
  } from "../chunk-KKHDIONI.js";
9
- import "../chunk-PZ5AY32C.js";
9
+ import "../chunk-NSBPE2FW.js";
10
10
  export {
11
11
  classifyEuAiRisk,
12
12
  euAiActReport,
package/dist/index.js CHANGED
@@ -16,7 +16,7 @@ import {
16
16
  BENCHMARK_SPLIT_SEED,
17
17
  benchmarks_exports,
18
18
  deterministicSplit
19
- } from "./chunk-6QDKWHLS.js";
19
+ } from "./chunk-MHQPVHXU.js";
20
20
  import {
21
21
  DEFAULT_RULES,
22
22
  classifyFailure,
@@ -243,7 +243,7 @@ import {
243
243
  ValidationError,
244
244
  VerificationError
245
245
  } from "./chunk-QYJT52YW.js";
246
- import "./chunk-PZ5AY32C.js";
246
+ import "./chunk-NSBPE2FW.js";
247
247
 
248
248
  // src/run-score.ts
249
249
  var DEFAULT_RUN_SCORE_WEIGHTS = {
@@ -7,7 +7,7 @@ import {
7
7
  } from "../chunk-3CKU6VGU.js";
8
8
  import "../chunk-NCRFYPS3.js";
9
9
  import "../chunk-TVVP3ZZQ.js";
10
- import "../chunk-PZ5AY32C.js";
10
+ import "../chunk-NSBPE2FW.js";
11
11
  export {
12
12
  acquisitionPlansForKnowledgeGaps,
13
13
  blockingKnowledgeEval,
@@ -3,7 +3,7 @@ import {
3
3
  runAgentMatrix,
4
4
  summariseRows
5
5
  } from "../chunk-QWV226SL.js";
6
- import "../chunk-PZ5AY32C.js";
6
+ import "../chunk-NSBPE2FW.js";
7
7
  export {
8
8
  buildByAxis,
9
9
  runAgentMatrix,
@@ -6,7 +6,7 @@ import {
6
6
  llmSpans
7
7
  } from "../chunk-47X6LRCE.js";
8
8
  import "../chunk-5BKGXME7.js";
9
- import "../chunk-PZ5AY32C.js";
9
+ import "../chunk-NSBPE2FW.js";
10
10
 
11
11
  // src/meta-eval/calibration.ts
12
12
  async function calibrationCurve(traceStore, outcomeStore, evalMetric, outcomeMetric, options = {}) {
@@ -1,7 +1,7 @@
1
1
  import {
2
2
  runAgentMatrix
3
3
  } from "../chunk-QWV226SL.js";
4
- import "../chunk-PZ5AY32C.js";
4
+ import "../chunk-NSBPE2FW.js";
5
5
 
6
6
  // src/multishot/router.ts
7
7
  async function routerCompletion(req) {
package/dist/openapi.json CHANGED
@@ -2,7 +2,7 @@
2
2
  "openapi": "3.1.0",
3
3
  "info": {
4
4
  "title": "@tangle-network/agent-eval — wire protocol",
5
- "version": "0.42.0",
5
+ "version": "0.43.1",
6
6
  "description": "HTTP and stdio RPC interface to agent-eval. The TypeScript runtime is the source of truth; this spec is the contract that cross-language clients (Python, Rust, Go) generate from.\n\nWire-protocol version: 1.0.0. Bumps on breaking changes to request/response schemas.",
7
7
  "contact": {
8
8
  "name": "Tangle Network",
@@ -19,7 +19,7 @@ import {
19
19
  } from "../chunk-47X6LRCE.js";
20
20
  import "../chunk-5BKGXME7.js";
21
21
  import "../chunk-QYJT52YW.js";
22
- import "../chunk-PZ5AY32C.js";
22
+ import "../chunk-NSBPE2FW.js";
23
23
 
24
24
  // src/pipelines/budget-breach.ts
25
25
  async function budgetBreachView(store, options = {}) {
package/dist/prm/index.js CHANGED
@@ -9,7 +9,7 @@ import "../chunk-5BKGXME7.js";
9
9
  import {
10
10
  TraceEmitter
11
11
  } from "../chunk-TVVP3ZZQ.js";
12
- import "../chunk-PZ5AY32C.js";
12
+ import "../chunk-NSBPE2FW.js";
13
13
 
14
14
  // src/prm/builtin-rubrics.ts
15
15
  function outputLengthRubric(args = {}) {
package/dist/reporting.js CHANGED
@@ -24,7 +24,7 @@ import {
24
24
  } from "./chunk-WP7SY7AI.js";
25
25
  import "./chunk-VSMTAMNK.js";
26
26
  import "./chunk-QYJT52YW.js";
27
- import "./chunk-PZ5AY32C.js";
27
+ import "./chunk-NSBPE2FW.js";
28
28
  export {
29
29
  RESEARCH_REPORT_HARD_PAIR_FLOOR,
30
30
  assertReleaseConfidence,
package/dist/rl.js CHANGED
@@ -26,7 +26,7 @@ import "./chunk-PC4UYEBM.js";
26
26
  import {
27
27
  ValidationError
28
28
  } from "./chunk-QYJT52YW.js";
29
- import "./chunk-PZ5AY32C.js";
29
+ import "./chunk-NSBPE2FW.js";
30
30
 
31
31
  // src/rl/compute-curves.ts
32
32
  async function runComputeCurve(opts) {
@@ -1,10 +1,10 @@
1
1
  import {
2
2
  runCampaign
3
- } from "./chunk-YNMCYUWT.js";
3
+ } from "./chunk-RXK7FXLV.js";
4
4
  import "./chunk-WP7SY7AI.js";
5
5
  import "./chunk-QYJT52YW.js";
6
- import "./chunk-PZ5AY32C.js";
6
+ import "./chunk-NSBPE2FW.js";
7
7
  export {
8
8
  runCampaign
9
9
  };
10
- //# sourceMappingURL=run-campaign-KEJK5KFT.js.map
10
+ //# sourceMappingURL=run-campaign-GNDO66B4.js.map
@@ -1,4 +1,4 @@
1
- import "../chunk-PZ5AY32C.js";
1
+ import "../chunk-NSBPE2FW.js";
2
2
 
3
3
  // src/telemetry/sink-file.ts
4
4
  import * as fs from "fs";
@@ -1,4 +1,4 @@
1
- import "../chunk-PZ5AY32C.js";
1
+ import "../chunk-NSBPE2FW.js";
2
2
 
3
3
  // src/telemetry/schema.ts
4
4
  var TELEMETRY_SCHEMA_VERSION = 1;
package/dist/traces.js CHANGED
@@ -76,7 +76,7 @@ import {
76
76
  providerFromBaseUrl
77
77
  } from "./chunk-PC4UYEBM.js";
78
78
  import "./chunk-QYJT52YW.js";
79
- import "./chunk-PZ5AY32C.js";
79
+ import "./chunk-NSBPE2FW.js";
80
80
  export {
81
81
  DEFAULT_REDACTION_RULES,
82
82
  DEFAULT_TRACE_ANALYST_BUDGETS,
@@ -38,7 +38,7 @@ import {
38
38
  import "../chunk-VXNVVBZO.js";
39
39
  import "../chunk-PC4UYEBM.js";
40
40
  import "../chunk-QYJT52YW.js";
41
- import "../chunk-PZ5AY32C.js";
41
+ import "../chunk-NSBPE2FW.js";
42
42
  export {
43
43
  BUILTIN_RUBRICS,
44
44
  ErrorResponseSchema,
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@tangle-network/agent-eval",
3
- "version": "0.42.0",
3
+ "version": "0.43.1",
4
4
  "description": "Substrate for self-improving agents: traces, verifiable rewards, preferences, GEPA / reflective mutation, auto-research, replay, sequential anytime-valid stats, and release gates.",
5
5
  "homepage": "https://github.com/tangle-network/agent-eval#readme",
6
6
  "repository": {
@@ -119,6 +119,18 @@
119
119
  "publishConfig": {
120
120
  "access": "public"
121
121
  },
122
+ "scripts": {
123
+ "build": "tsup && pnpm openapi",
124
+ "dev": "tsup --watch",
125
+ "prepare": "husky",
126
+ "prepublishOnly": "pnpm build",
127
+ "test": "vitest run",
128
+ "test:watch": "vitest",
129
+ "typecheck": "tsc --noEmit",
130
+ "lint": "biome check src",
131
+ "format": "biome format --write src",
132
+ "openapi": "node dist/cli.js openapi --out dist/openapi.json"
133
+ },
122
134
  "dependencies": {
123
135
  "@asteasolutions/zod-to-openapi": "^8.5.0",
124
136
  "@ax-llm/ax": "^19.0.25",
@@ -129,7 +141,7 @@
129
141
  },
130
142
  "peerDependencies": {
131
143
  "@tangle-network/agent-runtime": "^0.21.0",
132
- "@tangle-network/sandbox": "^0.2.1"
144
+ "@tangle-network/sandbox": ">=0.2.1 <0.4.0"
133
145
  },
134
146
  "peerDependenciesMeta": {
135
147
  "@tangle-network/agent-runtime": {
@@ -142,7 +154,7 @@
142
154
  "devDependencies": {
143
155
  "@biomejs/biome": "^2.4.15",
144
156
  "@tangle-network/agent-runtime": "^0.21.0",
145
- "@tangle-network/sandbox": "^0.2.1",
157
+ "@tangle-network/sandbox": "0.3.0",
146
158
  "@types/node": "^25.6.0",
147
159
  "husky": "^9.1.7",
148
160
  "lint-staged": "^17.0.5",
@@ -151,6 +163,17 @@
151
163
  "typescript": "^5.7.0",
152
164
  "vitest": "^3.0.0"
153
165
  },
166
+ "pnpm": {
167
+ "minimumReleaseAge": 4320,
168
+ "minimumReleaseAgeExclude": [
169
+ "@tangle-network/sandbox",
170
+ "@tangle-network/agent-runtime"
171
+ ],
172
+ "overrides": {
173
+ "postcss@<8.5.10": "^8.5.10",
174
+ "ws@>=8.0.0 <8.20.1": "^8.20.1"
175
+ }
176
+ },
154
177
  "engines": {
155
178
  "node": ">=20"
156
179
  },
@@ -163,14 +186,5 @@
163
186
  ]
164
187
  },
165
188
  "license": "MIT",
166
- "scripts": {
167
- "build": "tsup && pnpm openapi",
168
- "dev": "tsup --watch",
169
- "test": "vitest run",
170
- "test:watch": "vitest",
171
- "typecheck": "tsc --noEmit",
172
- "lint": "biome check src",
173
- "format": "biome format --write src",
174
- "openapi": "node dist/cli.js openapi --out dist/openapi.json"
175
- }
176
- }
189
+ "packageManager": "pnpm@10.22.0"
190
+ }
@@ -1,10 +0,0 @@
1
- var __defProp = Object.defineProperty;
2
- var __export = (target, all) => {
3
- for (var name in all)
4
- __defProp(target, name, { get: all[name], enumerable: true });
5
- };
6
-
7
- export {
8
- __export
9
- };
10
- //# sourceMappingURL=chunk-PZ5AY32C.js.map
@@ -1 +0,0 @@
1
- {"version":3,"sources":["../src/campaign/run-campaign.ts"],"sourcesContent":["/**\n * @experimental\n *\n * `runCampaign` — Pass A substrate primitive. ONE function that orchestrates\n * scenarios → dispatch → artifacts → judges → aggregates, with full\n * reproducibility (seed + manifest hash), cell-level resumability, bootstrap\n * CIs, and the `LabeledScenarioStore` capture flywheel.\n *\n * Improvement loops (optimizer / gate / autoOnPromote) ride on top of this\n * primitive but live in `presets/run-improvement-loop.ts`. This file keeps\n * the core orchestrator minimal — Phase 1 of the Pass A track.\n */\n\nimport { createHash } from 'node:crypto'\nimport { existsSync, mkdirSync, readFileSync, writeFileSync } from 'node:fs'\nimport { join } from 'node:path'\nimport { confidenceInterval } from '../statistics'\nimport type {\n CampaignAggregates,\n CampaignArtifactWriter,\n CampaignCellResult,\n CampaignCostMeter,\n CampaignResult,\n CampaignTraceWriter,\n DispatchContext,\n DispatchFn,\n JudgeAggregate,\n JudgeConfig,\n JudgeScore,\n LabeledScenarioStore,\n Scenario,\n ScenarioAggregate,\n TraceSpan,\n} from './types'\n\nexport interface RunCampaignOptions<TScenario extends Scenario, TArtifact> {\n scenarios: TScenario[]\n dispatch: DispatchFn<TScenario, TArtifact>\n judges?: JudgeConfig<TArtifact, TScenario>[]\n /** Required for reproducibility. Default 42. */\n seed?: number\n /** Per-scenario replicates for CI bands. Default 1; raise to 5+ for\n * bootstrap-tight intervals on critical eval. */\n reps?: number\n /** When true (default), completed cells are cached by\n * (manifestHash, scenarioId, rep, generation). Re-runs skip cached cells. */\n resumable?: boolean\n /** Optional store — when present, every artifact + judge score is captured\n * with the configured `captureSource`. Capture is default ON; pass `'off'`\n * to disable. */\n labeledStore?: LabeledScenarioStore | 'off'\n captureSource?: 'production-trace' | 'eval-run' | 'manual' | 'red-team' | 'synthetic'\n captureSourceVersionHash?: string\n /** Wall-clock cost cap across all cells. Cells beyond ceiling are skipped. */\n costCeiling?: number\n /** Max concurrent cells. Default 2. */\n maxConcurrency?: number\n /** Required: where artifacts + traces land. */\n runDir: string\n /** Tracing posture. Default is the substrate's `FileSystemTraceStore` rooted\n * at `<runDir>/traces/`. `'off'` disables capture entirely — substrate\n * refuses this when the caller wires `autoOnPromote !== 'none'`. */\n tracing?: 'on' | 'off'\n /** Test seam — override the wall clock for deterministic tests. */\n now?: () => Date\n /** Test seam — override per-cell trace writer factory. */\n buildTraceWriter?: (cellId: string, dir: string) => CampaignTraceWriter\n}\n\nexport async function runCampaign<TScenario extends Scenario, TArtifact>(\n opts: RunCampaignOptions<TScenario, TArtifact>,\n): Promise<CampaignResult<TArtifact, TScenario>> {\n const seed = opts.seed ?? 42\n const reps = opts.reps ?? 1\n const resumable = opts.resumable ?? true\n const maxConcurrency = opts.maxConcurrency ?? 2\n const now = opts.now ?? (() => new Date())\n const judges = opts.judges ?? []\n\n if (!existsSync(opts.runDir)) mkdirSync(opts.runDir, { recursive: true })\n\n const manifestHash = computeManifestHash({\n scenarios: opts.scenarios,\n judges: judges as unknown as JudgeConfig<unknown>[],\n dispatchRef: opts.dispatch.name || 'anonymous',\n seed,\n reps,\n })\n\n const startedAt = now()\n const cells: CampaignCellResult<TArtifact>[] = []\n const artifactsByPath: Record<string, string> = {}\n\n // Build the cell schedule (scenario × rep).\n const schedule: Array<{ scenario: TScenario; rep: number; cellId: string; cellSeed: number }> = []\n let cellIndex = 0\n for (const scenario of opts.scenarios) {\n for (let rep = 0; rep < reps; rep++) {\n const cellId = `${scenario.id}:${rep}`\n const cellSeed = seed + cellIndex\n schedule.push({ scenario, rep, cellId, cellSeed })\n cellIndex += 1\n }\n }\n\n // Concurrency-limited execution.\n let totalCostUsd = 0\n let costCeilingReached = false\n const abortController = new AbortController()\n // Concurrency lanes that drain the cell schedule. Named \"lanes\" — not\n // \"workers\" — to avoid clashing with the taxonomy's worker (= the agent\n // harness in a sandbox, invoked behind `dispatch`). See loop-taxonomy.md.\n const lanes: Promise<void>[] = []\n let nextIdx = 0\n const cellsRef = cells\n\n for (let i = 0; i < maxConcurrency; i++) {\n lanes.push(\n (async () => {\n while (true) {\n const myIdx = nextIdx++\n if (myIdx >= schedule.length) return\n const slot = schedule[myIdx]!\n if (costCeilingReached) {\n cellsRef.push(skippedCell(slot, 'cost_ceiling_reached'))\n continue\n }\n const result = await executeCell({\n slot,\n opts,\n manifestHash,\n resumable,\n now,\n buildTraceWriter: opts.buildTraceWriter ?? defaultBuildTraceWriter,\n signal: abortController.signal,\n })\n cellsRef.push(result.cell)\n totalCostUsd += result.cell.costUsd\n Object.assign(artifactsByPath, result.artifactsByPath)\n if (opts.costCeiling !== undefined && totalCostUsd >= opts.costCeiling) {\n costCeilingReached = true\n }\n // Capture into LabeledScenarioStore unless explicitly disabled.\n if (opts.labeledStore && opts.labeledStore !== 'off' && !result.cell.error) {\n await captureToStore({\n store: opts.labeledStore,\n cell: result.cell,\n scenario: slot.scenario,\n opts,\n now,\n }).catch((err) => {\n // Capture failures are non-fatal — log but don't crash the campaign.\n // (Trace would normally land here.)\n console.warn(\n `[runCampaign] capture failed for ${result.cell.cellId}: ${err instanceof Error ? err.message : String(err)}`,\n )\n })\n }\n }\n })(),\n )\n }\n await Promise.all(lanes)\n\n const endedAt = now()\n cellsRef.sort((a, b) => a.cellId.localeCompare(b.cellId))\n\n const aggregates = computeAggregates(\n cellsRef,\n judges as unknown as JudgeConfig<TArtifact>[],\n seed,\n )\n\n return {\n manifestHash,\n seed,\n startedAt: startedAt.toISOString(),\n endedAt: endedAt.toISOString(),\n durationMs: endedAt.getTime() - startedAt.getTime(),\n cells: cellsRef,\n aggregates,\n runDir: opts.runDir,\n artifactsByPath,\n scenarios: opts.scenarios.map((s) => ({ id: s.id, kind: s.kind })),\n }\n}\n\n// ── Internals ─────────────────────────────────────────────────────────\n\ninterface ExecuteCellArgs<TScenario extends Scenario, TArtifact> {\n slot: { scenario: TScenario; rep: number; cellId: string; cellSeed: number }\n opts: RunCampaignOptions<TScenario, TArtifact>\n manifestHash: string\n resumable: boolean\n now: () => Date\n buildTraceWriter: (cellId: string, dir: string) => CampaignTraceWriter\n signal: AbortSignal\n}\n\nasync function executeCell<TScenario extends Scenario, TArtifact>(\n args: ExecuteCellArgs<TScenario, TArtifact>,\n): Promise<{ cell: CampaignCellResult<TArtifact>; artifactsByPath: Record<string, string> }> {\n const cellDir = join(args.opts.runDir, args.slot.cellId.replace(/[^a-zA-Z0-9_-]/g, '_'))\n if (!existsSync(cellDir)) mkdirSync(cellDir, { recursive: true })\n\n // Resumability: cache key = (manifestHash, scenarioId, rep)\n const cachePath = join(cellDir, 'cached-result.json')\n if (args.resumable && existsSync(cachePath)) {\n try {\n const cached = JSON.parse(readFileSync(cachePath, 'utf8')) as CampaignCellResult<TArtifact>\n if (cached.cellId === args.slot.cellId) {\n return { cell: { ...cached, cached: true }, artifactsByPath: {} }\n }\n } catch {\n // Corrupt cache — fall through to re-run.\n }\n }\n\n const startMs = Date.now()\n const trace = args.buildTraceWriter(args.slot.cellId, cellDir)\n const artifactsByPath: Record<string, string> = {}\n const artifacts: CampaignArtifactWriter = {\n async write(path, content) {\n const fullPath = join(cellDir, path)\n const dir = join(fullPath, '..')\n if (!existsSync(dir)) mkdirSync(dir, { recursive: true })\n writeFileSync(fullPath, content as Uint8Array)\n artifactsByPath[`${args.slot.cellId}/${path}`] = fullPath\n return fullPath\n },\n async writeJson(path, value) {\n return artifacts.write(path, JSON.stringify(value, null, 2))\n },\n }\n let costSoFar = 0\n const cost: CampaignCostMeter = {\n observe(amount, source) {\n costSoFar += amount\n trace.span(`cost.${source}`, { amountUsd: amount }).end()\n },\n current() {\n return costSoFar\n },\n }\n\n const ctx: DispatchContext = {\n cellId: args.slot.cellId,\n rep: args.slot.rep,\n seed: args.slot.cellSeed,\n signal: args.signal,\n trace,\n artifacts,\n cost,\n }\n\n let artifact: TArtifact | undefined\n let errorMessage: string | undefined\n try {\n artifact = await args.opts.dispatch(args.slot.scenario, ctx)\n } catch (err) {\n errorMessage = err instanceof Error ? err.message : String(err)\n }\n\n // Run judges (only if we have an artifact). A judge that throws invalidates\n // the cell — recorded as `error`, NOT folded into a fake composite:0 (a fake\n // zero is indistinguishable from a real zero and poisons every aggregate).\n const judgeScores: Record<string, JudgeScore> = {}\n if (artifact !== undefined) {\n for (const judge of args.opts.judges ?? []) {\n if (judge.appliesTo && !judge.appliesTo(args.slot.scenario)) continue\n try {\n judgeScores[judge.name] = await runJudgeCell(judge, {\n artifact,\n scenario: args.slot.scenario,\n signal: args.signal,\n })\n } catch (err) {\n errorMessage = `judge '${judge.name}' failed: ${err instanceof Error ? err.message : String(err)}`\n break\n }\n }\n }\n\n await trace.flush()\n\n const cell: CampaignCellResult<TArtifact> = {\n cellId: args.slot.cellId,\n scenarioId: args.slot.scenario.id,\n rep: args.slot.rep,\n artifact: (artifact ?? null) as TArtifact,\n judgeScores,\n costUsd: costSoFar,\n durationMs: Date.now() - startMs,\n seed: args.slot.cellSeed,\n cached: false,\n error: errorMessage,\n }\n\n if (!errorMessage && args.resumable) {\n writeFileSync(cachePath, JSON.stringify(cell))\n }\n\n return { cell, artifactsByPath }\n}\n\nasync function runJudgeCell<TArtifact, TScenario extends Scenario>(\n judge: JudgeConfig<TArtifact, TScenario>,\n input: { artifact: TArtifact; scenario: TScenario; signal: AbortSignal },\n): Promise<JudgeScore> {\n return judge.score(input)\n}\n\nfunction defaultBuildTraceWriter(cellId: string, dir: string): CampaignTraceWriter {\n const spans: Array<Record<string, unknown>> = []\n return {\n span(name, attributes) {\n const startMs = Date.now()\n const record: Record<string, unknown> = { name, cellId, startMs, ...(attributes ?? {}) }\n const finish: TraceSpan = {\n end(endAttrs) {\n record.durationMs = Date.now() - startMs\n if (endAttrs) Object.assign(record, endAttrs)\n spans.push(record)\n },\n setAttribute(key, value) {\n record[key] = value\n },\n }\n return finish\n },\n async flush() {\n const path = join(dir, 'spans.jsonl')\n writeFileSync(path, spans.map((s) => JSON.stringify(s)).join('\\n'))\n },\n }\n}\n\nfunction skippedCell<TScenario extends Scenario, TArtifact>(\n slot: { scenario: TScenario; rep: number; cellId: string; cellSeed: number },\n reason: string,\n): CampaignCellResult<TArtifact> {\n return {\n cellId: slot.cellId,\n scenarioId: slot.scenario.id,\n rep: slot.rep,\n artifact: null as unknown as TArtifact,\n judgeScores: {},\n costUsd: 0,\n durationMs: 0,\n seed: slot.cellSeed,\n cached: false,\n error: `skipped: ${reason}`,\n }\n}\n\ninterface CaptureArgs<TScenario extends Scenario, TArtifact> {\n store: LabeledScenarioStore\n cell: CampaignCellResult<TArtifact>\n scenario: TScenario\n opts: RunCampaignOptions<TScenario, TArtifact>\n now: () => Date\n}\n\nasync function captureToStore<TScenario extends Scenario, TArtifact>(\n args: CaptureArgs<TScenario, TArtifact>,\n): Promise<void> {\n await args.store.observe({\n scenario: args.scenario,\n artifact: args.cell.artifact,\n judgeScores: args.cell.judgeScores,\n source: args.opts.captureSource ?? 'eval-run',\n sourceVersionHash: args.opts.captureSourceVersionHash ?? 'unknown',\n capturedAt: args.now().toISOString(),\n redactionStatus: 'raw',\n })\n}\n\n// ── Aggregates + manifest hash ────────────────────────────────────────\n\nfunction computeManifestHash(input: {\n scenarios: Scenario[]\n judges: JudgeConfig<unknown>[]\n dispatchRef: string\n seed: number\n reps: number\n}): string {\n const canonical = {\n scenarios: input.scenarios.map((s) => ({ id: s.id, kind: s.kind })),\n judges: input.judges.map((j) => ({ name: j.name, dims: j.dimensions.map((d) => d.key) })),\n dispatch: input.dispatchRef,\n seed: input.seed,\n reps: input.reps,\n }\n return createHash('sha256').update(JSON.stringify(canonical)).digest('hex')\n}\n\nfunction computeAggregates<TArtifact>(\n cells: CampaignCellResult<TArtifact>[],\n judges: JudgeConfig<TArtifact>[],\n seed: number,\n): CampaignAggregates {\n const byJudge: Record<string, JudgeAggregate> = {}\n for (const judge of judges) {\n const scores: number[] = []\n for (const cell of cells) {\n const s = cell.judgeScores[judge.name]\n if (s !== undefined) scores.push(s.composite)\n }\n byJudge[judge.name] = aggregate(scores, seed)\n }\n const byScenario: Record<string, ScenarioAggregate> = {}\n const scenarioGroups = new Map<string, number[]>()\n for (const cell of cells) {\n const composites = Object.values(cell.judgeScores).map((s) => s.composite)\n if (composites.length === 0) continue\n const mean = composites.reduce((a, b) => a + b, 0) / composites.length\n const arr = scenarioGroups.get(cell.scenarioId) ?? []\n arr.push(mean)\n scenarioGroups.set(cell.scenarioId, arr)\n }\n for (const [scenarioId, samples] of scenarioGroups) {\n const ag = aggregate(samples, seed)\n byScenario[scenarioId] = { meanComposite: ag.mean, ci95: ag.ci95, n: ag.n }\n }\n return {\n byJudge,\n byScenario,\n totalCostUsd: cells.reduce((a, c) => a + c.costUsd, 0),\n cellsExecuted: cells.filter((c) => !c.error).length,\n cellsSkipped: cells.filter((c) => c.error?.startsWith('skipped:')).length,\n cellsCached: cells.filter((c) => c.cached).length,\n cellsFailed: cells.filter((c) => c.error && !c.error.startsWith('skipped:')).length,\n }\n}\n\n// Percentile bootstrap CI95 via seeded resampling. Deterministic for a given\n// seed — same campaign re-run produces identical CI bands. Falls back to\n// degenerate intervals at n<=1 (the bootstrap is undefined there).\nfunction aggregate(samples: number[], seed: number): JudgeAggregate {\n const n = samples.length\n if (n === 0) return { mean: 0, stdev: 0, ci95: [0, 0], n: 0 }\n const mean = samples.reduce((a, b) => a + b, 0) / n\n const variance = samples.reduce((a, b) => a + (b - mean) ** 2, 0) / Math.max(1, n - 1)\n const stdev = Math.sqrt(variance)\n const ci = confidenceInterval(samples, 0.95, { seed, resamples: 1000 })\n return { mean, stdev, ci95: [ci.lower, ci.upper], n }\n}\n"],"mappings":";;;;;AAaA,SAAS,kBAAkB;AAC3B,SAAS,YAAY,WAAW,cAAc,qBAAqB;AACnE,SAAS,YAAY;AAsDrB,eAAsB,YACpB,MAC+C;AAC/C,QAAM,OAAO,KAAK,QAAQ;AAC1B,QAAM,OAAO,KAAK,QAAQ;AAC1B,QAAM,YAAY,KAAK,aAAa;AACpC,QAAM,iBAAiB,KAAK,kBAAkB;AAC9C,QAAM,MAAM,KAAK,QAAQ,MAAM,oBAAI,KAAK;AACxC,QAAM,SAAS,KAAK,UAAU,CAAC;AAE/B,MAAI,CAAC,WAAW,KAAK,MAAM,EAAG,WAAU,KAAK,QAAQ,EAAE,WAAW,KAAK,CAAC;AAExE,QAAM,eAAe,oBAAoB;AAAA,IACvC,WAAW,KAAK;AAAA,IAChB;AAAA,IACA,aAAa,KAAK,SAAS,QAAQ;AAAA,IACnC;AAAA,IACA;AAAA,EACF,CAAC;AAED,QAAM,YAAY,IAAI;AACtB,QAAM,QAAyC,CAAC;AAChD,QAAM,kBAA0C,CAAC;AAGjD,QAAM,WAA0F,CAAC;AACjG,MAAI,YAAY;AAChB,aAAW,YAAY,KAAK,WAAW;AACrC,aAAS,MAAM,GAAG,MAAM,MAAM,OAAO;AACnC,YAAM,SAAS,GAAG,SAAS,EAAE,IAAI,GAAG;AACpC,YAAM,WAAW,OAAO;AACxB,eAAS,KAAK,EAAE,UAAU,KAAK,QAAQ,SAAS,CAAC;AACjD,mBAAa;AAAA,IACf;AAAA,EACF;AAGA,MAAI,eAAe;AACnB,MAAI,qBAAqB;AACzB,QAAM,kBAAkB,IAAI,gBAAgB;AAI5C,QAAM,QAAyB,CAAC;AAChC,MAAI,UAAU;AACd,QAAM,WAAW;AAEjB,WAAS,IAAI,GAAG,IAAI,gBAAgB,KAAK;AACvC,UAAM;AAAA,OACH,YAAY;AACX,eAAO,MAAM;AACX,gBAAM,QAAQ;AACd,cAAI,SAAS,SAAS,OAAQ;AAC9B,gBAAM,OAAO,SAAS,KAAK;AAC3B,cAAI,oBAAoB;AACtB,qBAAS,KAAK,YAAY,MAAM,sBAAsB,CAAC;AACvD;AAAA,UACF;AACA,gBAAM,SAAS,MAAM,YAAY;AAAA,YAC/B;AAAA,YACA;AAAA,YACA;AAAA,YACA;AAAA,YACA;AAAA,YACA,kBAAkB,KAAK,oBAAoB;AAAA,YAC3C,QAAQ,gBAAgB;AAAA,UAC1B,CAAC;AACD,mBAAS,KAAK,OAAO,IAAI;AACzB,0BAAgB,OAAO,KAAK;AAC5B,iBAAO,OAAO,iBAAiB,OAAO,eAAe;AACrD,cAAI,KAAK,gBAAgB,UAAa,gBAAgB,KAAK,aAAa;AACtE,iCAAqB;AAAA,UACvB;AAEA,cAAI,KAAK,gBAAgB,KAAK,iBAAiB,SAAS,CAAC,OAAO,KAAK,OAAO;AAC1E,kBAAM,eAAe;AAAA,cACnB,OAAO,KAAK;AAAA,cACZ,MAAM,OAAO;AAAA,cACb,UAAU,KAAK;AAAA,cACf;AAAA,cACA;AAAA,YACF,CAAC,EAAE,MAAM,CAAC,QAAQ;AAGhB,sBAAQ;AAAA,gBACN,oCAAoC,OAAO,KAAK,MAAM,KAAK,eAAe,QAAQ,IAAI,UAAU,OAAO,GAAG,CAAC;AAAA,cAC7G;AAAA,YACF,CAAC;AAAA,UACH;AAAA,QACF;AAAA,MACF,GAAG;AAAA,IACL;AAAA,EACF;AACA,QAAM,QAAQ,IAAI,KAAK;AAEvB,QAAM,UAAU,IAAI;AACpB,WAAS,KAAK,CAAC,GAAG,MAAM,EAAE,OAAO,cAAc,EAAE,MAAM,CAAC;AAExD,QAAM,aAAa;AAAA,IACjB;AAAA,IACA;AAAA,IACA;AAAA,EACF;AAEA,SAAO;AAAA,IACL;AAAA,IACA;AAAA,IACA,WAAW,UAAU,YAAY;AAAA,IACjC,SAAS,QAAQ,YAAY;AAAA,IAC7B,YAAY,QAAQ,QAAQ,IAAI,UAAU,QAAQ;AAAA,IAClD,OAAO;AAAA,IACP;AAAA,IACA,QAAQ,KAAK;AAAA,IACb;AAAA,IACA,WAAW,KAAK,UAAU,IAAI,CAAC,OAAO,EAAE,IAAI,EAAE,IAAI,MAAM,EAAE,KAAK,EAAE;AAAA,EACnE;AACF;AAcA,eAAe,YACb,MAC2F;AAC3F,QAAM,UAAU,KAAK,KAAK,KAAK,QAAQ,KAAK,KAAK,OAAO,QAAQ,mBAAmB,GAAG,CAAC;AACvF,MAAI,CAAC,WAAW,OAAO,EAAG,WAAU,SAAS,EAAE,WAAW,KAAK,CAAC;AAGhE,QAAM,YAAY,KAAK,SAAS,oBAAoB;AACpD,MAAI,KAAK,aAAa,WAAW,SAAS,GAAG;AAC3C,QAAI;AACF,YAAM,SAAS,KAAK,MAAM,aAAa,WAAW,MAAM,CAAC;AACzD,UAAI,OAAO,WAAW,KAAK,KAAK,QAAQ;AACtC,eAAO,EAAE,MAAM,EAAE,GAAG,QAAQ,QAAQ,KAAK,GAAG,iBAAiB,CAAC,EAAE;AAAA,MAClE;AAAA,IACF,QAAQ;AAAA,IAER;AAAA,EACF;AAEA,QAAM,UAAU,KAAK,IAAI;AACzB,QAAM,QAAQ,KAAK,iBAAiB,KAAK,KAAK,QAAQ,OAAO;AAC7D,QAAM,kBAA0C,CAAC;AACjD,QAAM,YAAoC;AAAA,IACxC,MAAM,MAAM,MAAM,SAAS;AACzB,YAAM,WAAW,KAAK,SAAS,IAAI;AACnC,YAAM,MAAM,KAAK,UAAU,IAAI;AAC/B,UAAI,CAAC,WAAW,GAAG,EAAG,WAAU,KAAK,EAAE,WAAW,KAAK,CAAC;AACxD,oBAAc,UAAU,OAAqB;AAC7C,sBAAgB,GAAG,KAAK,KAAK,MAAM,IAAI,IAAI,EAAE,IAAI;AACjD,aAAO;AAAA,IACT;AAAA,IACA,MAAM,UAAU,MAAM,OAAO;AAC3B,aAAO,UAAU,MAAM,MAAM,KAAK,UAAU,OAAO,MAAM,CAAC,CAAC;AAAA,IAC7D;AAAA,EACF;AACA,MAAI,YAAY;AAChB,QAAM,OAA0B;AAAA,IAC9B,QAAQ,QAAQ,QAAQ;AACtB,mBAAa;AACb,YAAM,KAAK,QAAQ,MAAM,IAAI,EAAE,WAAW,OAAO,CAAC,EAAE,IAAI;AAAA,IAC1D;AAAA,IACA,UAAU;AACR,aAAO;AAAA,IACT;AAAA,EACF;AAEA,QAAM,MAAuB;AAAA,IAC3B,QAAQ,KAAK,KAAK;AAAA,IAClB,KAAK,KAAK,KAAK;AAAA,IACf,MAAM,KAAK,KAAK;AAAA,IAChB,QAAQ,KAAK;AAAA,IACb;AAAA,IACA;AAAA,IACA;AAAA,EACF;AAEA,MAAI;AACJ,MAAI;AACJ,MAAI;AACF,eAAW,MAAM,KAAK,KAAK,SAAS,KAAK,KAAK,UAAU,GAAG;AAAA,EAC7D,SAAS,KAAK;AACZ,mBAAe,eAAe,QAAQ,IAAI,UAAU,OAAO,GAAG;AAAA,EAChE;AAKA,QAAM,cAA0C,CAAC;AACjD,MAAI,aAAa,QAAW;AAC1B,eAAW,SAAS,KAAK,KAAK,UAAU,CAAC,GAAG;AAC1C,UAAI,MAAM,aAAa,CAAC,MAAM,UAAU,KAAK,KAAK,QAAQ,EAAG;AAC7D,UAAI;AACF,oBAAY,MAAM,IAAI,IAAI,MAAM,aAAa,OAAO;AAAA,UAClD;AAAA,UACA,UAAU,KAAK,KAAK;AAAA,UACpB,QAAQ,KAAK;AAAA,QACf,CAAC;AAAA,MACH,SAAS,KAAK;AACZ,uBAAe,UAAU,MAAM,IAAI,aAAa,eAAe,QAAQ,IAAI,UAAU,OAAO,GAAG,CAAC;AAChG;AAAA,MACF;AAAA,IACF;AAAA,EACF;AAEA,QAAM,MAAM,MAAM;AAElB,QAAM,OAAsC;AAAA,IAC1C,QAAQ,KAAK,KAAK;AAAA,IAClB,YAAY,KAAK,KAAK,SAAS;AAAA,IAC/B,KAAK,KAAK,KAAK;AAAA,IACf,UAAW,YAAY;AAAA,IACvB;AAAA,IACA,SAAS;AAAA,IACT,YAAY,KAAK,IAAI,IAAI;AAAA,IACzB,MAAM,KAAK,KAAK;AAAA,IAChB,QAAQ;AAAA,IACR,OAAO;AAAA,EACT;AAEA,MAAI,CAAC,gBAAgB,KAAK,WAAW;AACnC,kBAAc,WAAW,KAAK,UAAU,IAAI,CAAC;AAAA,EAC/C;AAEA,SAAO,EAAE,MAAM,gBAAgB;AACjC;AAEA,eAAe,aACb,OACA,OACqB;AACrB,SAAO,MAAM,MAAM,KAAK;AAC1B;AAEA,SAAS,wBAAwB,QAAgB,KAAkC;AACjF,QAAM,QAAwC,CAAC;AAC/C,SAAO;AAAA,IACL,KAAK,MAAM,YAAY;AACrB,YAAM,UAAU,KAAK,IAAI;AACzB,YAAM,SAAkC,EAAE,MAAM,QAAQ,SAAS,GAAI,cAAc,CAAC,EAAG;AACvF,YAAM,SAAoB;AAAA,QACxB,IAAI,UAAU;AACZ,iBAAO,aAAa,KAAK,IAAI,IAAI;AACjC,cAAI,SAAU,QAAO,OAAO,QAAQ,QAAQ;AAC5C,gBAAM,KAAK,MAAM;AAAA,QACnB;AAAA,QACA,aAAa,KAAK,OAAO;AACvB,iBAAO,GAAG,IAAI;AAAA,QAChB;AAAA,MACF;AACA,aAAO;AAAA,IACT;AAAA,IACA,MAAM,QAAQ;AACZ,YAAM,OAAO,KAAK,KAAK,aAAa;AACpC,oBAAc,MAAM,MAAM,IAAI,CAAC,MAAM,KAAK,UAAU,CAAC,CAAC,EAAE,KAAK,IAAI,CAAC;AAAA,IACpE;AAAA,EACF;AACF;AAEA,SAAS,YACP,MACA,QAC+B;AAC/B,SAAO;AAAA,IACL,QAAQ,KAAK;AAAA,IACb,YAAY,KAAK,SAAS;AAAA,IAC1B,KAAK,KAAK;AAAA,IACV,UAAU;AAAA,IACV,aAAa,CAAC;AAAA,IACd,SAAS;AAAA,IACT,YAAY;AAAA,IACZ,MAAM,KAAK;AAAA,IACX,QAAQ;AAAA,IACR,OAAO,YAAY,MAAM;AAAA,EAC3B;AACF;AAUA,eAAe,eACb,MACe;AACf,QAAM,KAAK,MAAM,QAAQ;AAAA,IACvB,UAAU,KAAK;AAAA,IACf,UAAU,KAAK,KAAK;AAAA,IACpB,aAAa,KAAK,KAAK;AAAA,IACvB,QAAQ,KAAK,KAAK,iBAAiB;AAAA,IACnC,mBAAmB,KAAK,KAAK,4BAA4B;AAAA,IACzD,YAAY,KAAK,IAAI,EAAE,YAAY;AAAA,IACnC,iBAAiB;AAAA,EACnB,CAAC;AACH;AAIA,SAAS,oBAAoB,OAMlB;AACT,QAAM,YAAY;AAAA,IAChB,WAAW,MAAM,UAAU,IAAI,CAAC,OAAO,EAAE,IAAI,EAAE,IAAI,MAAM,EAAE,KAAK,EAAE;AAAA,IAClE,QAAQ,MAAM,OAAO,IAAI,CAAC,OAAO,EAAE,MAAM,EAAE,MAAM,MAAM,EAAE,WAAW,IAAI,CAAC,MAAM,EAAE,GAAG,EAAE,EAAE;AAAA,IACxF,UAAU,MAAM;AAAA,IAChB,MAAM,MAAM;AAAA,IACZ,MAAM,MAAM;AAAA,EACd;AACA,SAAO,WAAW,QAAQ,EAAE,OAAO,KAAK,UAAU,SAAS,CAAC,EAAE,OAAO,KAAK;AAC5E;AAEA,SAAS,kBACP,OACA,QACA,MACoB;AACpB,QAAM,UAA0C,CAAC;AACjD,aAAW,SAAS,QAAQ;AAC1B,UAAM,SAAmB,CAAC;AAC1B,eAAW,QAAQ,OAAO;AACxB,YAAM,IAAI,KAAK,YAAY,MAAM,IAAI;AACrC,UAAI,MAAM,OAAW,QAAO,KAAK,EAAE,SAAS;AAAA,IAC9C;AACA,YAAQ,MAAM,IAAI,IAAI,UAAU,QAAQ,IAAI;AAAA,EAC9C;AACA,QAAM,aAAgD,CAAC;AACvD,QAAM,iBAAiB,oBAAI,IAAsB;AACjD,aAAW,QAAQ,OAAO;AACxB,UAAM,aAAa,OAAO,OAAO,KAAK,WAAW,EAAE,IAAI,CAAC,MAAM,EAAE,SAAS;AACzE,QAAI,WAAW,WAAW,EAAG;AAC7B,UAAM,OAAO,WAAW,OAAO,CAAC,GAAG,MAAM,IAAI,GAAG,CAAC,IAAI,WAAW;AAChE,UAAM,MAAM,eAAe,IAAI,KAAK,UAAU,KAAK,CAAC;AACpD,QAAI,KAAK,IAAI;AACb,mBAAe,IAAI,KAAK,YAAY,GAAG;AAAA,EACzC;AACA,aAAW,CAAC,YAAY,OAAO,KAAK,gBAAgB;AAClD,UAAM,KAAK,UAAU,SAAS,IAAI;AAClC,eAAW,UAAU,IAAI,EAAE,eAAe,GAAG,MAAM,MAAM,GAAG,MAAM,GAAG,GAAG,EAAE;AAAA,EAC5E;AACA,SAAO;AAAA,IACL;AAAA,IACA;AAAA,IACA,cAAc,MAAM,OAAO,CAAC,GAAG,MAAM,IAAI,EAAE,SAAS,CAAC;AAAA,IACrD,eAAe,MAAM,OAAO,CAAC,MAAM,CAAC,EAAE,KAAK,EAAE;AAAA,IAC7C,cAAc,MAAM,OAAO,CAAC,MAAM,EAAE,OAAO,WAAW,UAAU,CAAC,EAAE;AAAA,IACnE,aAAa,MAAM,OAAO,CAAC,MAAM,EAAE,MAAM,EAAE;AAAA,IAC3C,aAAa,MAAM,OAAO,CAAC,MAAM,EAAE,SAAS,CAAC,EAAE,MAAM,WAAW,UAAU,CAAC,EAAE;AAAA,EAC/E;AACF;AAKA,SAAS,UAAU,SAAmB,MAA8B;AAClE,QAAM,IAAI,QAAQ;AAClB,MAAI,MAAM,EAAG,QAAO,EAAE,MAAM,GAAG,OAAO,GAAG,MAAM,CAAC,GAAG,CAAC,GAAG,GAAG,EAAE;AAC5D,QAAM,OAAO,QAAQ,OAAO,CAAC,GAAG,MAAM,IAAI,GAAG,CAAC,IAAI;AAClD,QAAM,WAAW,QAAQ,OAAO,CAAC,GAAG,MAAM,KAAK,IAAI,SAAS,GAAG,CAAC,IAAI,KAAK,IAAI,GAAG,IAAI,CAAC;AACrF,QAAM,QAAQ,KAAK,KAAK,QAAQ;AAChC,QAAM,KAAK,mBAAmB,SAAS,MAAM,EAAE,MAAM,WAAW,IAAK,CAAC;AACtE,SAAO,EAAE,MAAM,OAAO,MAAM,CAAC,GAAG,OAAO,GAAG,KAAK,GAAG,EAAE;AACtD;","names":[]}