@tangle-network/agent-eval 0.187.2 → 0.189.0

package/dist/emitter-DeQHiDMm.js

@@ -1,257 +0,0 @@
-import { t as newRecordId } from "./record-id-DUgsK5qp.js";
-//#region src/trace/emitter.ts
-/**
-* TraceEmitter — hierarchical span builder that auto-parents using an
-* internal stack. One emitter per Run; emitters do NOT share state.
-*
-* Convenience methods (`llm`, `tool`, `retrieval`, `judge`, `sandbox`)
-* return a `SpanHandle` with `.end()` / `.fail()` so callers don't
-* have to thread spanIds manually. For async workflows that can't use
-* the stack (e.g. fan-out parallel calls), pass `parentSpanId`
-* explicitly.
-*/
-var TraceEmitter = class {
-	store;
-	stack = [];
-	_runId;
-	now;
-	id;
-	hooks;
-	hookErrors;
-	constructor(store, options = {}) {
-		this.store = store;
-		this.now = options.now ?? (() => Date.now());
-		this.id = options.id ?? (() => newRecordId());
-		this._runId = options.runId ?? this.id();
-		this.hooks = options.onRunComplete ?? [];
-		this.hookErrors = options.hookErrors ?? "swallow";
-	}
-	get runId() {
-		return this._runId;
-	}
-	get traceStore() {
-		return this.store;
-	}
-	/** Append a hook after construction (e.g. attach the trace analyst). */
-	addRunCompleteHook(hook) {
-		this.hooks.push(hook);
-	}
-	/**
-	* Begin a Run.
-	*
-	* `scenarioId` is required on the persisted Run shape — every Run downstream
-	* gets a non-empty scenarioId so filters and aggregations stay simple — but
-	* the INPUT here accepts it as optional. When omitted, startRun substitutes
-	* a sensible default (`run.layer ?? run.tags?.['kind'] ?? 'runtime'`) so
-	* runtime / operator / meta-eval runs that have no curated-scenario corpus
-	* to anchor to don't have to invent placeholder strings at the call site.
-	*/
-	async startRun(run) {
-		const scenarioId = run.scenarioId ?? run.layer ?? run.tags?.kind ?? "runtime";
-		const full = {
-			...run,
-			scenarioId,
-			runId: this._runId,
-			startedAt: this.now(),
-			status: "running"
-		};
-		await this.store.appendRun(full);
-		return full;
-	}
-	async endRun(outcome) {
-		const status = outcome?.pass === false ? "failed" : "completed";
-		await this.store.updateRun(this._runId, {
-			endedAt: this.now(),
-			status,
-			outcome
-		});
-		await this.runHooks({
-			runId: this._runId,
-			emitter: this,
-			store: this.store,
-			outcome,
-			status
-		});
-	}
-	async abortRun(reason) {
-		const outcome = {
-			pass: false,
-			notes: reason
-		};
-		await this.store.updateRun(this._runId, {
-			endedAt: this.now(),
-			status: "aborted",
-			outcome
-		});
-		await this.runHooks({
-			runId: this._runId,
-			emitter: this,
-			store: this.store,
-			outcome,
-			status: "aborted"
-		});
-	}
-	async runHooks(ctx) {
-		for (const hook of this.hooks) try {
-			await hook(ctx);
-		} catch (err) {
-			if (this.hookErrors === "throw") throw err;
-			try {
-				await this.store.appendEvent({
-					eventId: this.id(),
-					runId: this._runId,
-					kind: "log",
-					timestamp: this.now(),
-					payload: {
-						source: "run_complete_hook",
-						error: err instanceof Error ? err.message : String(err)
-					}
-				});
-			} catch {}
-		}
-	}
-	async span(init) {
-		const spanId = this.id();
-		const span = {
-			spanId,
-			parentSpanId: init.parentSpanId ?? this.stack[this.stack.length - 1],
-			runId: this._runId,
-			startedAt: this.now(),
-			...init
-		};
-		await this.store.appendSpan(span);
-		this.stack.push(spanId);
-		return this.handle(span);
-	}
-	handle(span) {
-		return {
-			span,
-			end: async (patch) => {
-				const endedAt = this.now();
-				await this.store.updateSpan(span.spanId, {
-					endedAt,
-					status: "ok",
-					...patch
-				});
-				this.pop(span.spanId);
-			},
-			fail: async (error, patch) => {
-				const endedAt = this.now();
-				const errStr = error instanceof Error ? error.message : error;
-				await this.store.updateSpan(span.spanId, {
-					endedAt,
-					status: "error",
-					error: errStr,
-					...patch
-				});
-				this.pop(span.spanId);
-			}
-		};
-	}
-	pop(spanId) {
-		const idx = this.stack.lastIndexOf(spanId);
-		if (idx >= 0) this.stack.splice(idx, 1);
-	}
-	llm(init) {
-		return this.span({
-			kind: "llm",
-			...init
-		});
-	}
-	tool(init) {
-		return this.span({
-			kind: "tool",
-			...init
-		});
-	}
-	retrieval(init) {
-		return this.span({
-			kind: "retrieval",
-			...init
-		});
-	}
-	async recordJudge(verdict) {
-		const spanId = this.id();
-		const now = this.now();
-		const full = {
-			spanId,
-			runId: this._runId,
-			kind: "judge",
-			startedAt: now,
-			endedAt: now,
-			status: "ok",
-			...verdict
-		};
-		await this.store.appendSpan(full);
-		return full;
-	}
-	sandbox(init) {
-		return this.span({
-			kind: "sandbox",
-			...init
-		});
-	}
-	async emit(event) {
-		const full = {
-			eventId: this.id(),
-			runId: this._runId,
-			spanId: event.spanId ?? this.stack[this.stack.length - 1],
-			kind: event.kind,
-			timestamp: this.now(),
-			payload: event.payload ?? {}
-		};
-		await this.store.appendEvent(full);
-		return full;
-	}
-	async recordBudget(entry) {
-		const full = {
-			runId: this._runId,
-			timestamp: entry.timestamp ?? this.now(),
-			dimension: entry.dimension,
-			limit: entry.limit,
-			consumed: entry.consumed,
-			remaining: entry.remaining,
-			breached: entry.breached,
-			spanId: entry.spanId ?? this.stack[this.stack.length - 1]
-		};
-		await this.store.appendBudgetEntry(full);
-		if (full.breached) await this.emit({
-			kind: "budget_breach",
-			spanId: full.spanId,
-			payload: {
-				dimension: full.dimension,
-				limit: full.limit,
-				consumed: full.consumed
-			}
-		});
-		return full;
-	}
-	async recordArtifact(artifact) {
-		const full = {
-			artifactId: this.id(),
-			runId: this._runId,
-			...artifact
-		};
-		await this.store.appendArtifact(full);
-		return full;
-	}
-	/**
-	* Runs `fn` inside a span; auto-ends on success, auto-fails on throw.
-	* Returns the fn's return value. Use this for the 95% case.
-	*/
-	async within(init, fn) {
-		const handle = await this.span(init);
-		try {
-			const result = await fn(handle);
-			await handle.end();
-			return result;
-		} catch (err) {
-			await handle.fail(err instanceof Error ? err : String(err));
-			throw err;
-		}
-	}
-};
-//#endregion
-export { TraceEmitter as t };
-
-//# sourceMappingURL=emitter-DeQHiDMm.js.map

package/dist/emitter-DeQHiDMm.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"emitter-DeQHiDMm.js","names":[],"sources":["../src/trace/emitter.ts"],"sourcesContent":["/**\n * TraceEmitter — hierarchical span builder that auto-parents using an\n * internal stack. One emitter per Run; emitters do NOT share state.\n *\n * Convenience methods (`llm`, `tool`, `retrieval`, `judge`, `sandbox`)\n * return a `SpanHandle` with `.end()` / `.fail()` so callers don't\n * have to thread spanIds manually. For async workflows that can't use\n * the stack (e.g. fan-out parallel calls), pass `parentSpanId`\n * explicitly.\n */\n\nimport { newRecordId } from '../record-id'\nimport type {\n  Artifact,\n  BudgetLedgerEntry,\n  EventKind,\n  JudgeSpan,\n  LlmSpan,\n  RetrievalSpan,\n  Run,\n  RunOutcome,\n  SandboxSpan,\n  Span,\n  SpanKind,\n  ToolSpan,\n  TraceEvent,\n} from './schema'\nimport type { TraceStore } from './store'\n\nexport interface SpanHandle<S extends Span = Span> {\n  span: S\n  end(patch?: Partial<S>): Promise<void>\n  fail(error: string | Error, patch?: Partial<S>): Promise<void>\n}\n\nexport interface RunCompleteHookContext {\n  runId: string\n  emitter: TraceEmitter\n  store: TraceStore\n  /** Outcome the caller passed to `endRun` (undefined for `abortRun`). */\n  outcome?: RunOutcome\n  /** Final run status. */\n  status: 'completed' | 'failed' | 'aborted'\n}\n\nexport type RunCompleteHook = (ctx: RunCompleteHookContext) => Promise<void> | void\n\nexport interface TraceEmitterOptions {\n  runId?: string\n  /** Inject a clock for deterministic tests. */\n  now?: () => number\n  /** Inject an id generator for deterministic tests. */\n  id?: () => string\n  /**\n   * Hooks fired after `endRun` / `abortRun` writes the final run state.\n   * Designed for trace-analyst auto-execution, integrity assertions, and\n   * outbound notifications. Hooks run sequentially in the order supplied.\n   *\n   * By default a hook that throws is swallowed and logged as a `note` event\n   * on the run — auto-orchestration must not crash the underlying flow.\n   * Set `hookErrors: 'throw'` to propagate.\n   */\n  onRunComplete?: RunCompleteHook[]\n  /** `'swallow'` (default) | `'throw'`. */\n  hookErrors?: 'swallow' | 'throw'\n}\n\nexport class TraceEmitter {\n  private store: TraceStore\n  private stack: string[] = []\n  private _runId: string\n  private now: () => number\n  private id: () => string\n  private hooks: RunCompleteHook[]\n  private hookErrors: 'swallow' | 'throw'\n\n  constructor(store: TraceStore, options: TraceEmitterOptions = {}) {\n    this.store = store\n    this.now = options.now ?? (() => Date.now())\n    this.id = options.id ?? (() => newRecordId())\n    this._runId = options.runId ?? this.id()\n    this.hooks = options.onRunComplete ?? []\n    this.hookErrors = options.hookErrors ?? 'swallow'\n  }\n\n  get runId(): string {\n    return this._runId\n  }\n\n  get traceStore(): TraceStore {\n    return this.store\n  }\n\n  /** Append a hook after construction (e.g. attach the trace analyst). */\n  addRunCompleteHook(hook: RunCompleteHook): void {\n    this.hooks.push(hook)\n  }\n\n  // ── Run lifecycle ──────────────────────────────────────────────────\n\n  /**\n   * Begin a Run.\n   *\n   * `scenarioId` is required on the persisted Run shape — every Run downstream\n   * gets a non-empty scenarioId so filters and aggregations stay simple — but\n   * the INPUT here accepts it as optional. When omitted, startRun substitutes\n   * a sensible default (`run.layer ?? run.tags?.['kind'] ?? 'runtime'`) so\n   * runtime / operator / meta-eval runs that have no curated-scenario corpus\n   * to anchor to don't have to invent placeholder strings at the call site.\n   */\n  async startRun(\n    run: Omit<Run, 'runId' | 'scenarioId' | 'startedAt' | 'status'> & { scenarioId?: string },\n  ): Promise<Run> {\n    const scenarioId = run.scenarioId ?? run.layer ?? run.tags?.kind ?? 'runtime'\n    const full: Run = {\n      ...run,\n      scenarioId,\n      runId: this._runId,\n      startedAt: this.now(),\n      status: 'running',\n    }\n    await this.store.appendRun(full)\n    return full\n  }\n\n  async endRun(outcome?: RunOutcome): Promise<void> {\n    const status: 'completed' | 'failed' = outcome?.pass === false ? 'failed' : 'completed'\n    await this.store.updateRun(this._runId, { endedAt: this.now(), status, outcome })\n    await this.runHooks({ runId: this._runId, emitter: this, store: this.store, outcome, status })\n  }\n\n  async abortRun(reason: string): Promise<void> {\n    const outcome = { pass: false, notes: reason }\n    await this.store.updateRun(this._runId, {\n      endedAt: this.now(),\n      status: 'aborted',\n      outcome,\n    })\n    await this.runHooks({\n      runId: this._runId,\n      emitter: this,\n      store: this.store,\n      outcome,\n      status: 'aborted',\n    })\n  }\n\n  private async runHooks(ctx: RunCompleteHookContext): Promise<void> {\n    for (const hook of this.hooks) {\n      try {\n        await hook(ctx)\n      } catch (err) {\n        if (this.hookErrors === 'throw') throw err\n        try {\n          await this.store.appendEvent({\n            eventId: this.id(),\n            runId: this._runId,\n            kind: 'log',\n            timestamp: this.now(),\n            payload: {\n              source: 'run_complete_hook',\n              error: err instanceof Error ? err.message : String(err),\n            },\n          })\n        } catch {\n          // best-effort\n        }\n      }\n    }\n  }\n\n  // ── Generic span ───────────────────────────────────────────────────\n\n  async span<S extends Span = Span>(\n    init: {\n      kind: SpanKind\n      name: string\n      parentSpanId?: string\n      attributes?: Record<string, unknown>\n    } & Partial<Omit<S, 'spanId' | 'runId' | 'startedAt' | 'kind' | 'name'>>,\n  ): Promise<SpanHandle<S>> {\n    const spanId = this.id()\n    const parent = init.parentSpanId ?? this.stack[this.stack.length - 1]\n    const span = {\n      spanId,\n      parentSpanId: parent,\n      runId: this._runId,\n      startedAt: this.now(),\n      ...init,\n    } as unknown as S\n    await this.store.appendSpan(span)\n    this.stack.push(spanId)\n    return this.handle<S>(span)\n  }\n\n  private handle<S extends Span>(span: S): SpanHandle<S> {\n    return {\n      span,\n      end: async (patch?: Partial<S>) => {\n        const endedAt = this.now()\n        await this.store.updateSpan(span.spanId, {\n          endedAt,\n          status: 'ok',\n          ...patch,\n        } as Partial<Span>)\n        this.pop(span.spanId)\n      },\n      fail: async (error: string | Error, patch?: Partial<S>) => {\n        const endedAt = this.now()\n        const errStr = error instanceof Error ? error.message : error\n        await this.store.updateSpan(span.spanId, {\n          endedAt,\n          status: 'error',\n          error: errStr,\n          ...patch,\n        } as Partial<Span>)\n        this.pop(span.spanId)\n      },\n    }\n  }\n\n  private pop(spanId: string): void {\n    const idx = this.stack.lastIndexOf(spanId)\n    if (idx >= 0) this.stack.splice(idx, 1)\n  }\n\n  // ── Typed span conveniences ────────────────────────────────────────\n\n  llm(\n    init: Omit<LlmSpan, 'spanId' | 'runId' | 'kind' | 'startedAt'>,\n  ): Promise<SpanHandle<LlmSpan>> {\n    return this.span<LlmSpan>({ kind: 'llm', ...init })\n  }\n\n  tool(\n    init: Omit<ToolSpan, 'spanId' | 'runId' | 'kind' | 'startedAt'>,\n  ): Promise<SpanHandle<ToolSpan>> {\n    return this.span<ToolSpan>({ kind: 'tool', ...init })\n  }\n\n  retrieval(\n    init: Omit<RetrievalSpan, 'spanId' | 'runId' | 'kind' | 'startedAt'>,\n  ): Promise<SpanHandle<RetrievalSpan>> {\n    return this.span<RetrievalSpan>({ kind: 'retrieval', ...init })\n  }\n\n  async recordJudge(\n    verdict: Omit<JudgeSpan, 'spanId' | 'runId' | 'kind' | 'startedAt' | 'endedAt'>,\n  ): Promise<JudgeSpan> {\n    const spanId = this.id()\n    const now = this.now()\n    const full: JudgeSpan = {\n      spanId,\n      runId: this._runId,\n      kind: 'judge',\n      startedAt: now,\n      endedAt: now,\n      status: 'ok',\n      ...verdict,\n    }\n    await this.store.appendSpan(full)\n    return full\n  }\n\n  sandbox(\n    init: Omit<SandboxSpan, 'spanId' | 'runId' | 'kind' | 'startedAt'>,\n  ): Promise<SpanHandle<SandboxSpan>> {\n    return this.span<SandboxSpan>({ kind: 'sandbox', ...init })\n  }\n\n  // ── Events ─────────────────────────────────────────────────────────\n\n  async emit(event: {\n    kind: EventKind\n    spanId?: string\n    payload?: Record<string, unknown>\n  }): Promise<TraceEvent> {\n    const full: TraceEvent = {\n      eventId: this.id(),\n      runId: this._runId,\n      spanId: event.spanId ?? this.stack[this.stack.length - 1],\n      kind: event.kind,\n      timestamp: this.now(),\n      payload: event.payload ?? {},\n    }\n    await this.store.appendEvent(full)\n    return full\n  }\n\n  // ── Budget ledger ──────────────────────────────────────────────────\n\n  async recordBudget(\n    entry: Omit<BudgetLedgerEntry, 'runId' | 'timestamp'> & { timestamp?: number },\n  ): Promise<BudgetLedgerEntry> {\n    const full: BudgetLedgerEntry = {\n      runId: this._runId,\n      timestamp: entry.timestamp ?? this.now(),\n      dimension: entry.dimension,\n      limit: entry.limit,\n      consumed: entry.consumed,\n      remaining: entry.remaining,\n      breached: entry.breached,\n      spanId: entry.spanId ?? this.stack[this.stack.length - 1],\n    }\n    await this.store.appendBudgetEntry(full)\n    if (full.breached) {\n      await this.emit({\n        kind: 'budget_breach',\n        spanId: full.spanId,\n        payload: { dimension: full.dimension, limit: full.limit, consumed: full.consumed },\n      })\n    }\n    return full\n  }\n\n  // ── Artifacts ──────────────────────────────────────────────────────\n\n  async recordArtifact(artifact: Omit<Artifact, 'artifactId' | 'runId'>): Promise<Artifact> {\n    const full: Artifact = { artifactId: this.id(), runId: this._runId, ...artifact }\n    await this.store.appendArtifact(full)\n    return full\n  }\n\n  // ── Nested composition ─────────────────────────────────────────────\n\n  /**\n   * Runs `fn` inside a span; auto-ends on success, auto-fails on throw.\n   * Returns the fn's return value. Use this for the 95% case.\n   */\n  async within<T>(\n    init: Parameters<TraceEmitter['span']>[0],\n    fn: (handle: SpanHandle) => Promise<T>,\n  ): Promise<T> {\n    const handle = await this.span(init)\n    try {\n      const result = await fn(handle)\n      await handle.end()\n      return result\n    } catch (err) {\n      await handle.fail(err instanceof Error ? err : String(err))\n      throw err\n    }\n  }\n}\n\n// Helpers -------------------------------------------------------------\n"],"mappings":";;;;;;;;;;;;AAmEA,IAAa,eAAb,MAA0B;CACxB;CACA,QAA0B,CAAC;CAC3B;CACA;CACA;CACA;CACA;CAEA,YAAY,OAAmB,UAA+B,CAAC,GAAG;EAChE,KAAK,QAAQ;EACb,KAAK,MAAM,QAAQ,cAAc,KAAK,IAAI;EAC1C,KAAK,KAAK,QAAQ,aAAa,YAAY;EAC3C,KAAK,SAAS,QAAQ,SAAS,KAAK,GAAG;EACvC,KAAK,QAAQ,QAAQ,iBAAiB,CAAC;EACvC,KAAK,aAAa,QAAQ,cAAc;CAC1C;CAEA,IAAI,QAAgB;EAClB,OAAO,KAAK;CACd;CAEA,IAAI,aAAyB;EAC3B,OAAO,KAAK;CACd;;CAGA,mBAAmB,MAA6B;EAC9C,KAAK,MAAM,KAAK,IAAI;CACtB;;;;;;;;;;;CAcA,MAAM,SACJ,KACc;EACd,MAAM,aAAa,IAAI,cAAc,IAAI,SAAS,IAAI,MAAM,QAAQ;EACpE,MAAM,OAAY;GAChB,GAAG;GACH;GACA,OAAO,KAAK;GACZ,WAAW,KAAK,IAAI;GACpB,QAAQ;EACV;EACA,MAAM,KAAK,MAAM,UAAU,IAAI;EAC/B,OAAO;CACT;CAEA,MAAM,OAAO,SAAqC;EAChD,MAAM,SAAiC,SAAS,SAAS,QAAQ,WAAW;EAC5E,MAAM,KAAK,MAAM,UAAU,KAAK,QAAQ;GAAE,SAAS,KAAK,IAAI;GAAG;GAAQ;EAAQ,CAAC;EAChF,MAAM,KAAK,SAAS;GAAE,OAAO,KAAK;GAAQ,SAAS;GAAM,OAAO,KAAK;GAAO;GAAS;EAAO,CAAC;CAC/F;CAEA,MAAM,SAAS,QAA+B;EAC5C,MAAM,UAAU;GAAE,MAAM;GAAO,OAAO;EAAO;EAC7C,MAAM,KAAK,MAAM,UAAU,KAAK,QAAQ;GACtC,SAAS,KAAK,IAAI;GAClB,QAAQ;GACR;EACF,CAAC;EACD,MAAM,KAAK,SAAS;GAClB,OAAO,KAAK;GACZ,SAAS;GACT,OAAO,KAAK;GACZ;GACA,QAAQ;EACV,CAAC;CACH;CAEA,MAAc,SAAS,KAA4C;EACjE,KAAK,MAAM,QAAQ,KAAK,OACtB,IAAI;GACF,MAAM,KAAK,GAAG;EAChB,SAAS,KAAK;GACZ,IAAI,KAAK,eAAe,SAAS,MAAM;GACvC,IAAI;IACF,MAAM,KAAK,MAAM,YAAY;KAC3B,SAAS,KAAK,GAAG;KACjB,OAAO,KAAK;KACZ,MAAM;KACN,WAAW,KAAK,IAAI;KACpB,SAAS;MACP,QAAQ;MACR,OAAO,eAAe,QAAQ,IAAI,UAAU,OAAO,GAAG;KACxD;IACF,CAAC;GACH,QAAQ,CAER;EACF;CAEJ;CAIA,MAAM,KACJ,MAMwB;EACxB,MAAM,SAAS,KAAK,GAAG;EAEvB,MAAM,OAAO;GACX;GACA,cAHa,KAAK,gBAAgB,KAAK,MAAM,KAAK,MAAM,SAAS;GAIjE,OAAO,KAAK;GACZ,WAAW,KAAK,IAAI;GACpB,GAAG;EACL;EACA,MAAM,KAAK,MAAM,WAAW,IAAI;EAChC,KAAK,MAAM,KAAK,MAAM;EACtB,OAAO,KAAK,OAAU,IAAI;CAC5B;CAEA,OAA+B,MAAwB;EACrD,OAAO;GACL;GACA,KAAK,OAAO,UAAuB;IACjC,MAAM,UAAU,KAAK,IAAI;IACzB,MAAM,KAAK,MAAM,WAAW,KAAK,QAAQ;KACvC;KACA,QAAQ;KACR,GAAG;IACL,CAAkB;IAClB,KAAK,IAAI,KAAK,MAAM;GACtB;GACA,MAAM,OAAO,OAAuB,UAAuB;IACzD,MAAM,UAAU,KAAK,IAAI;IACzB,MAAM,SAAS,iBAAiB,QAAQ,MAAM,UAAU;IACxD,MAAM,KAAK,MAAM,WAAW,KAAK,QAAQ;KACvC;KACA,QAAQ;KACR,OAAO;KACP,GAAG;IACL,CAAkB;IAClB,KAAK,IAAI,KAAK,MAAM;GACtB;EACF;CACF;CAEA,IAAY,QAAsB;EAChC,MAAM,MAAM,KAAK,MAAM,YAAY,MAAM;EACzC,IAAI,OAAO,GAAG,KAAK,MAAM,OAAO,KAAK,CAAC;CACxC;CAIA,IACE,MAC8B;EAC9B,OAAO,KAAK,KAAc;GAAE,MAAM;GAAO,GAAG;EAAK,CAAC;CACpD;CAEA,KACE,MAC+B;EAC/B,OAAO,KAAK,KAAe;GAAE,MAAM;GAAQ,GAAG;EAAK,CAAC;CACtD;CAEA,UACE,MACoC;EACpC,OAAO,KAAK,KAAoB;GAAE,MAAM;GAAa,GAAG;EAAK,CAAC;CAChE;CAEA,MAAM,YACJ,SACoB;EACpB,MAAM,SAAS,KAAK,GAAG;EACvB,MAAM,MAAM,KAAK,IAAI;EACrB,MAAM,OAAkB;GACtB;GACA,OAAO,KAAK;GACZ,MAAM;GACN,WAAW;GACX,SAAS;GACT,QAAQ;GACR,GAAG;EACL;EACA,MAAM,KAAK,MAAM,WAAW,IAAI;EAChC,OAAO;CACT;CAEA,QACE,MACkC;EAClC,OAAO,KAAK,KAAkB;GAAE,MAAM;GAAW,GAAG;EAAK,CAAC;CAC5D;CAIA,MAAM,KAAK,OAIa;EACtB,MAAM,OAAmB;GACvB,SAAS,KAAK,GAAG;GACjB,OAAO,KAAK;GACZ,QAAQ,MAAM,UAAU,KAAK,MAAM,KAAK,MAAM,SAAS;GACvD,MAAM,MAAM;GACZ,WAAW,KAAK,IAAI;GACpB,SAAS,MAAM,WAAW,CAAC;EAC7B;EACA,MAAM,KAAK,MAAM,YAAY,IAAI;EACjC,OAAO;CACT;CAIA,MAAM,aACJ,OAC4B;EAC5B,MAAM,OAA0B;GAC9B,OAAO,KAAK;GACZ,WAAW,MAAM,aAAa,KAAK,IAAI;GACvC,WAAW,MAAM;GACjB,OAAO,MAAM;GACb,UAAU,MAAM;GAChB,WAAW,MAAM;GACjB,UAAU,MAAM;GAChB,QAAQ,MAAM,UAAU,KAAK,MAAM,KAAK,MAAM,SAAS;EACzD;EACA,MAAM,KAAK,MAAM,kBAAkB,IAAI;EACvC,IAAI,KAAK,UACP,MAAM,KAAK,KAAK;GACd,MAAM;GACN,QAAQ,KAAK;GACb,SAAS;IAAE,WAAW,KAAK;IAAW,OAAO,KAAK;IAAO,UAAU,KAAK;GAAS;EACnF,CAAC;EAEH,OAAO;CACT;CAIA,MAAM,eAAe,UAAqE;EACxF,MAAM,OAAiB;GAAE,YAAY,KAAK,GAAG;GAAG,OAAO,KAAK;GAAQ,GAAG;EAAS;EAChF,MAAM,KAAK,MAAM,eAAe,IAAI;EACpC,OAAO;CACT;;;;;CAQA,MAAM,OACJ,MACA,IACY;EACZ,MAAM,SAAS,MAAM,KAAK,KAAK,IAAI;EACnC,IAAI;GACF,MAAM,SAAS,MAAM,GAAG,MAAM;GAC9B,MAAM,OAAO,IAAI;GACjB,OAAO;EACT,SAAS,KAAK;GACZ,MAAM,OAAO,KAAK,eAAe,QAAQ,MAAM,OAAO,GAAG,CAAC;GAC1D,MAAM;EACR;CACF;AACF"}

package/dist/engine-DNlKhIm0.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"engine-DNlKhIm0.d.ts","names":[],"sources":["../src/eval-tools.ts","../src/trace-analyst/tools.ts","../src/analyst/finding-signature.ts","../src/analyst/engine.ts"],"mappings":";;;;;;;UA4BiB;EACf;EACA;EACA,YAAY;EACZ,UAAU,eAAe;IAAQ,SAAS;QAAkB;;;;cClBjD;UAEI,uCAAuC;EACtD,OAAO;;UAGQ,oCAAoC;EACnD,kBAAkB;;;iBAIJ,kCACd,SAAS,iCACR;iBAoIa,0BAA0B,SAAS;EACjD;EACA;EACA;EACA;EACA,WAAW;;;;cCxJP,0BAAwB,EAAA;;;GAKnB,EAAA,KAAA;KAEC,qBAAqB,EAAE,aAAa;cAiBnC,yBAAuB,EAAA;;;;;;;;;;;;;;;;;GAKzB,EAAA,KAAA;KAEC,oBAAoB,EAAE,aAAa;;;;;;cAOlC;;iBAYG,2BAA2B,SAAS,oBAAoB;iBAQxD,gBACd,cACA,OAAO,aAAa,SAAS,mCAC5B;;;;UCpEc;;EAEf;;EAEA;;EAEA;;EAEA;;;;;;;;;;;;;;;;cAwBW;UAEI;EACf;EACA;EACA;EACA,gBAAgB;;;;;;;;;;;EAWhB,aAAa,SAAS;EACtB,QAAQ;EACR,YAAY;EACZ;EACA,WAAW;EACX,SAAS;EACT,OAAO,iBAAiB,SAAS;;;UAIlB;EACf;EACA,UAAU;;EAEV;;;;;EAKA;;EAEA;EACA,SAAS;;;;;;;;;UAUM;WACN;WACA;WACA;;WAEA;;;;;;;;WAQA,iBAAiB,SAAS;EACnC,QAAQ,SAAS,6BAA6B,QAAQ;;iBAGxC,0BACd,QAAQ,QAAQ,kCACf"}

package/dist/experiment-tracker-BKEumQug.js

@@ -1,280 +0,0 @@
-import { s as ValidationError } from "./errors-Dngq5h35.js";
-import { n as iqr } from "./baseline-BC-eBZ7U.js";
-import { execSync } from "node:child_process";
-//#region src/experiment-tracker.ts
-/**
-* Experiment tracker — git-provenanced experiment log with N-rep stats and a
-* KEEP / REGRESSION / NOISE verdict against a parent.
-*
-* Every loop the fleet runs reduces to the same question: "I ran the candidate
-* N times — is the median measurably better than the parent, or is the delta
-* inside the noise band?" The hand-rolled copies bake a fixed score scale
-* (percentage points), a fixed store path (`.evolve/experiments-v2.json`), and
-* `execSync('git …')` straight into the module. This is the canonical version:
-* provenance and persistence are injected, thresholds are configurable, and the
-* stats + verdict are pure functions you can unit-test without a git repo or a
-* filesystem.
-*
-* Stats per experiment: median / mean / min / max / iqr / stddev / passRate /
-* n, plus a `stable` flag (`iqr < iqrUnstableAbove && stddev < stddevUnstableAbove`).
-*
-* Verdict against a parent (both must have `n >= minRepsForVerdict`):
-*   - NOISE       — the candidate is too unstable to judge (`!stable`)
-*   - KEEP        — `medianDelta >  keepThreshold`
-*   - REGRESSION  — `medianDelta < -regressionThreshold`
-*   - NOISE       — otherwise (delta inside the band)
-* With no parent (or insufficient reps) the verdict is the neutral ITERATE.
-*/
-const DEFAULTS = {
-	keepThreshold: 5,
-	regressionThreshold: 5,
-	iqrUnstableAbove: 10,
-	stddevUnstableAbove: Number.POSITIVE_INFINITY,
-	minRepsForVerdict: 3
-};
-function resolveThresholds(t) {
-	const r = {
-		...DEFAULTS,
-		...t ?? {}
-	};
-	if (r.keepThreshold < 0) throw new ValidationError(`experiment-tracker: keepThreshold must be >= 0, got ${r.keepThreshold}`);
-	if (r.regressionThreshold < 0) throw new ValidationError(`experiment-tracker: regressionThreshold must be >= 0, got ${r.regressionThreshold}`);
-	if (r.minRepsForVerdict < 1) throw new ValidationError(`experiment-tracker: minRepsForVerdict must be >= 1, got ${r.minRepsForVerdict}`);
-	return r;
-}
-function median(sorted) {
-	const n = sorted.length;
-	if (n === 0) return 0;
-	const mid = Math.floor(n / 2);
-	return n % 2 === 0 ? (sorted[mid - 1] + sorted[mid]) / 2 : sorted[mid];
-}
-/** Population standard deviation (÷n). 0 for fewer than 2 values. */
-function stddev(values, mean) {
-	if (values.length < 2) return 0;
-	const variance = values.reduce((acc, v) => acc + (v - mean) ** 2, 0) / values.length;
-	return Math.sqrt(variance);
-}
-/**
-* Compute the N-rep statistics for a set of reps. Pure — no I/O. The `stable`
-* flag is the trust gate the verdict depends on: a sample whose spread exceeds
-* the configured bounds can't distinguish a real delta from run-to-run noise.
-*/
-function computeExperimentStats(reps, thresholds) {
-	const t = resolveThresholds(thresholds);
-	const n = reps.length;
-	if (n === 0) return {
-		median: 0,
-		mean: 0,
-		min: 0,
-		max: 0,
-		iqr: 0,
-		stddev: 0,
-		passRate: null,
-		n: 0,
-		stable: false
-	};
-	const scores = reps.map((r) => {
-		if (!Number.isFinite(r.score)) throw new ValidationError(`experiment-tracker: rep ${r.rep} has non-finite score ${r.score}`);
-		return r.score;
-	});
-	const sorted = [...scores].sort((a, b) => a - b);
-	const mean = scores.reduce((s, v) => s + v, 0) / n;
-	const sd = stddev(scores, mean);
-	const spread = iqr(scores);
-	const rated = reps.filter((r) => typeof r.passed === "boolean");
-	const passRate = rated.length === 0 ? null : rated.filter((r) => r.passed).length / rated.length;
-	const stable = spread < t.iqrUnstableAbove && sd < t.stddevUnstableAbove;
-	return {
-		median: median(sorted),
-		mean,
-		min: sorted[0],
-		max: sorted[n - 1],
-		iqr: spread,
-		stddev: sd,
-		passRate,
-		n,
-		stable
-	};
-}
-/**
-* Verdict for a candidate against its parent. Pure — operates on already-computed
-* stats. KEEP/REGRESSION require both sides to have `>= minRepsForVerdict` reps
-* AND the candidate to be `stable`; otherwise the result is NOISE (unstable) or
-* ITERATE (not enough reps / no parent).
-*/
-function improvementVerdict(candidate, parent, thresholds) {
-	const t = resolveThresholds(thresholds);
-	if (!parent) return {
-		verdict: "ITERATE",
-		medianDelta: null,
-		reason: "no parent experiment to compare against"
-	};
-	if (candidate.n < t.minRepsForVerdict || parent.n < t.minRepsForVerdict) return {
-		verdict: "ITERATE",
-		medianDelta: null,
-		reason: `need >= ${t.minRepsForVerdict} reps on both sides (candidate n=${candidate.n}, parent n=${parent.n})`
-	};
-	if (!candidate.stable) return {
-		verdict: "NOISE",
-		medianDelta: candidate.median - parent.median,
-		reason: `candidate unstable (iqr=${candidate.iqr}, stddev=${candidate.stddev.toFixed(2)})`
-	};
-	const medianDelta = candidate.median - parent.median;
-	if (medianDelta > t.keepThreshold) return {
-		verdict: "KEEP",
-		medianDelta,
-		reason: `median +${medianDelta} > +${t.keepThreshold}`
-	};
-	if (medianDelta < -t.regressionThreshold) return {
-		verdict: "REGRESSION",
-		medianDelta,
-		reason: `median ${medianDelta} < -${t.regressionThreshold}`
-	};
-	return {
-		verdict: "NOISE",
-		medianDelta,
-		reason: `median delta ${medianDelta} inside noise band [-${t.regressionThreshold}, +${t.keepThreshold}]`
-	};
-}
-/**
-* Default provenance reader: `git rev-parse HEAD`, the subject line, and the
-* files changed vs `HEAD~1`. Fail-loud — a tracker that silently logs
-* `commit: 'unknown'` corrupts the provenance the whole point of the log is to
-* carry. When the working tree genuinely has no parent commit, pass an override.
-*/
-const gitProvenanceReader = () => {
-	const run = (cmd) => execSync(cmd, { encoding: "utf8" }).trim();
-	const commit = run("git rev-parse --short HEAD");
-	const message = run("git log -1 --format=%s");
-	const changedRaw = run("git diff --name-only HEAD~1");
-	return {
-		commit,
-		message,
-		changedFiles: changedRaw.length === 0 ? [] : changedRaw.split("\n").filter(Boolean)
-	};
-};
-/** In-memory store — the default when no persistence is wanted (tests, ephemeral
-*  runs). State lives on the instance. */
-function inMemoryExperimentStore(initial = []) {
-	let state = initial.map((e) => structuredClone(e));
-	return {
-		async load() {
-			return state.map((e) => structuredClone(e));
-		},
-		async save(experiments) {
-			state = experiments.map((e) => structuredClone(e));
-		}
-	};
-}
-/** Filesystem store — a single JSON array at `path`, created on first save. */
-function fileExperimentStore(path) {
-	return {
-		async load() {
-			const fs = await import("node:fs/promises");
-			try {
-				const raw = await fs.readFile(path, "utf8");
-				const parsed = JSON.parse(raw);
-				if (!Array.isArray(parsed)) throw new ValidationError(`experiment-tracker: store at ${path} is not a JSON array`);
-				return parsed;
-			} catch (err) {
-				if (err.code === "ENOENT") return [];
-				throw err;
-			}
-		},
-		async save(experiments) {
-			const fs = await import("node:fs/promises");
-			const pathMod = await import("node:path");
-			await fs.mkdir(pathMod.dirname(path), { recursive: true });
-			await fs.writeFile(path, JSON.stringify(experiments, null, 2), "utf8");
-		}
-	};
-}
-/**
-* Stateful tracker over an `ExperimentStore`. Create an experiment (provenance
-* is captured once), append reps as they complete (stats + verdict recompute on
-* every append), and read the log back for a dashboard. All persistence and git
-* access flow through the injected seams, so the tracker is fully testable
-* without a repo or disk.
-*/
-var ExperimentTracker = class {
-	store;
-	provenanceReader;
-	thresholds;
-	now;
-	constructor(options = {}) {
-		this.store = options.store ?? inMemoryExperimentStore();
-		this.provenanceReader = options.provenanceReader ?? gitProvenanceReader;
-		this.thresholds = resolveThresholds(options.thresholds);
-		this.now = options.now ?? Date.now;
-	}
-	async create(input) {
-		const experiments = await this.store.load();
-		if (experiments.some((e) => e.id === input.id)) throw new ValidationError(`experiment-tracker: experiment id "${input.id}" already exists`);
-		if (input.parentId && !experiments.some((e) => e.id === input.parentId)) throw new ValidationError(`experiment-tracker: parent experiment "${input.parentId}" not found`);
-		const provenance = input.provenance ?? await this.provenanceReader();
-		const experiment = {
-			id: input.id,
-			label: input.label,
-			provenance,
-			parentId: input.parentId,
-			changeSummary: input.changeSummary,
-			reps: [],
-			stats: computeExperimentStats([], this.thresholds),
-			verdict: "ITERATE",
-			createdAt: new Date(this.now()).toISOString()
-		};
-		experiments.push(experiment);
-		await this.store.save(experiments);
-		return structuredClone(experiment);
-	}
-	/** Append a rep (its `rep` index defaults to the current rep count) and
-	*  recompute stats + verdict. Returns the updated experiment. */
-	async addRep(experimentId, rep) {
-		const experiments = await this.store.load();
-		const exp = experiments.find((e) => e.id === experimentId);
-		if (!exp) throw new ValidationError(`experiment-tracker: experiment "${experimentId}" not found`);
-		if (rep.runId !== void 0 && rep.runId.trim().length === 0) throw new ValidationError("experiment-tracker: rep runId must be non-empty when present");
-		const evidence = rep.evidence?.map((reference, index) => {
-			if (reference.uri.trim().length === 0) throw new ValidationError(`experiment-tracker: rep evidence[${index}].uri must be non-empty`);
-			return {
-				kind: reference.kind,
-				uri: reference.uri,
-				...reference.excerpt === void 0 ? {} : { excerpt: reference.excerpt }
-			};
-		});
-		const fullRep = {
-			rep: rep.rep ?? exp.reps.length,
-			score: rep.score,
-			timestamp: rep.timestamp ?? new Date(this.now()).toISOString(),
-			...rep.runId === void 0 ? {} : { runId: rep.runId },
-			...evidence === void 0 ? {} : { evidence },
-			...rep.passed === void 0 ? {} : { passed: rep.passed },
-			...rep.metrics === void 0 ? {} : { metrics: { ...rep.metrics } }
-		};
-		exp.reps.push(fullRep);
-		exp.stats = computeExperimentStats(exp.reps, this.thresholds);
-		const parent = exp.parentId ? experiments.find((e) => e.id === exp.parentId) : void 0;
-		exp.verdict = improvementVerdict(exp.stats, parent?.stats ?? null, this.thresholds).verdict;
-		await this.store.save(experiments);
-		return structuredClone(exp);
-	}
-	async get(experimentId) {
-		const found = (await this.store.load()).find((e) => e.id === experimentId);
-		return found ? structuredClone(found) : void 0;
-	}
-	async list() {
-		return this.store.load();
-	}
-	/** Full verdict (not just the enum) for an experiment vs its parent. */
-	async verdictFor(experimentId) {
-		const experiments = await this.store.load();
-		const exp = experiments.find((e) => e.id === experimentId);
-		if (!exp) throw new ValidationError(`experiment-tracker: experiment "${experimentId}" not found`);
-		const parent = exp.parentId ? experiments.find((e) => e.id === exp.parentId) : void 0;
-		return improvementVerdict(exp.stats, parent?.stats ?? null, this.thresholds);
-	}
-};
-//#endregion
-export { inMemoryExperimentStore as a, improvementVerdict as i, computeExperimentStats as n, fileExperimentStore as r, ExperimentTracker as t };
-
-//# sourceMappingURL=experiment-tracker-BKEumQug.js.map

package/dist/experiment-tracker-BKEumQug.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"experiment-tracker-BKEumQug.js","names":[],"sources":["../src/experiment-tracker.ts"],"sourcesContent":["/**\n * Experiment tracker — git-provenanced experiment log with N-rep stats and a\n * KEEP / REGRESSION / NOISE verdict against a parent.\n *\n * Every loop the fleet runs reduces to the same question: \"I ran the candidate\n * N times — is the median measurably better than the parent, or is the delta\n * inside the noise band?\" The hand-rolled copies bake a fixed score scale\n * (percentage points), a fixed store path (`.evolve/experiments-v2.json`), and\n * `execSync('git …')` straight into the module. This is the canonical version:\n * provenance and persistence are injected, thresholds are configurable, and the\n * stats + verdict are pure functions you can unit-test without a git repo or a\n * filesystem.\n *\n * Stats per experiment: median / mean / min / max / iqr / stddev / passRate /\n * n, plus a `stable` flag (`iqr < iqrUnstableAbove && stddev < stddevUnstableAbove`).\n *\n * Verdict against a parent (both must have `n >= minRepsForVerdict`):\n *   - NOISE       — the candidate is too unstable to judge (`!stable`)\n *   - KEEP        — `medianDelta >  keepThreshold`\n *   - REGRESSION  — `medianDelta < -regressionThreshold`\n *   - NOISE       — otherwise (delta inside the band)\n * With no parent (or insufficient reps) the verdict is the neutral ITERATE.\n */\n\nimport { execSync } from 'node:child_process'\nimport type { EvidenceRef } from './analyst/types'\nimport { iqr } from './baseline'\nimport { ValidationError } from './errors'\n\n/** Verdict for one experiment relative to its parent. ITERATE is the neutral\n *  \"keep collecting reps / no parent to compare against\" state. */\nexport type ExperimentVerdict = 'KEEP' | 'ITERATE' | 'NOISE' | 'REGRESSION'\n\n/** Git provenance for the working tree an experiment was run from. */\nexport interface ExperimentProvenance {\n  /** Commit sha (short or full — the tracker does not interpret it). */\n  commit: string\n  /** First line of the commit message. */\n  message: string\n  /** Files changed vs the parent commit, or a marker like 'uncommitted'. */\n  changedFiles: string[]\n}\n\n/** A single repetition of an experiment, carrying the score the verdict is\n *  computed on plus any free-form per-rep metrics the consumer wants kept. */\nexport interface ExperimentRep {\n  /** 0-indexed repetition number within the experiment. */\n  rep: number\n  /** The score this rep is judged on (same scale as the thresholds). */\n  score: number\n  /** ISO timestamp the rep completed. */\n  timestamp: string\n  /** Stable execution/run identity that produced this score. */\n  runId?: string\n  /** Mechanically resolvable trace, artifact, metric, or finding evidence. */\n  evidence?: EvidenceRef[]\n  /** Whether this rep passed the consumer's own gate — folded into `passRate`. */\n  passed?: boolean\n  /** Free-form numeric metrics retained for later analysis. */\n  metrics?: Record<string, number>\n}\n\nexport interface ExperimentStats {\n  median: number\n  mean: number\n  min: number\n  max: number\n  /** Inter-quartile range of the rep scores. */\n  iqr: number\n  /** Population standard deviation of the rep scores. */\n  stddev: number\n  /** Fraction of reps with `passed === true`, over reps that set `passed`.\n   *  null when no rep declared a pass/fail outcome. */\n  passRate: number | null\n  /** Number of reps. */\n  n: number\n  /** True when the sample is tight enough to trust for a verdict. */\n  stable: boolean\n}\n\nexport interface Experiment {\n  /** Stable id for the experiment. */\n  id: string\n  /** Free-form label / config descriptor. */\n  label: string\n  /** Git provenance captured when the experiment was created. */\n  provenance: ExperimentProvenance\n  /** Parent experiment id this candidate is compared against, if any. */\n  parentId?: string\n  /** One-line summary of what changed from the parent. */\n  changeSummary: string\n  reps: ExperimentRep[]\n  stats: ExperimentStats\n  verdict: ExperimentVerdict\n  /** ISO timestamp the experiment was created. */\n  createdAt: string\n}\n\nexport interface ImprovementThresholds {\n  /** medianDelta strictly above this ⇒ KEEP. Default 5. */\n  keepThreshold?: number\n  /** medianDelta strictly below the negative of this ⇒ REGRESSION. Default 5. */\n  regressionThreshold?: number\n  /** iqr at or above this ⇒ unstable. Default 10. */\n  iqrUnstableAbove?: number\n  /** stddev at or above this ⇒ unstable. Default Infinity (iqr-only stability). */\n  stddevUnstableAbove?: number\n  /** Reps required on BOTH candidate and parent before a verdict is rendered.\n   *  Default 3. */\n  minRepsForVerdict?: number\n}\n\nexport interface ImprovementVerdictResult {\n  verdict: ExperimentVerdict\n  /** candidate.median − parent.median; null when no parent or insufficient reps. */\n  medianDelta: number | null\n  /** Human-readable reason for the verdict — for dashboards and logs. */\n  reason: string\n}\n\nconst DEFAULTS: Required<ImprovementThresholds> = {\n  keepThreshold: 5,\n  regressionThreshold: 5,\n  iqrUnstableAbove: 10,\n  stddevUnstableAbove: Number.POSITIVE_INFINITY,\n  minRepsForVerdict: 3,\n}\n\nfunction resolveThresholds(t: ImprovementThresholds | undefined): Required<ImprovementThresholds> {\n  const r = { ...DEFAULTS, ...(t ?? {}) }\n  if (r.keepThreshold < 0) {\n    throw new ValidationError(\n      `experiment-tracker: keepThreshold must be >= 0, got ${r.keepThreshold}`,\n    )\n  }\n  if (r.regressionThreshold < 0) {\n    throw new ValidationError(\n      `experiment-tracker: regressionThreshold must be >= 0, got ${r.regressionThreshold}`,\n    )\n  }\n  if (r.minRepsForVerdict < 1) {\n    throw new ValidationError(\n      `experiment-tracker: minRepsForVerdict must be >= 1, got ${r.minRepsForVerdict}`,\n    )\n  }\n  return r\n}\n\nfunction median(sorted: number[]): number {\n  const n = sorted.length\n  if (n === 0) return 0\n  const mid = Math.floor(n / 2)\n  return n % 2 === 0 ? (sorted[mid - 1]! + sorted[mid]!) / 2 : sorted[mid]!\n}\n\n/** Population standard deviation (÷n). 0 for fewer than 2 values. */\nfunction stddev(values: number[], mean: number): number {\n  if (values.length < 2) return 0\n  const variance = values.reduce((acc, v) => acc + (v - mean) ** 2, 0) / values.length\n  return Math.sqrt(variance)\n}\n\n/**\n * Compute the N-rep statistics for a set of reps. Pure — no I/O. The `stable`\n * flag is the trust gate the verdict depends on: a sample whose spread exceeds\n * the configured bounds can't distinguish a real delta from run-to-run noise.\n */\nexport function computeExperimentStats(\n  reps: ExperimentRep[],\n  thresholds?: ImprovementThresholds,\n): ExperimentStats {\n  const t = resolveThresholds(thresholds)\n  const n = reps.length\n  if (n === 0) {\n    return {\n      median: 0,\n      mean: 0,\n      min: 0,\n      max: 0,\n      iqr: 0,\n      stddev: 0,\n      passRate: null,\n      n: 0,\n      stable: false,\n    }\n  }\n  const scores = reps.map((r) => {\n    if (!Number.isFinite(r.score)) {\n      throw new ValidationError(`experiment-tracker: rep ${r.rep} has non-finite score ${r.score}`)\n    }\n    return r.score\n  })\n  const sorted = [...scores].sort((a, b) => a - b)\n  const mean = scores.reduce((s, v) => s + v, 0) / n\n  const sd = stddev(scores, mean)\n  const spread = iqr(scores)\n  const rated = reps.filter((r) => typeof r.passed === 'boolean')\n  const passRate = rated.length === 0 ? null : rated.filter((r) => r.passed).length / rated.length\n  const stable = spread < t.iqrUnstableAbove && sd < t.stddevUnstableAbove\n  return {\n    median: median(sorted),\n    mean,\n    min: sorted[0]!,\n    max: sorted[n - 1]!,\n    iqr: spread,\n    stddev: sd,\n    passRate,\n    n,\n    stable,\n  }\n}\n\n/**\n * Verdict for a candidate against its parent. Pure — operates on already-computed\n * stats. KEEP/REGRESSION require both sides to have `>= minRepsForVerdict` reps\n * AND the candidate to be `stable`; otherwise the result is NOISE (unstable) or\n * ITERATE (not enough reps / no parent).\n */\nexport function improvementVerdict(\n  candidate: ExperimentStats,\n  parent: ExperimentStats | null,\n  thresholds?: ImprovementThresholds,\n): ImprovementVerdictResult {\n  const t = resolveThresholds(thresholds)\n  if (!parent) {\n    return {\n      verdict: 'ITERATE',\n      medianDelta: null,\n      reason: 'no parent experiment to compare against',\n    }\n  }\n  if (candidate.n < t.minRepsForVerdict || parent.n < t.minRepsForVerdict) {\n    return {\n      verdict: 'ITERATE',\n      medianDelta: null,\n      reason: `need >= ${t.minRepsForVerdict} reps on both sides (candidate n=${candidate.n}, parent n=${parent.n})`,\n    }\n  }\n  if (!candidate.stable) {\n    return {\n      verdict: 'NOISE',\n      medianDelta: candidate.median - parent.median,\n      reason: `candidate unstable (iqr=${candidate.iqr}, stddev=${candidate.stddev.toFixed(2)})`,\n    }\n  }\n  const medianDelta = candidate.median - parent.median\n  if (medianDelta > t.keepThreshold) {\n    return { verdict: 'KEEP', medianDelta, reason: `median +${medianDelta} > +${t.keepThreshold}` }\n  }\n  if (medianDelta < -t.regressionThreshold) {\n    return {\n      verdict: 'REGRESSION',\n      medianDelta,\n      reason: `median ${medianDelta} < -${t.regressionThreshold}`,\n    }\n  }\n  return {\n    verdict: 'NOISE',\n    medianDelta,\n    reason: `median delta ${medianDelta} inside noise band [-${t.regressionThreshold}, +${t.keepThreshold}]`,\n  }\n}\n\n// ── Provenance + persistence seams ───────────────────────────────────\n\n/** Reads git provenance for the working tree. Inject a fake in tests; the\n *  default implementation shells out to `git`. */\nexport type ProvenanceReader = () => ExperimentProvenance | Promise<ExperimentProvenance>\n\n/** Persistence seam for the experiment log. Inject in-memory in tests; the\n *  filesystem implementation is `fileExperimentStore`. */\nexport interface ExperimentStore {\n  load(): Promise<Experiment[]>\n  save(experiments: Experiment[]): Promise<void>\n}\n\n/**\n * Default provenance reader: `git rev-parse HEAD`, the subject line, and the\n * files changed vs `HEAD~1`. Fail-loud — a tracker that silently logs\n * `commit: 'unknown'` corrupts the provenance the whole point of the log is to\n * carry. When the working tree genuinely has no parent commit, pass an override.\n */\nexport const gitProvenanceReader: ProvenanceReader = () => {\n  const run = (cmd: string): string => execSync(cmd, { encoding: 'utf8' }).trim()\n  const commit = run('git rev-parse --short HEAD')\n  const message = run('git log -1 --format=%s')\n  const changedRaw = run('git diff --name-only HEAD~1')\n  const changedFiles = changedRaw.length === 0 ? [] : changedRaw.split('\\n').filter(Boolean)\n  return { commit, message, changedFiles }\n}\n\n/** In-memory store — the default when no persistence is wanted (tests, ephemeral\n *  runs). State lives on the instance. */\nexport function inMemoryExperimentStore(initial: Experiment[] = []): ExperimentStore {\n  let state = initial.map((e) => structuredClone(e))\n  return {\n    async load() {\n      return state.map((e) => structuredClone(e))\n    },\n    async save(experiments) {\n      state = experiments.map((e) => structuredClone(e))\n    },\n  }\n}\n\n/** Filesystem store — a single JSON array at `path`, created on first save. */\nexport function fileExperimentStore(path: string): ExperimentStore {\n  return {\n    async load() {\n      const fs = await import('node:fs/promises')\n      try {\n        const raw = await fs.readFile(path, 'utf8')\n        const parsed = JSON.parse(raw)\n        if (!Array.isArray(parsed)) {\n          throw new ValidationError(`experiment-tracker: store at ${path} is not a JSON array`)\n        }\n        return parsed as Experiment[]\n      } catch (err) {\n        if ((err as NodeJS.ErrnoException).code === 'ENOENT') return []\n        throw err\n      }\n    },\n    async save(experiments) {\n      const fs = await import('node:fs/promises')\n      const pathMod = await import('node:path')\n      await fs.mkdir(pathMod.dirname(path), { recursive: true })\n      await fs.writeFile(path, JSON.stringify(experiments, null, 2), 'utf8')\n    },\n  }\n}\n\nexport interface ExperimentTrackerOptions {\n  store?: ExperimentStore\n  provenanceReader?: ProvenanceReader\n  thresholds?: ImprovementThresholds\n  /** Clock seam for deterministic timestamps in tests. Default `Date.now`. */\n  now?: () => number\n}\n\nexport interface CreateExperimentInput {\n  id: string\n  label: string\n  changeSummary: string\n  parentId?: string\n  /** Override provenance instead of reading from git (e.g. CI metadata). */\n  provenance?: ExperimentProvenance\n}\n\n/**\n * Stateful tracker over an `ExperimentStore`. Create an experiment (provenance\n * is captured once), append reps as they complete (stats + verdict recompute on\n * every append), and read the log back for a dashboard. All persistence and git\n * access flow through the injected seams, so the tracker is fully testable\n * without a repo or disk.\n */\nexport class ExperimentTracker {\n  private readonly store: ExperimentStore\n  private readonly provenanceReader: ProvenanceReader\n  private readonly thresholds: Required<ImprovementThresholds>\n  private readonly now: () => number\n\n  constructor(options: ExperimentTrackerOptions = {}) {\n    this.store = options.store ?? inMemoryExperimentStore()\n    this.provenanceReader = options.provenanceReader ?? gitProvenanceReader\n    this.thresholds = resolveThresholds(options.thresholds)\n    this.now = options.now ?? Date.now\n  }\n\n  async create(input: CreateExperimentInput): Promise<Experiment> {\n    const experiments = await this.store.load()\n    if (experiments.some((e) => e.id === input.id)) {\n      throw new ValidationError(`experiment-tracker: experiment id \"${input.id}\" already exists`)\n    }\n    if (input.parentId && !experiments.some((e) => e.id === input.parentId)) {\n      throw new ValidationError(\n        `experiment-tracker: parent experiment \"${input.parentId}\" not found`,\n      )\n    }\n    const provenance = input.provenance ?? (await this.provenanceReader())\n    const experiment: Experiment = {\n      id: input.id,\n      label: input.label,\n      provenance,\n      parentId: input.parentId,\n      changeSummary: input.changeSummary,\n      reps: [],\n      stats: computeExperimentStats([], this.thresholds),\n      verdict: 'ITERATE',\n      createdAt: new Date(this.now()).toISOString(),\n    }\n    experiments.push(experiment)\n    await this.store.save(experiments)\n    return structuredClone(experiment)\n  }\n\n  /** Append a rep (its `rep` index defaults to the current rep count) and\n   *  recompute stats + verdict. Returns the updated experiment. */\n  async addRep(\n    experimentId: string,\n    rep: Omit<ExperimentRep, 'rep' | 'timestamp'> & { rep?: number; timestamp?: string },\n  ): Promise<Experiment> {\n    const experiments = await this.store.load()\n    const exp = experiments.find((e) => e.id === experimentId)\n    if (!exp)\n      throw new ValidationError(`experiment-tracker: experiment \"${experimentId}\" not found`)\n    if (rep.runId !== undefined && rep.runId.trim().length === 0) {\n      throw new ValidationError('experiment-tracker: rep runId must be non-empty when present')\n    }\n    const evidence = rep.evidence?.map((reference, index) => {\n      if (reference.uri.trim().length === 0) {\n        throw new ValidationError(\n          `experiment-tracker: rep evidence[${index}].uri must be non-empty`,\n        )\n      }\n      return {\n        kind: reference.kind,\n        uri: reference.uri,\n        ...(reference.excerpt === undefined ? {} : { excerpt: reference.excerpt }),\n      }\n    })\n    const fullRep: ExperimentRep = {\n      rep: rep.rep ?? exp.reps.length,\n      score: rep.score,\n      timestamp: rep.timestamp ?? new Date(this.now()).toISOString(),\n      ...(rep.runId === undefined ? {} : { runId: rep.runId }),\n      ...(evidence === undefined ? {} : { evidence }),\n      ...(rep.passed === undefined ? {} : { passed: rep.passed }),\n      ...(rep.metrics === undefined ? {} : { metrics: { ...rep.metrics } }),\n    }\n    exp.reps.push(fullRep)\n    exp.stats = computeExperimentStats(exp.reps, this.thresholds)\n    const parent = exp.parentId ? experiments.find((e) => e.id === exp.parentId) : undefined\n    exp.verdict = improvementVerdict(exp.stats, parent?.stats ?? null, this.thresholds).verdict\n    await this.store.save(experiments)\n    return structuredClone(exp)\n  }\n\n  async get(experimentId: string): Promise<Experiment | undefined> {\n    const experiments = await this.store.load()\n    const found = experiments.find((e) => e.id === experimentId)\n    return found ? structuredClone(found) : undefined\n  }\n\n  async list(): Promise<Experiment[]> {\n    return this.store.load()\n  }\n\n  /** Full verdict (not just the enum) for an experiment vs its parent. */\n  async verdictFor(experimentId: string): Promise<ImprovementVerdictResult> {\n    const experiments = await this.store.load()\n    const exp = experiments.find((e) => e.id === experimentId)\n    if (!exp)\n      throw new ValidationError(`experiment-tracker: experiment \"${experimentId}\" not found`)\n    const parent = exp.parentId ? experiments.find((e) => e.id === exp.parentId) : undefined\n    return improvementVerdict(exp.stats, parent?.stats ?? null, this.thresholds)\n  }\n}\n"],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;AAwHA,MAAM,WAA4C;CAChD,eAAe;CACf,qBAAqB;CACrB,kBAAkB;CAClB,qBAAqB,OAAO;CAC5B,mBAAmB;AACrB;AAEA,SAAS,kBAAkB,GAAuE;CAChG,MAAM,IAAI;EAAE,GAAG;EAAU,GAAI,KAAK,CAAC;CAAG;CACtC,IAAI,EAAE,gBAAgB,GACpB,MAAM,IAAI,gBACR,uDAAuD,EAAE,eAC3D;CAEF,IAAI,EAAE,sBAAsB,GAC1B,MAAM,IAAI,gBACR,6DAA6D,EAAE,qBACjE;CAEF,IAAI,EAAE,oBAAoB,GACxB,MAAM,IAAI,gBACR,2DAA2D,EAAE,mBAC/D;CAEF,OAAO;AACT;AAEA,SAAS,OAAO,QAA0B;CACxC,MAAM,IAAI,OAAO;CACjB,IAAI,MAAM,GAAG,OAAO;CACpB,MAAM,MAAM,KAAK,MAAM,IAAI,CAAC;CAC5B,OAAO,IAAI,MAAM,KAAK,OAAO,MAAM,KAAM,OAAO,QAAS,IAAI,OAAO;AACtE;;AAGA,SAAS,OAAO,QAAkB,MAAsB;CACtD,IAAI,OAAO,SAAS,GAAG,OAAO;CAC9B,MAAM,WAAW,OAAO,QAAQ,KAAK,MAAM,OAAO,IAAI,SAAS,GAAG,CAAC,IAAI,OAAO;CAC9E,OAAO,KAAK,KAAK,QAAQ;AAC3B;;;;;;AAOA,SAAgB,uBACd,MACA,YACiB;CACjB,MAAM,IAAI,kBAAkB,UAAU;CACtC,MAAM,IAAI,KAAK;CACf,IAAI,MAAM,GACR,OAAO;EACL,QAAQ;EACR,MAAM;EACN,KAAK;EACL,KAAK;EACL,KAAK;EACL,QAAQ;EACR,UAAU;EACV,GAAG;EACH,QAAQ;CACV;CAEF,MAAM,SAAS,KAAK,KAAK,MAAM;EAC7B,IAAI,CAAC,OAAO,SAAS,EAAE,KAAK,GAC1B,MAAM,IAAI,gBAAgB,2BAA2B,EAAE,IAAI,wBAAwB,EAAE,OAAO;EAE9F,OAAO,EAAE;CACX,CAAC;CACD,MAAM,SAAS,CAAC,GAAG,MAAM,CAAC,CAAC,MAAM,GAAG,MAAM,IAAI,CAAC;CAC/C,MAAM,OAAO,OAAO,QAAQ,GAAG,MAAM,IAAI,GAAG,CAAC,IAAI;CACjD,MAAM,KAAK,OAAO,QAAQ,IAAI;CAC9B,MAAM,SAAS,IAAI,MAAM;CACzB,MAAM,QAAQ,KAAK,QAAQ,MAAM,OAAO,EAAE,WAAW,SAAS;CAC9D,MAAM,WAAW,MAAM,WAAW,IAAI,OAAO,MAAM,QAAQ,MAAM,EAAE,MAAM,CAAC,CAAC,SAAS,MAAM;CAC1F,MAAM,SAAS,SAAS,EAAE,oBAAoB,KAAK,EAAE;CACrD,OAAO;EACL,QAAQ,OAAO,MAAM;EACrB;EACA,KAAK,OAAO;EACZ,KAAK,OAAO,IAAI;EAChB,KAAK;EACL,QAAQ;EACR;EACA;EACA;CACF;AACF;;;;;;;AAQA,SAAgB,mBACd,WACA,QACA,YAC0B;CAC1B,MAAM,IAAI,kBAAkB,UAAU;CACtC,IAAI,CAAC,QACH,OAAO;EACL,SAAS;EACT,aAAa;EACb,QAAQ;CACV;CAEF,IAAI,UAAU,IAAI,EAAE,qBAAqB,OAAO,IAAI,EAAE,mBACpD,OAAO;EACL,SAAS;EACT,aAAa;EACb,QAAQ,WAAW,EAAE,kBAAkB,mCAAmC,UAAU,EAAE,aAAa,OAAO,EAAE;CAC9G;CAEF,IAAI,CAAC,UAAU,QACb,OAAO;EACL,SAAS;EACT,aAAa,UAAU,SAAS,OAAO;EACvC,QAAQ,2BAA2B,UAAU,IAAI,WAAW,UAAU,OAAO,QAAQ,CAAC,EAAE;CAC1F;CAEF,MAAM,cAAc,UAAU,SAAS,OAAO;CAC9C,IAAI,cAAc,EAAE,eAClB,OAAO;EAAE,SAAS;EAAQ;EAAa,QAAQ,WAAW,YAAY,MAAM,EAAE;CAAgB;CAEhG,IAAI,cAAc,CAAC,EAAE,qBACnB,OAAO;EACL,SAAS;EACT;EACA,QAAQ,UAAU,YAAY,MAAM,EAAE;CACxC;CAEF,OAAO;EACL,SAAS;EACT;EACA,QAAQ,gBAAgB,YAAY,uBAAuB,EAAE,oBAAoB,KAAK,EAAE,cAAc;CACxG;AACF;;;;;;;AAqBA,MAAa,4BAA8C;CACzD,MAAM,OAAO,QAAwB,SAAS,KAAK,EAAE,UAAU,OAAO,CAAC,CAAC,CAAC,KAAK;CAC9E,MAAM,SAAS,IAAI,4BAA4B;CAC/C,MAAM,UAAU,IAAI,wBAAwB;CAC5C,MAAM,aAAa,IAAI,6BAA6B;CAEpD,OAAO;EAAE;EAAQ;EAAS,cADL,WAAW,WAAW,IAAI,CAAC,IAAI,WAAW,MAAM,IAAI,CAAC,CAAC,OAAO,OAAO;CAClD;AACzC;;;AAIA,SAAgB,wBAAwB,UAAwB,CAAC,GAAoB;CACnF,IAAI,QAAQ,QAAQ,KAAK,MAAM,gBAAgB,CAAC,CAAC;CACjD,OAAO;EACL,MAAM,OAAO;GACX,OAAO,MAAM,KAAK,MAAM,gBAAgB,CAAC,CAAC;EAC5C;EACA,MAAM,KAAK,aAAa;GACtB,QAAQ,YAAY,KAAK,MAAM,gBAAgB,CAAC,CAAC;EACnD;CACF;AACF;;AAGA,SAAgB,oBAAoB,MAA+B;CACjE,OAAO;EACL,MAAM,OAAO;GACX,MAAM,KAAK,MAAM,OAAO;GACxB,IAAI;IACF,MAAM,MAAM,MAAM,GAAG,SAAS,MAAM,MAAM;IAC1C,MAAM,SAAS,KAAK,MAAM,GAAG;IAC7B,IAAI,CAAC,MAAM,QAAQ,MAAM,GACvB,MAAM,IAAI,gBAAgB,gCAAgC,KAAK,qBAAqB;IAEtF,OAAO;GACT,SAAS,KAAK;IACZ,IAAK,IAA8B,SAAS,UAAU,OAAO,CAAC;IAC9D,MAAM;GACR;EACF;EACA,MAAM,KAAK,aAAa;GACtB,MAAM,KAAK,MAAM,OAAO;GACxB,MAAM,UAAU,MAAM,OAAO;GAC7B,MAAM,GAAG,MAAM,QAAQ,QAAQ,IAAI,GAAG,EAAE,WAAW,KAAK,CAAC;GACzD,MAAM,GAAG,UAAU,MAAM,KAAK,UAAU,aAAa,MAAM,CAAC,GAAG,MAAM;EACvE;CACF;AACF;;;;;;;;AA0BA,IAAa,oBAAb,MAA+B;CAC7B;CACA;CACA;CACA;CAEA,YAAY,UAAoC,CAAC,GAAG;EAClD,KAAK,QAAQ,QAAQ,SAAS,wBAAwB;EACtD,KAAK,mBAAmB,QAAQ,oBAAoB;EACpD,KAAK,aAAa,kBAAkB,QAAQ,UAAU;EACtD,KAAK,MAAM,QAAQ,OAAO,KAAK;CACjC;CAEA,MAAM,OAAO,OAAmD;EAC9D,MAAM,cAAc,MAAM,KAAK,MAAM,KAAK;EAC1C,IAAI,YAAY,MAAM,MAAM,EAAE,OAAO,MAAM,EAAE,GAC3C,MAAM,IAAI,gBAAgB,sCAAsC,MAAM,GAAG,iBAAiB;EAE5F,IAAI,MAAM,YAAY,CAAC,YAAY,MAAM,MAAM,EAAE,OAAO,MAAM,QAAQ,GACpE,MAAM,IAAI,gBACR,0CAA0C,MAAM,SAAS,YAC3D;EAEF,MAAM,aAAa,MAAM,cAAe,MAAM,KAAK,iBAAiB;EACpE,MAAM,aAAyB;GAC7B,IAAI,MAAM;GACV,OAAO,MAAM;GACb;GACA,UAAU,MAAM;GAChB,eAAe,MAAM;GACrB,MAAM,CAAC;GACP,OAAO,uBAAuB,CAAC,GAAG,KAAK,UAAU;GACjD,SAAS;GACT,WAAW,IAAI,KAAK,KAAK,IAAI,CAAC,CAAC,CAAC,YAAY;EAC9C;EACA,YAAY,KAAK,UAAU;EAC3B,MAAM,KAAK,MAAM,KAAK,WAAW;EACjC,OAAO,gBAAgB,UAAU;CACnC;;;CAIA,MAAM,OACJ,cACA,KACqB;EACrB,MAAM,cAAc,MAAM,KAAK,MAAM,KAAK;EAC1C,MAAM,MAAM,YAAY,MAAM,MAAM,EAAE,OAAO,YAAY;EACzD,IAAI,CAAC,KACH,MAAM,IAAI,gBAAgB,mCAAmC,aAAa,YAAY;EACxF,IAAI,IAAI,UAAU,KAAA,KAAa,IAAI,MAAM,KAAK,CAAC,CAAC,WAAW,GACzD,MAAM,IAAI,gBAAgB,8DAA8D;EAE1F,MAAM,WAAW,IAAI,UAAU,KAAK,WAAW,UAAU;GACvD,IAAI,UAAU,IAAI,KAAK,CAAC,CAAC,WAAW,GAClC,MAAM,IAAI,gBACR,oCAAoC,MAAM,wBAC5C;GAEF,OAAO;IACL,MAAM,UAAU;IAChB,KAAK,UAAU;IACf,GAAI,UAAU,YAAY,KAAA,IAAY,CAAC,IAAI,EAAE,SAAS,UAAU,QAAQ;GAC1E;EACF,CAAC;EACD,MAAM,UAAyB;GAC7B,KAAK,IAAI,OAAO,IAAI,KAAK;GACzB,OAAO,IAAI;GACX,WAAW,IAAI,aAAa,IAAI,KAAK,KAAK,IAAI,CAAC,CAAC,CAAC,YAAY;GAC7D,GAAI,IAAI,UAAU,KAAA,IAAY,CAAC,IAAI,EAAE,OAAO,IAAI,MAAM;GACtD,GAAI,aAAa,KAAA,IAAY,CAAC,IAAI,EAAE,SAAS;GAC7C,GAAI,IAAI,WAAW,KAAA,IAAY,CAAC,IAAI,EAAE,QAAQ,IAAI,OAAO;GACzD,GAAI,IAAI,YAAY,KAAA,IAAY,CAAC,IAAI,EAAE,SAAS,EAAE,GAAG,IAAI,QAAQ,EAAE;EACrE;EACA,IAAI,KAAK,KAAK,OAAO;EACrB,IAAI,QAAQ,uBAAuB,IAAI,MAAM,KAAK,UAAU;EAC5D,MAAM,SAAS,IAAI,WAAW,YAAY,MAAM,MAAM,EAAE,OAAO,IAAI,QAAQ,IAAI,KAAA;EAC/E,IAAI,UAAU,mBAAmB,IAAI,OAAO,QAAQ,SAAS,MAAM,KAAK,UAAU,CAAC,CAAC;EACpF,MAAM,KAAK,MAAM,KAAK,WAAW;EACjC,OAAO,gBAAgB,GAAG;CAC5B;CAEA,MAAM,IAAI,cAAuD;EAE/D,MAAM,SAAQ,MADY,KAAK,MAAM,KAAK,EAAA,CAChB,MAAM,MAAM,EAAE,OAAO,YAAY;EAC3D,OAAO,QAAQ,gBAAgB,KAAK,IAAI,KAAA;CAC1C;CAEA,MAAM,OAA8B;EAClC,OAAO,KAAK,MAAM,KAAK;CACzB;;CAGA,MAAM,WAAW,cAAyD;EACxE,MAAM,cAAc,MAAM,KAAK,MAAM,KAAK;EAC1C,MAAM,MAAM,YAAY,MAAM,MAAM,EAAE,OAAO,YAAY;EACzD,IAAI,CAAC,KACH,MAAM,IAAI,gBAAgB,mCAAmC,aAAa,YAAY;EACxF,MAAM,SAAS,IAAI,WAAW,YAAY,MAAM,MAAM,EAAE,OAAO,IAAI,QAAQ,IAAI,KAAA;EAC/E,OAAO,mBAAmB,IAAI,OAAO,QAAQ,SAAS,MAAM,KAAK,UAAU;CAC7E;AACF"}