@tangle-network/agent-eval 0.221.3 → 0.221.6

package/CHANGELOG.md

@@ -4,6 +4,165 @@ All notable changes to `@tangle-network/agent-eval` and its sibling `agent-eval-
 
 ---
 
+## [0.221.6] — 2026-10-25
+
+- feat: own judge-panel calibration, multishot conversation legs and served-panel identity
+- chore(release): 0.221.5
+- ci: retain private adopter failure evidence (#1106)
+- test(process): isolate process-tree fixture environment
+- fix(adopt): invoke each resolved Playwright CLI (#1096)
+- test(systemone): close disposable provider connections
+- chore(release): 0.221.4
+- test(adoption): inherit the maintained integration deadline
+- fix(adoption): scope guard allowance to disposable Git fixtures
+- fix(analyst): carry root-cause target into failure-mode rubric
+- ci: check locally and in publication paths
+- test(adopt): give real-git fixtures their own commit identity (#1101)
+- chore(release): 0.221.3 (#1099)
+- fix(analyst): repin the benchmark implementation digest after #1097 (#1098)
+- fix(trace-analyst): build OTLP stores from bytes, not one joined string
+- fix(adopt): prepare caller-locked Chromium before gates
+- test(adopt): retain admitted identity in real Git fixtures
+- fix(adopt): retain the sparse caller entrypoint
+- fix(adopt): settle pushed proposal reads against remote refs
+- chore(release): 0.221.2
+- Return plain analyst findings from decision replay
+- chore(release): 0.221.1 (#1091)
+- fix(deps): pin zod 4.6.6 (GHSA-mrhx-c7xc-m3w5) (#1088)
+- Give the disposable adoption fixture its own Git author
+- Fix real adoption drift fixture under the declared package manager
+- Capture supporting trace events in the original native decision
+- test: inherit the integration default for behavioral questions
+- test: set the integration default from the observed loaded tail
+- chore(release): 0.221.0
+- test: own the measured integration deadline in Vitest config
+- test: await observed integration completion across tool paths
+- feat: preserve finding composition and assigned arm populations
+- test: await decision-point HTTP and ledger completion
+- test: await durable readiness in filesystem integration cases
+- test(process): wait for argv child readiness before abort
+- feat(analyst): add bounded calibrated trace question observations
+- fix(analyst): reject surrounding whitespace in caller gold ids
+- test(analyst): reproduce whitespace source-id bypass
+- test(optimizer): retire fake CLI metering harness
+- test(analyst): isolate recorded gold admission scenarios
+- fix(systemone): validate decision boundaries before routing
+- ci: pin the eval publication dispatcher body
+- fix(analyst): reserve excluded source ids across caller gold
+- fix(analyst): validate root provenance before gold admission
+- fix(analyst): preserve actor boundaries in AgentRx calibration gold
+- Preserve manual drafts observed while recording failed adoption (#1076)
+- Recover updater-owned failure drafts while preserving manual holds (#1075)
+- feat(systemone): persist calibrated decision points and outcomes
+- ci: move eval verification and Python publishing to owned runners
+- chore(deps): bump opentelemetry-api in /clients/python (#1068)
+- fix(adopt): quote optional workflow flags
+- fix(adopt): preserve shallow history and caller-owned dependency references
+- fix(adopt): reconcile effective workspace pins and preserve proposal history
+
+## [0.221.5] — 2026-10-25
+
+- ci: retain private adopter failure evidence (#1106)
+- test(process): isolate process-tree fixture environment
+- fix(adopt): invoke each resolved Playwright CLI (#1096)
+- test(systemone): close disposable provider connections
+- chore(release): 0.221.4
+- test(adoption): inherit the maintained integration deadline
+- fix(adoption): scope guard allowance to disposable Git fixtures
+- fix(analyst): carry root-cause target into failure-mode rubric
+- ci: check locally and in publication paths
+- test(adopt): give real-git fixtures their own commit identity (#1101)
+- chore(release): 0.221.3 (#1099)
+- fix(analyst): repin the benchmark implementation digest after #1097 (#1098)
+- fix(trace-analyst): build OTLP stores from bytes, not one joined string
+- fix(adopt): prepare caller-locked Chromium before gates
+- test(adopt): retain admitted identity in real Git fixtures
+- fix(adopt): retain the sparse caller entrypoint
+- fix(adopt): settle pushed proposal reads against remote refs
+- chore(release): 0.221.2
+- Return plain analyst findings from decision replay
+- chore(release): 0.221.1 (#1091)
+- fix(deps): pin zod 4.6.6 (GHSA-mrhx-c7xc-m3w5) (#1088)
+- Give the disposable adoption fixture its own Git author
+- Fix real adoption drift fixture under the declared package manager
+- Capture supporting trace events in the original native decision
+- test: inherit the integration default for behavioral questions
+- test: set the integration default from the observed loaded tail
+- chore(release): 0.221.0
+- test: own the measured integration deadline in Vitest config
+- test: await observed integration completion across tool paths
+- feat: preserve finding composition and assigned arm populations
+- test: await decision-point HTTP and ledger completion
+- test: await durable readiness in filesystem integration cases
+- test(process): wait for argv child readiness before abort
+- feat(analyst): add bounded calibrated trace question observations
+- fix(analyst): reject surrounding whitespace in caller gold ids
+- test(analyst): reproduce whitespace source-id bypass
+- test(optimizer): retire fake CLI metering harness
+- test(analyst): isolate recorded gold admission scenarios
+- fix(systemone): validate decision boundaries before routing
+- ci: pin the eval publication dispatcher body
+- fix(analyst): reserve excluded source ids across caller gold
+- fix(analyst): validate root provenance before gold admission
+- fix(analyst): preserve actor boundaries in AgentRx calibration gold
+- Preserve manual drafts observed while recording failed adoption (#1076)
+- Recover updater-owned failure drafts while preserving manual holds (#1075)
+- feat(systemone): persist calibrated decision points and outcomes
+- ci: move eval verification and Python publishing to owned runners
+- chore(deps): bump opentelemetry-api in /clients/python (#1068)
+- fix(adopt): quote optional workflow flags
+- fix(adopt): preserve shallow history and caller-owned dependency references
+- fix(adopt): reconcile effective workspace pins and preserve proposal history
+
+## [0.221.4] — 2026-10-25
+
+- test(adoption): inherit the maintained integration deadline
+- fix(adoption): scope guard allowance to disposable Git fixtures
+- fix(analyst): carry root-cause target into failure-mode rubric
+- ci: check locally and in publication paths
+- test(adopt): give real-git fixtures their own commit identity (#1101)
+- chore(release): 0.221.3 (#1099)
+- fix(analyst): repin the benchmark implementation digest after #1097 (#1098)
+- fix(trace-analyst): build OTLP stores from bytes, not one joined string
+- fix(adopt): prepare caller-locked Chromium before gates
+- test(adopt): retain admitted identity in real Git fixtures
+- fix(adopt): retain the sparse caller entrypoint
+- fix(adopt): settle pushed proposal reads against remote refs
+- chore(release): 0.221.2
+- Return plain analyst findings from decision replay
+- chore(release): 0.221.1 (#1091)
+- fix(deps): pin zod 4.6.6 (GHSA-mrhx-c7xc-m3w5) (#1088)
+- Give the disposable adoption fixture its own Git author
+- Fix real adoption drift fixture under the declared package manager
+- Capture supporting trace events in the original native decision
+- test: inherit the integration default for behavioral questions
+- test: set the integration default from the observed loaded tail
+- chore(release): 0.221.0
+- test: own the measured integration deadline in Vitest config
+- test: await observed integration completion across tool paths
+- feat: preserve finding composition and assigned arm populations
+- test: await decision-point HTTP and ledger completion
+- test: await durable readiness in filesystem integration cases
+- test(process): wait for argv child readiness before abort
+- feat(analyst): add bounded calibrated trace question observations
+- fix(analyst): reject surrounding whitespace in caller gold ids
+- test(analyst): reproduce whitespace source-id bypass
+- test(optimizer): retire fake CLI metering harness
+- test(analyst): isolate recorded gold admission scenarios
+- fix(systemone): validate decision boundaries before routing
+- ci: pin the eval publication dispatcher body
+- fix(analyst): reserve excluded source ids across caller gold
+- fix(analyst): validate root provenance before gold admission
+- fix(analyst): preserve actor boundaries in AgentRx calibration gold
+- Preserve manual drafts observed while recording failed adoption (#1076)
+- Recover updater-owned failure drafts while preserving manual holds (#1075)
+- feat(systemone): persist calibrated decision points and outcomes
+- ci: move eval verification and Python publishing to owned runners
+- chore(deps): bump opentelemetry-api in /clients/python (#1068)
+- fix(adopt): quote optional workflow flags
+- fix(adopt): preserve shallow history and caller-owned dependency references
+- fix(adopt): reconcile effective workspace pins and preserve proposal history
+
 ## [0.221.3] — 2026-10-24
 
 - fix(analyst): repin the benchmark implementation digest after #1097 (#1098)

package/README.md

@@ -4,7 +4,6 @@ Run agent evaluations, compare changes on the same cases, and decide whether a c
 
 [![npm](https://img.shields.io/npm/v/@tangle-network/agent-eval.svg)](https://www.npmjs.com/package/@tangle-network/agent-eval)
 [![pypi](https://img.shields.io/pypi/v/agent-eval-rpc.svg)](https://pypi.org/project/agent-eval-rpc/)
-[![tests](https://github.com/tangle-network/agent-eval/actions/workflows/ci.yml/badge.svg)](https://github.com/tangle-network/agent-eval/actions/workflows/ci.yml)
 [![license: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](./LICENSE)
 
 Eval runs in your TypeScript process.

package/dist/adapters/http.d.ts

@@ -1,6 +1,6 @@
 import { d as CostReceipt } from "../cost-ledger-C1W3ICmP.js";
-import { R as Scenario, d as DispatchContext, f as DispatchFn } from "../types-DDKfH-N6.js";
-import "../index-y7i8l6Y_.js";
+import { R as Scenario, d as DispatchContext, f as DispatchFn } from "../types-M6eF9m-r.js";
+import "../index-jnuFLxOs.js";
 //#region src/adapters/http.d.ts
 export interface HttpDispatchOptions<TScenario extends Scenario, _TArtifact> {
   /** Static endpoint URL. Mutually exclusive with `resolveUrl`. */

package/dist/{agent-profile-Bc7ADJY4.d.ts → agent-profile-CNmwfaqq.d.ts}

@@ -1,6 +1,6 @@
 import { f as CostReceiptInput, m as CustomTokenPricing } from "./cost-ledger-C1W3ICmP.js";
-import { J as TraceAnalysisStore, i as AnalystFinding, m as EvidenceRef, p as AnalystUsageReceipt } from "./types-DGImvHIr.js";
-import { C as ChatResponse, S as ChatRequest } from "./types-eMhdacVp.js";
+import { J as TraceAnalysisStore, i as AnalystFinding, m as EvidenceRef, p as AnalystUsageReceipt } from "./types-6Lfdlshp.js";
+import { C as ChatResponse, S as ChatRequest } from "./types-B_IFcKp5.js";
 import { AgentProfile, AgentProfile as AgentProfile$1, HarnessType, HarnessType as HarnessType$1 } from "@tangle-network/agent-interface";
 //#region src/campaign/external-optimizer-contracts.d.ts
 interface ExternalOptimizerProcessLimits {
@@ -472,4 +472,4 @@ declare function agentProfileId(profile: AgentProfile): string;
 declare function agentProfileHash(profile: AgentProfile): string;
 //#endregion
 export { traceStoreEvidenceResolver as A, ExternalOptimizerModelCallRequest as B, AnalystEvidenceResolutionError as C, AnalystLatencyDistribution as D, AnalystIssueExpectation as E, ExternalOptimizerEndpointFormat as F, ExternalOptimizerRunnerCommand as G, ExternalOptimizerModelExecutionObservation as H, ExternalOptimizerEvaluationObservation as I, ExternalTextEvaluationRequest as J, ExternalOptimizerWireCounts as K, ExternalOptimizerEvaluationRefusalReason as L, DEFAULT_EXTERNAL_OPTIMIZER_PROCESS_LIMITS as M, ExternalOptimizerCallbackLimits as N, RunAnalystBenchmarkOptions as O, ExternalOptimizerChatRequest as P, ExternalOptimizerModelBudget as R, AnalystEvidenceResolution as S, AnalystFindingScore as T, ExternalOptimizerProcessLimits as U, ExternalOptimizerModelCallResult as V, ExternalOptimizerResumeMode as W, AnalystBenchmarkProvenance as _, ProfileAxisSpec as a, AnalystBenchmarkSummary as b, expandProfileAxes as c, AnalystBenchmarkDatasetRef as d, AnalystBenchmarkDescriptor as f, AnalystBenchmarkOutput as g, AnalystBenchmarkObservation as h, HarnessType$1 as i, DEFAULT_EXTERNAL_OPTIMIZER_CALLBACK_LIMITS as j, runAnalystBenchmark as k, harnessAxisOf as l, AnalystBenchmarkLabelState as m, CODING_HARNESSES as n, agentProfileHash as o, AnalystBenchmarkError as p, ExternalTextCandidate as q, HARNESS_NATIVE_MODEL as r, agentProfileId as s, AgentProfile$1 as t, AnalystBenchmarkCase as u, AnalystBenchmarkResult as v, AnalystEvidenceResolver as w, AnalystEvidenceExpectation as x, AnalystBenchmarkRunner as y, ExternalOptimizerModelCall as z };
-//# sourceMappingURL=agent-profile-Bc7ADJY4.d.ts.map
+//# sourceMappingURL=agent-profile-CNmwfaqq.d.ts.map

package/dist/{agent-profile-Bc7ADJY4.d.ts.map → agent-profile-CNmwfaqq.d.ts.map}

@@ -1 +1 @@
-{"version":3,"file":"agent-profile-Bc7ADJY4.d.ts","names":["_$___cost_ledger0.CostReceiptInput"],"sources":["../src/campaign/external-optimizer-contracts.ts","../src/analyst/benchmark.ts","../src/agent-profile.ts"],"mappings":";;;;;UAciB;;EAEf;;EAEA;;EAEA;;cAGW,2CAA2C,SAAS;UAOhD;;EAEf;;EAEA;;cAGW,4CAA4C,SAAS;UAMjD;EACf;EACA;EACA,MAAM,OAAO;;EAEb,SAAS,QAAQ;;KAGP;KAEA,iCAAiC;UAE5B;EACf,WAAW;EACX;;KAqDG,aAAa,KAAK,eAAc,6BACjC,IACA,0BAA0B,gBACf,aAAa,OACtB,+BACc,WAAW,IAAI,aAAa,EAAE,SAC1C;;KAGI,+BAA+B,aACzC,KAAK;EAA0B;;KAGrB;;UAMK;;WAEN;;WAEA,SAAS;;WAET,iBAAiB;WACjB,QAAQ;;;KAIP;WAEG;;WAEA,UAAU;;;;;;;WAOV,SAPUA;;WASV;;WAGA;;WAEA;;WAEA,SATkCA;;WAWlC;;;;;;;;;;;;KAaH,8BACV,SAAS,sCACN,QAAQ;;KAGD;WAEG;WACA;WACA;WACA;WACA;WACA;WACA;WACA;;WAGA;WACA;WACA;WACA;WACA;WACA;WACA;WACA;;KAGH;;;;;;KAUA;WAEG;WACA;WACA,WAAW;WACX;;WAGA;WACA;WACA,WAAW;WACX;WACA;;WAEA;WACA;;WAGA;WACA;WACA,QAAQ;WACR,YAAY;WACZ;WACA;;UAGE;;EAEf;;EAEA;;EAEA;;EAEA;;EAEA;;;;;;;;;;EAUA;;;;;;EAMA,UAAU;;EAEV;;;UAIe;;EAEf;;EAEA;;;;UCpQe;EACf;EACA,OAAO;;UAGQ;EACf;EACA;EACA;EACA;EACA,oBAAoB;EACpB;;EAEA,4BAA4B;;KAGlB;UAEK,qBAAqB;EACpC;;EAEA;;EAEA,YAAY;EACZ,OAAO;EACP,yBAAyB;;EAEzB,2BAA2B;EAC3B;EACA,WAAW;;UAGI;EACf;EACA;EACA;EACA;EACA;EACA,mBAAmB;EACnB;EACA;EACA;EACA;;EAEA;;EAEA;;EAEA;EACA;;UAGe;EACf,UAAU;EACV;EACA;;UAGe;EACf;EACA;EACA,oBAAoB;EACpB,QAAQ;;EAER;;KAGU,wBAAwB,qBAAqB;EACvD;EACA,WAAW;EACX,UAAU;EACV,SAAS;gBACK;;;;;iBAMA,2BAA2B,QACzC,WAAW,OAAO,WAAW,qBAC5B,wBAAwB;UAoBV;EACf,mBAAmB;EACnB,QAAQ;EACR,WAAW;;;;;EAKX;;EAEA,QAAQ;;UAGO;EACf;EACA;EACA;EACA;;UAGe,uBAAuB;EACtC;EACA,QACE,OAAO,QACP;IAAW;IAAgB;IAAoB,SAAS;MACvD,yBAAyB,QAAQ;;UAGrB;EACf;EACA;EACA;EACA,YAAY;EACZ;EACA;EACA;EACA;EACA,mBAAmB;EACnB,OAAO;EACP,qBAAqB;EACrB;EACA,eAAe;EACf,QAAQ;EACR,iBAAiB;EACjB,QAAQ;;UAGO;EACf;EACA;EACA;EACA;EACA;;UAGe;EACf;EACA;EACA;EACA;EACA;EACA;EACA;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;;EAEA;;EAEA;;EAEA;;EAEA;EACA,WAAW;EACX;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;;UAGe;EACf;EACA;EACA;;UAGe;EACf;EACA,UAAU;EACV;EACA,cAAc;EACd,WAAW;;UAGI,mCAAmC;EAClD;EACA;EACA;EACA;EACA;EACA;EACA;;UAGe;EACf,YAAY;EACZ,cAAc;EACd,WAAW;;UAGI,2BAA2B;EAC1C,gBAAgB,qBAAqB;EACrC,kBAAkB,uBAAuB;EACzC;EACA;EACA;EACA,kBAAkB,wBAAwB;EAC1C,YAAY;;EAEZ,+BAA+B;EAC/B,iBAAiB,aAAa,uCAAuC;EACrE,SAAS;;iBAGW,oBAAoB,QACxC,SAAS,2BAA2B,UACnC,QAAQ;;;;;;;;;;cCnQE,2BAA2B;UAOvB;;;EAGf,MAAM;;EAEN,qBAAqB;;;EAGrB;;;;;;EAMA;;;;;;;cAQW;;;;;;;;;;;;;;;;;;;iBAoBG,kBAAkB,MAAM,kBAAkB;;;;;;;;iBAwD1C,cACd,SAAS,KAAK;EACX,SAAS;EAAa;;;;;;;;;iBAiBX,eAAe,SAAS;;;;;;;;;;;iBAmDxB,iBAAiB,SAAS"}
+{"version":3,"file":"agent-profile-CNmwfaqq.d.ts","names":["_$___cost_ledger0.CostReceiptInput"],"sources":["../src/campaign/external-optimizer-contracts.ts","../src/analyst/benchmark.ts","../src/agent-profile.ts"],"mappings":";;;;;UAciB;;EAEf;;EAEA;;EAEA;;cAGW,2CAA2C,SAAS;UAOhD;;EAEf;;EAEA;;cAGW,4CAA4C,SAAS;UAMjD;EACf;EACA;EACA,MAAM,OAAO;;EAEb,SAAS,QAAQ;;KAGP;KAEA,iCAAiC;UAE5B;EACf,WAAW;EACX;;KAqDG,aAAa,KAAK,eAAc,6BACjC,IACA,0BAA0B,gBACf,aAAa,OACtB,+BACc,WAAW,IAAI,aAAa,EAAE,SAC1C;;KAGI,+BAA+B,aACzC,KAAK;EAA0B;;KAGrB;;UAMK;;WAEN;;WAEA,SAAS;;WAET,iBAAiB;WACjB,QAAQ;;;KAIP;WAEG;;WAEA,UAAU;;;;;;;WAOV,SAPUA;;WASV;;WAGA;;WAEA;;WAEA,SATkCA;;WAWlC;;;;;;;;;;;;KAaH,8BACV,SAAS,sCACN,QAAQ;;KAGD;WAEG;WACA;WACA;WACA;WACA;WACA;WACA;WACA;;WAGA;WACA;WACA;WACA;WACA;WACA;WACA;WACA;;KAGH;;;;;;KAUA;WAEG;WACA;WACA,WAAW;WACX;;WAGA;WACA;WACA,WAAW;WACX;WACA;;WAEA;WACA;;WAGA;WACA;WACA,QAAQ;WACR,YAAY;WACZ;WACA;;UAGE;;EAEf;;EAEA;;EAEA;;EAEA;;EAEA;;;;;;;;;;EAUA;;;;;;EAMA,UAAU;;EAEV;;;UAIe;;EAEf;;EAEA;;;;UCpQe;EACf;EACA,OAAO;;UAGQ;EACf;EACA;EACA;EACA;EACA,oBAAoB;EACpB;;EAEA,4BAA4B;;KAGlB;UAEK,qBAAqB;EACpC;;EAEA;;EAEA,YAAY;EACZ,OAAO;EACP,yBAAyB;;EAEzB,2BAA2B;EAC3B;EACA,WAAW;;UAGI;EACf;EACA;EACA;EACA;EACA;EACA,mBAAmB;EACnB;EACA;EACA;EACA;;EAEA;;EAEA;;EAEA;EACA;;UAGe;EACf,UAAU;EACV;EACA;;UAGe;EACf;EACA;EACA,oBAAoB;EACpB,QAAQ;;EAER;;KAGU,wBAAwB,qBAAqB;EACvD;EACA,WAAW;EACX,UAAU;EACV,SAAS;gBACK;;;;;iBAMA,2BAA2B,QACzC,WAAW,OAAO,WAAW,qBAC5B,wBAAwB;UAoBV;EACf,mBAAmB;EACnB,QAAQ;EACR,WAAW;;;;;EAKX;;EAEA,QAAQ;;UAGO;EACf;EACA;EACA;EACA;;UAGe,uBAAuB;EACtC;EACA,QACE,OAAO,QACP;IAAW;IAAgB;IAAoB,SAAS;MACvD,yBAAyB,QAAQ;;UAGrB;EACf;EACA;EACA;EACA,YAAY;EACZ;EACA;EACA;EACA;EACA,mBAAmB;EACnB,OAAO;EACP,qBAAqB;EACrB;EACA,eAAe;EACf,QAAQ;EACR,iBAAiB;EACjB,QAAQ;;UAGO;EACf;EACA;EACA;EACA;EACA;;UAGe;EACf;EACA;EACA;EACA;EACA;EACA;EACA;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;;EAEA;;EAEA;;EAEA;;EAEA;EACA,WAAW;EACX;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;EACA;;UAGe;EACf;EACA;EACA;;UAGe;EACf;EACA,UAAU;EACV;EACA,cAAc;EACd,WAAW;;UAGI,mCAAmC;EAClD;EACA;EACA;EACA;EACA;EACA;EACA;;UAGe;EACf,YAAY;EACZ,cAAc;EACd,WAAW;;UAGI,2BAA2B;EAC1C,gBAAgB,qBAAqB;EACrC,kBAAkB,uBAAuB;EACzC;EACA;EACA;EACA,kBAAkB,wBAAwB;EAC1C,YAAY;;EAEZ,+BAA+B;EAC/B,iBAAiB,aAAa,uCAAuC;EACrE,SAAS;;iBAGW,oBAAoB,QACxC,SAAS,2BAA2B,UACnC,QAAQ;;;;;;;;;;cCnQE,2BAA2B;UAOvB;;;EAGf,MAAM;;EAEN,qBAAqB;;;EAGrB;;;;;;EAMA;;;;;;;cAQW;;;;;;;;;;;;;;;;;;;iBAoBG,kBAAkB,MAAM,kBAAkB;;;;;;;;iBAwD1C,cACd,SAAS,KAAK;EACX,SAAS;EAAa;;;;;;;;;iBAiBX,eAAe,SAAS;;;;;;;;;;;iBAmDxB,iBAAiB,SAAS"}

package/dist/analyst/index.d.ts

@@ -1,19 +1,19 @@
 import { m as CustomTokenPricing, o as CostLedgerHandle } from "../cost-ledger-C1W3ICmP.js";
-import { A as FindingReaderLabel, B as TREATMENT_LAYERS, C as FINDING_LAYERS, D as FindingMeasure, E as FindingLayer, F as lineageRoots, I as parseAnalystFinding, J as TraceAnalysisStore, L as readFindingScope, M as FindingScopeReading, N as ProposalFindingSchema, O as FindingMethod, P as independentUnitCount, R as FindingConditions, S as makeProposalFinding, T as FindingEvidenceUnit, V as TreatmentLayer, _ as ExecutionProbeRequest, a as AnalystFindingSchemaVersion, at as TraceAnalystSpan, b as computeFindingId, c as AnalystRunEvent, d as AnalystRunSummary, f as AnalystSeverity, g as ExecutionProbeOutcome, h as ExecutionProbe, i as AnalystFinding, j as FindingScope, k as FindingParseOutcome, l as AnalystRunInputs, m as EvidenceRef, n as AnalystContext, o as AnalystInputKind, p as AnalystUsageReceipt, r as AnalystCost, s as AnalystRequirements, t as Analyst, u as AnalystRunResult, v as ProposalFinding, w as FindingCalibration, x as makeFinding, y as ProposalFindingOrigin, z as FindingTreatment } from "../types-DGImvHIr.js";
-import { A as createChatClient, C as ChatResponse, D as MockTransportOpts, E as CustomTransportOpts, O as OpenAiCompatibleTransportOpts, S as ChatRequest, T as CreateChatClientOpts, b as ChatCallOpts, k as SandboxSdkTransportOpts, w as ChatTransport, x as ChatClient } from "../types-eMhdacVp.js";
-import { A as traceStoreEvidenceResolver, C as AnalystEvidenceResolutionError, D as AnalystLatencyDistribution, E as AnalystIssueExpectation, G as ExternalOptimizerRunnerCommand, H as ExternalOptimizerModelExecutionObservation, O as RunAnalystBenchmarkOptions, S as AnalystEvidenceResolution, T as AnalystFindingScore, _ as AnalystBenchmarkProvenance, b as AnalystBenchmarkSummary, d as AnalystBenchmarkDatasetRef, f as AnalystBenchmarkDescriptor, g as AnalystBenchmarkOutput, h as AnalystBenchmarkObservation, k as runAnalystBenchmark, m as AnalystBenchmarkLabelState, p as AnalystBenchmarkError, t as AgentProfile, u as AnalystBenchmarkCase, v as AnalystBenchmarkResult, w as AnalystEvidenceResolver, x as AnalystEvidenceExpectation, y as AnalystBenchmarkRunner, z as ExternalOptimizerModelCall } from "../agent-profile-Bc7ADJY4.js";
+import { A as FindingReaderLabel, B as TREATMENT_LAYERS, C as FINDING_LAYERS, D as FindingMeasure, E as FindingLayer, F as lineageRoots, I as parseAnalystFinding, J as TraceAnalysisStore, L as readFindingScope, M as FindingScopeReading, N as ProposalFindingSchema, O as FindingMethod, P as independentUnitCount, R as FindingConditions, S as makeProposalFinding, T as FindingEvidenceUnit, V as TreatmentLayer, _ as ExecutionProbeRequest, a as AnalystFindingSchemaVersion, at as TraceAnalystSpan, b as computeFindingId, c as AnalystRunEvent, d as AnalystRunSummary, f as AnalystSeverity, g as ExecutionProbeOutcome, h as ExecutionProbe, i as AnalystFinding, j as FindingScope, k as FindingParseOutcome, l as AnalystRunInputs, m as EvidenceRef, n as AnalystContext, o as AnalystInputKind, p as AnalystUsageReceipt, r as AnalystCost, s as AnalystRequirements, t as Analyst, u as AnalystRunResult, v as ProposalFinding, w as FindingCalibration, x as makeFinding, y as ProposalFindingOrigin, z as FindingTreatment } from "../types-6Lfdlshp.js";
+import { A as createChatClient, C as ChatResponse, D as MockTransportOpts, E as CustomTransportOpts, O as OpenAiCompatibleTransportOpts, S as ChatRequest, T as CreateChatClientOpts, b as ChatCallOpts, k as SandboxSdkTransportOpts, w as ChatTransport, x as ChatClient } from "../types-B_IFcKp5.js";
+import { A as traceStoreEvidenceResolver, C as AnalystEvidenceResolutionError, D as AnalystLatencyDistribution, E as AnalystIssueExpectation, G as ExternalOptimizerRunnerCommand, H as ExternalOptimizerModelExecutionObservation, O as RunAnalystBenchmarkOptions, S as AnalystEvidenceResolution, T as AnalystFindingScore, _ as AnalystBenchmarkProvenance, b as AnalystBenchmarkSummary, d as AnalystBenchmarkDatasetRef, f as AnalystBenchmarkDescriptor, g as AnalystBenchmarkOutput, h as AnalystBenchmarkObservation, k as runAnalystBenchmark, m as AnalystBenchmarkLabelState, p as AnalystBenchmarkError, t as AgentProfile, u as AnalystBenchmarkCase, v as AnalystBenchmarkResult, w as AnalystEvidenceResolver, x as AnalystEvidenceExpectation, y as AnalystBenchmarkRunner, z as ExternalOptimizerModelCall } from "../agent-profile-CNmwfaqq.js";
 import { g as SystemOneState } from "../systemone-protocol-Bl2EI2KQ.js";
-import { A as FindingsLedger, B as FindingsDiff, F as LedgerFinding, H as PersistedFinding, I as diffFindingsOverTime, L as openFindingsLedger, M as FindingsLedgerRead, N as FindingsQuery, P as FindingsTimeStep, R as queryFindings, U as diffFindings, V as FindingsStore, j as FindingsLedgerOutcome, r as SystemOneEvaluatorOptions, z as DiffPolicy } from "../systemone-D3MrW63C.js";
-import { a as ExactAnalystRunPolicySnapshot, c as ExactAnalystSnapshot, d as ExactExecutionComponentSnapshot, i as ExactAnalystRunEvent, l as ExactCapableAnalyst, n as ExactAnalystExecutionPlanSnapshot, o as ExactAnalystRunResult, r as ExactAnalystRunCompletion, s as ExactAnalystRunSummary, t as ExactAnalystBudgetSnapshot, u as ExactExecutionComponentIdentity } from "../exact-types-CVn1QxFO.js";
-import { _ as BehavioralTraceQuestionsOptions, a as createTraceAnalyst, b as BehavioralWindowPolicy, c as behavioralAnalyst, d as BehavioralQuestionCalibration, f as BehavioralTraceAnswer, g as BehavioralTraceQuestionId, h as BehavioralTraceLayer, i as TraceAnalystDefinition, l as BEHAVIORAL_TRACE_QUESTIONS, m as BehavioralTraceEventMap, n as buildDefaultAnalystRegistry, o as runTraceAnalyst, p as BehavioralTraceEvent, r as CreateTraceAnalystOptions, s as BehavioralAnalystOptions, t as DefaultAnalystRegistryOptions, u as BEHAVIORAL_TRACE_QUESTIONS_VERSION, v as BehavioralTraceSupport, x as behavioralTraceQuestionDigest, y as BehavioralTraceWindow } from "../default-registry-BMHqNR6_.js";
-import { a as RAW_FINDING_SCHEMA_PROMPT, c as RawAnalystFindingSchema, i as TraceAnalystLimits, l as evidenceRefsFromRawFinding, n as TraceAnalysisEngineRequest, o as RawAnalystEvidence, r as TraceAnalysisEngineResult, s as RawAnalystFinding, t as TraceAnalysisEngine, u as parseRawFinding } from "../engine-BF95CWzH.js";
-import { n as buildTraceToolsForGroup, t as TraceToolGroupName } from "../tool-groups-DK1bu5TD.js";
-import { a as BudgetPolicy, c as RegistryRunOpts, i as AnalystRegistryOptions, n as ExactAnalystRunExecutionError, o as ExactAnalystBudgetPolicy, r as AnalystHooks, s as ExactRegistryRunOpts, t as AnalystRegistry } from "../registry-DDKaUrmM.js";
-import { T as SemanticConceptJudgeOptions, _ as DspyRlmTraceEngineOptions, a as FAILURE_MODE_KIND_SPEC, b as createChatTraceEngine, c as emitControlIntegrityFindings, d as FindingSubject, f as FindingSubjectKind, g as renderFindingSubject, h as parseFindingSubject, i as IMPROVEMENT_KIND_SPEC, l as FINDING_SUBJECT_KINDS, m as findingSubjectGrammarPromptFor, n as KNOWLEDGE_POISONING_KIND_SPEC, o as CONTROL_INTEGRITY_ANALYST, p as KIND_EXPECTED_SUBJECTS, r as KNOWLEDGE_GAP_KIND_SPEC, s as ControlIntegrityAnalyst, t as DEFAULT_TRACE_ANALYST_KINDS, u as FINDING_SUBJECT_SYNTAX, v as createDspyRlmTraceEngine, y as ChatTraceEngineOptions } from "../index-BJZ7s-sv.js";
-import { C as runPrimeExchange, D as ReplyRowDecoded, E as ReplyEnvelope, O as ReplyRowRejection, S as projectPrimeTrajectory, T as ReplyContract, _ as buildPrimePrompt, a as PrimeProjectionOutcome, b as normalizePrimeUsage, c as PrimeProtocolIdentity, d as PrimeRepairPromptSpec, f as PrimeRepairState, g as analystUsageReceiptFromPrimeUsage, h as PrimeTurnRecord, i as PrimeProjectionDelivery, l as PrimeRawUsage, m as PrimeRowDecoded, n as PrimeExchangeOutcome, o as PrimeProjectionSource, p as PrimeReplyContract, r as PrimeFailure, s as PrimePromptSpec, t as PrimeExchangeOptions, u as PrimeRejectedRow, v as buildPrimeRepairPrompt, w as DecodedReply, x as primeProtocolSha256, y as mergePrimeRawUsage } from "../prime-protocol-BEaqBoQE.js";
+import { A as FindingsLedger, B as FindingsDiff, F as LedgerFinding, H as PersistedFinding, I as diffFindingsOverTime, L as openFindingsLedger, M as FindingsLedgerRead, N as FindingsQuery, P as FindingsTimeStep, R as queryFindings, U as diffFindings, V as FindingsStore, j as FindingsLedgerOutcome, r as SystemOneEvaluatorOptions, z as DiffPolicy } from "../systemone-CHnqwmYg.js";
+import { a as ExactAnalystRunPolicySnapshot, c as ExactAnalystSnapshot, d as ExactExecutionComponentSnapshot, i as ExactAnalystRunEvent, l as ExactCapableAnalyst, n as ExactAnalystExecutionPlanSnapshot, o as ExactAnalystRunResult, r as ExactAnalystRunCompletion, s as ExactAnalystRunSummary, t as ExactAnalystBudgetSnapshot, u as ExactExecutionComponentIdentity } from "../exact-types-Do63KoKn.js";
+import { _ as BehavioralTraceQuestionsOptions, a as createTraceAnalyst, b as BehavioralWindowPolicy, c as behavioralAnalyst, d as BehavioralQuestionCalibration, f as BehavioralTraceAnswer, g as BehavioralTraceQuestionId, h as BehavioralTraceLayer, i as TraceAnalystDefinition, l as BEHAVIORAL_TRACE_QUESTIONS, m as BehavioralTraceEventMap, n as buildDefaultAnalystRegistry, o as runTraceAnalyst, p as BehavioralTraceEvent, r as CreateTraceAnalystOptions, s as BehavioralAnalystOptions, t as DefaultAnalystRegistryOptions, u as BEHAVIORAL_TRACE_QUESTIONS_VERSION, v as BehavioralTraceSupport, x as behavioralTraceQuestionDigest, y as BehavioralTraceWindow } from "../default-registry-CLG5YqGS.js";
+import { a as RAW_FINDING_SCHEMA_PROMPT, c as RawAnalystFindingSchema, i as TraceAnalystLimits, l as evidenceRefsFromRawFinding, n as TraceAnalysisEngineRequest, o as RawAnalystEvidence, r as TraceAnalysisEngineResult, s as RawAnalystFinding, t as TraceAnalysisEngine, u as parseRawFinding } from "../engine-dtRVeuQs.js";
+import { n as buildTraceToolsForGroup, t as TraceToolGroupName } from "../tool-groups-b3oU_D88.js";
+import { a as BudgetPolicy, c as RegistryRunOpts, i as AnalystRegistryOptions, n as ExactAnalystRunExecutionError, o as ExactAnalystBudgetPolicy, r as AnalystHooks, s as ExactRegistryRunOpts, t as AnalystRegistry } from "../registry-BcdJmyoU.js";
+import { T as SemanticConceptJudgeOptions, _ as DspyRlmTraceEngineOptions, a as FAILURE_MODE_KIND_SPEC, b as createChatTraceEngine, c as emitControlIntegrityFindings, d as FindingSubject, f as FindingSubjectKind, g as renderFindingSubject, h as parseFindingSubject, i as IMPROVEMENT_KIND_SPEC, l as FINDING_SUBJECT_KINDS, m as findingSubjectGrammarPromptFor, n as KNOWLEDGE_POISONING_KIND_SPEC, o as CONTROL_INTEGRITY_ANALYST, p as KIND_EXPECTED_SUBJECTS, r as KNOWLEDGE_GAP_KIND_SPEC, s as ControlIntegrityAnalyst, t as DEFAULT_TRACE_ANALYST_KINDS, u as FINDING_SUBJECT_SYNTAX, v as createDspyRlmTraceEngine, y as ChatTraceEngineOptions } from "../index-CTIRz-AB.js";
+import { C as runPrimeExchange, D as ReplyRowDecoded, E as ReplyEnvelope, O as ReplyRowRejection, S as projectPrimeTrajectory, T as ReplyContract, _ as buildPrimePrompt, a as PrimeProjectionOutcome, b as normalizePrimeUsage, c as PrimeProtocolIdentity, d as PrimeRepairPromptSpec, f as PrimeRepairState, g as analystUsageReceiptFromPrimeUsage, h as PrimeTurnRecord, i as PrimeProjectionDelivery, l as PrimeRawUsage, m as PrimeRowDecoded, n as PrimeExchangeOutcome, o as PrimeProjectionSource, p as PrimeReplyContract, r as PrimeFailure, s as PrimePromptSpec, t as PrimeExchangeOptions, u as PrimeRejectedRow, v as buildPrimeRepairPrompt, w as DecodedReply, x as primeProtocolSha256, y as mergePrimeRawUsage } from "../prime-protocol-BGxjfhM4.js";
 import { i as nodeHttpPrimeBridgeTransport, n as PrimeBridgeTransportRequest, r as PrimeBridgeTransportResult, t as PrimeBridgeTransport } from "../prime-bridge-transport-6feEglLf.js";
 import { b as SupervisorRunSources } from "../types-CmXY0PQf.js";
-import { a as askTraceQuestions, i as VerifiedTraceFinding, n as TraceQuestionOutcome, r as TraceQuestionSpec, t as AskTraceQuestionsOptions } from "../trace-questions-D_OKYZ6u.js";
+import { a as askTraceQuestions, i as VerifiedTraceFinding, n as TraceQuestionOutcome, r as TraceQuestionSpec, t as AskTraceQuestionsOptions } from "../trace-questions-BmT8M6UQ.js";
 import { z } from "zod";
 //#region src/analyst/adapters.d.ts
 interface SemanticConceptJudgeAdapterOpts {
@@ -1292,7 +1292,7 @@ export declare function lowestCommonAncestor(taxonomy: DomainTaxonomy, ids: read
 //#endregion
 //#region src/analyst/failure-modes.d.ts
 export declare const FAILURE_MODES_ANALYST_ID = "failure-modes";
-export declare const FAILURE_MODES_VERSION = "1";
+export declare const FAILURE_MODES_VERSION = "2";
 /** κ below this refuses the rubric. */
 export declare const FAILURE_MODES_MIN_KAPPA = 0.6;
 export declare const FAILURE_MODE_LABELS: {
@@ -1320,7 +1320,7 @@ type FailureModeLabel = keyof typeof FAILURE_MODE_LABELS;
 export declare const FAILURE_MODES_QUESTIONS: {
   readonly failure_mode: {
     readonly type: 'choice';
-    readonly instructions: 'Which single failure mode best explains why this agent or team fell short? Choose none when it did not fail.';
+    readonly instructions: 'Which agent or team root-cause failure best explains the observed shortfall? Identify the causal error rather than a downstream symptom; a corrected earlier mistake need not be the root cause. Use only the provided task, policy, actions, tool results and outcome evidence. Do not invent missing outcomes or infrastructure errors. Choose none when the evidence establishes that no failure occurred.';
     readonly criteria: {
       readonly 'disobey-task-specification': 'Ignored or violated the stated task requirements or constraints.';
       readonly 'disobey-role-specification': 'Acted outside its assigned role or took over another role.';

package/dist/analyst/index.js

@@ -2,16 +2,16 @@ import { n as assertConsistentFindingConditions, o as mergeFindingComposition, r
 import { i as nominalAgreement } from "../agreement-irr-BhAqjluA.js";
 import { t as FileLedgerJournal } from "../journal-C7om8Gum.js";
 import { r as merkleRoot } from "../ledger-core-CAxUdDk0.js";
-import { A as spanningWindow, C as UNKNOWN_LINEAGE, D as mostSevere, E as mergeEvidenceUnits, O as parseAnalystFinding, S as ProposalFindingSchema, T as lineageRoots, b as AnalystFindingSchema, k as readFindingScope, w as independentUnitCount, x as FINDING_LAYERS } from "../external-optimizer-subprocess-DA02c_L6.js";
+import { A as spanningWindow, C as UNKNOWN_LINEAGE, D as mostSevere, E as mergeEvidenceUnits, O as parseAnalystFinding, S as ProposalFindingSchema, T as lineageRoots, b as AnalystFindingSchema, k as readFindingScope, w as independentUnitCount, x as FINDING_LAYERS } from "../external-optimizer-subprocess-B_ukPtRD.js";
 import { a as makeProposalFinding, i as makeFinding, r as computeFindingId } from "../types-CnWmj8NW.js";
-import { t as assertProposalFindings } from "../proposal-findings-D0_zkQ2l.js";
+import { t as assertProposalFindings } from "../proposal-findings-DqLUWsAG.js";
 import { A as RawAnalystFindingSchema, B as parseFindingSubject, F as stripCodeFences, I as FINDING_SUBJECT_KINDS, L as FINDING_SUBJECT_SYNTAX, M as parseRawFinding, N as parseTraceSpanEvidenceUri, R as KIND_EXPECTED_SUBJECTS, V as renderFindingSubject, j as evidenceRefsFromRawFinding, k as RAW_FINDING_SCHEMA_PROMPT, n as runTraceAnalyst, r as buildTraceToolsForGroup, t as createTraceAnalyst, z as findingSubjectGrammarPromptFor } from "../kind-factory-Dl0PmNQL.js";
-import { _ as behavioralTraceQuestionDigest, a as KNOWLEDGE_POISONING_KIND_SPEC, c as FAILURE_MODE_KIND_SPEC, d as emitControlIntegrityFindings, f as behavioralAnalyst, g as BEHAVIORAL_TRACE_QUESTIONS_VERSION, h as BEHAVIORAL_TRACE_QUESTIONS, i as DEFAULT_TRACE_ANALYST_KINDS, l as CONTROL_INTEGRITY_ANALYST, m as behavioralTraceSupportSchema, n as AnalystRegistry, o as KNOWLEDGE_GAP_KIND_SPEC, p as validBehavioralCalibration, r as ExactAnalystRunExecutionError, s as IMPROVEMENT_KIND_SPEC, t as buildDefaultAnalystRegistry, u as ControlIntegrityAnalyst } from "../default-registry-BZ5FyYl9.js";
-import { t as createDspyRlmTraceEngine } from "../dspy-rlm-engine-BTy7u0jQ.js";
-import { n as diffFindings, r as createChatTraceEngine, t as FindingsStore } from "../findings-store-DRiFdoid.js";
+import { _ as behavioralTraceQuestionDigest, a as KNOWLEDGE_POISONING_KIND_SPEC, c as FAILURE_MODE_KIND_SPEC, d as emitControlIntegrityFindings, f as behavioralAnalyst, g as BEHAVIORAL_TRACE_QUESTIONS_VERSION, h as BEHAVIORAL_TRACE_QUESTIONS, i as DEFAULT_TRACE_ANALYST_KINDS, l as CONTROL_INTEGRITY_ANALYST, m as behavioralTraceSupportSchema, n as AnalystRegistry, o as KNOWLEDGE_GAP_KIND_SPEC, p as validBehavioralCalibration, r as ExactAnalystRunExecutionError, s as IMPROVEMENT_KIND_SPEC, t as buildDefaultAnalystRegistry, u as ControlIntegrityAnalyst } from "../default-registry-BBTDgkDM.js";
+import { t as createDspyRlmTraceEngine } from "../dspy-rlm-engine-DN1p_5jx.js";
+import { n as diffFindings, r as createChatTraceEngine, t as FindingsStore } from "../findings-store-Bg4sdS-e.js";
 import { a as systemOneAnalyst, p as validDecisionIdentity } from "../decision-point-0nJubuoM.js";
-import { t as createChatClient } from "../chat-client-CMQ8_h4J.js";
-import { A as agentRxBenchmarkCase, C as ANALYST_BENCHMARK_COST_LEDGER_FILE, D as AGENT_RX_UPSTREAM_REVISION, E as ANALYST_BENCHMARK_OBSERVATIONS_FILE, F as runAnalystBenchmark, I as traceStoreEvidenceResolver, M as normalizeAgentRxCategory, N as externalId, O as codeTraceBenchCase, P as positiveStep, S as ANALYST_BENCHMARK_IMPLEMENTATION_FILES, T as ANALYST_BENCHMARK_MANIFEST_FILE, _ as readAnalystBenchmarkArtifact, a as runInlineAnalystDefinition, b as ANALYST_BENCHMARK_EVIDENCE_DEPENDENCY_LOCK_SHA256, c as publicRlmAnalystDefinition, d as publicDirectAnalystDefinition, f as runChunkedAnalystDefinition, g as expandCodeTraceFailureBlocks, h as analystDefinitionProtocolSha256, i as primeCodeTraceAnalystDefinition, j as agentRxPredictionsToFindings, k as codeTracerPredictionsToFindings, l as runReplVariableAnalystDefinition, m as analystDefinitionAsymmetries, n as createPrimeBenchmarkRunner, o as loadPublicBenchmarkRows, p as AnalystExpressivenessError, r as primeAnalystProtocolSha256, s as createPublicBenchmarkRlmRunner, t as runAnalystBenchmarkCommand, u as createPublicBenchmarkDirectRunner, v as CODE_TRACE_BENCH_ANALYST_PROMPT, w as ANALYST_BENCHMARK_LOCAL_RECEIPT_FILE, x as ANALYST_BENCHMARK_EVIDENCE_IMPLEMENTATION_SHA256, y as publicBenchmarkProtocolSha256 } from "../benchmark-command-BDq0VgoP.js";
+import { t as createChatClient } from "../chat-client-BNATB3rJ.js";
+import { A as agentRxBenchmarkCase, C as ANALYST_BENCHMARK_COST_LEDGER_FILE, D as AGENT_RX_UPSTREAM_REVISION, E as ANALYST_BENCHMARK_OBSERVATIONS_FILE, F as runAnalystBenchmark, I as traceStoreEvidenceResolver, M as normalizeAgentRxCategory, N as externalId, O as codeTraceBenchCase, P as positiveStep, S as ANALYST_BENCHMARK_IMPLEMENTATION_FILES, T as ANALYST_BENCHMARK_MANIFEST_FILE, _ as readAnalystBenchmarkArtifact, a as runInlineAnalystDefinition, b as ANALYST_BENCHMARK_EVIDENCE_DEPENDENCY_LOCK_SHA256, c as publicRlmAnalystDefinition, d as publicDirectAnalystDefinition, f as runChunkedAnalystDefinition, g as expandCodeTraceFailureBlocks, h as analystDefinitionProtocolSha256, i as primeCodeTraceAnalystDefinition, j as agentRxPredictionsToFindings, k as codeTracerPredictionsToFindings, l as runReplVariableAnalystDefinition, m as analystDefinitionAsymmetries, n as createPrimeBenchmarkRunner, o as loadPublicBenchmarkRows, p as AnalystExpressivenessError, r as primeAnalystProtocolSha256, s as createPublicBenchmarkRlmRunner, t as runAnalystBenchmarkCommand, u as createPublicBenchmarkDirectRunner, v as CODE_TRACE_BENCH_ANALYST_PROMPT, w as ANALYST_BENCHMARK_LOCAL_RECEIPT_FILE, x as ANALYST_BENCHMARK_EVIDENCE_IMPLEMENTATION_SHA256, y as publicBenchmarkProtocolSha256 } from "../benchmark-command-Cnjfo0Zh.js";
 import { a as normalizePrimeUsage, c as runPrimeExchange, i as mergePrimeRawUsage, n as buildPrimePrompt, o as primeProtocolSha256, r as buildPrimeRepairPrompt, s as projectPrimeTrajectory, t as analystUsageReceiptFromPrimeUsage } from "../prime-protocol-BFgGrEMP.js";
 import { t as nodeHttpPrimeBridgeTransport } from "../prime-bridge-transport-Z3cpKGoV.js";
 import { t as clusterBootstrapMean } from "../cluster-bootstrap-DEwwhl0C.js";
@@ -400,7 +400,7 @@ function nodeIndex(taxonomy) {
 * gold set has nothing to report, so it reports nothing.
 */
 const FAILURE_MODES_ANALYST_ID = "failure-modes";
-const FAILURE_MODES_VERSION = "1";
+const FAILURE_MODES_VERSION = "2";
 /** κ below this refuses the rubric. */
 const FAILURE_MODES_MIN_KAPPA = .6;
 const FAILURE_MODE_LABELS = {
@@ -450,7 +450,7 @@ const AGENT_RX_AGENT_ACTORS = /* @__PURE__ */ new Set([
 const FAILURE_MODES_QUESTIONS = {
 	failure_mode: {
 		type: "choice",
-		instructions: "Which single failure mode best explains why this agent or team fell short? Choose none when it did not fail.",
+		instructions: "Which agent or team root-cause failure best explains the observed shortfall? Identify the causal error rather than a downstream symptom; a corrected earlier mistake need not be the root cause. Use only the provided task, policy, actions, tool results and outcome evidence. Do not invent missing outcomes or infrastructure errors. Choose none when the evidence establishes that no failure occurred.",
 		criteria: FAILURE_MODE_LABELS
 	},
 	delegation_brief: {
@@ -661,7 +661,7 @@ function createFailureModesAnalyst(options) {
 	const method = {
 		kind: "judged",
 		name: FAILURE_MODES_ANALYST_ID,
-		version: "1",
+		version: "2",
 		calibration: {
 			gold_n: calibration.gold_n,
 			agreement: calibration.agreement,
@@ -677,7 +677,7 @@ function createFailureModesAnalyst(options) {
 			id: FAILURE_MODES_ANALYST_ID,
 			description: "Labels agent, team and run failures with a calibrated fixed-choice rubric.",
 			inputKind: "custom",
-			version: "1",
+			version: "2",
 			questions: FAILURE_MODES_QUESTIONS,
 			findings: (value, input, context) => {
 				const judgedSubject = subject?.(input, context);