@tangle-network/agent-eval 0.205.2 → 0.206.0

package/docs/public-api.md

@@ -7,11 +7,11 @@ Generated by `pnpm api:census` on demand — this is a dated reading, not a gate
 | measure | count |
 | --- | --- |
 | export subpaths | 33 |
-| published value exports (subpath x symbol) | 1589 |
-| distinct symbols | 1401 |
-| production | 1048 |
-| planned | 278 |
-| none | 263 |
+| published value exports (subpath x symbol) | 1614 |
+| distinct symbols | 1422 |
+| production | 1052 |
+| planned | 296 |
+| none | 266 |
 
 Type-only exports are not listed: a type binds no runtime surface, and removing one cannot break a caller at run time.
 
@@ -84,7 +84,7 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 
 ### `.`
 
-360 value exports — 279 production, 40 planned, 41 none.
+371 value exports — 281 production, 44 planned, 46 none.
 
 | symbol | consumer | evidence |
 | --- | --- | --- |
@@ -150,6 +150,8 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `checkCanaries` | production | blueprint-agent:scripts/experiments/lib/contamination-preflight.ts |
 | `checkServedModel` | production | this package: src/integrity/preflight.ts:33 |
 | `checkTraceContracts` | production | creative-agent:eval/lib/trace-contracts.ts |
+| `chiSquared` | none | only this package's tests: src/statistics/inference.test.ts:2 |
+| `chiSquaredPValue` | none | only this package's tests: src/statistics/inference.test.ts:2 |
 | `clamp01` | production | agent-dev-container:products/intelligence/api/src/lib/consultant/executor.ts |
 | `classifyFailure` | production | phony:products/builder/api/src/eval/autoresearch/trace-analyst.ts |
 | `classifyFailureReason` | production | this package: src/diagnosis/first-failure.ts:39 |
@@ -246,6 +248,7 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `FileSystemTraceStore` | production | agent-builder:scripts/eval.ts |
 | `fileVerdictCache` | production | blueprint-agent:scripts/experiments/calibrate-instrument.ts |
 | `FindingsStore` | production | agent-builder:scripts/run-canonical-analyst-loop.ts |
+| `fisherExactTwoSided` | none | only this package's tests: src/statistics/inference.test.ts:2 |
 | `formatScorecardDiff` | production | ai-trading-blueprint:evals/src/trading/scorecard-integration.ts |
 | `gainHistogram` | production | blueprint-agent:scripts/experiments/vb-runrecord-analysis.ts |
 | `gateTreatmentApplied` | production | blueprint-agent:scripts/experiments/lib/validity-gates.ts |
@@ -286,6 +289,7 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `judgeSpans` | production | this package: src/builder-eval/three-layer-eval.ts:25 |
 | `knowledgeReadinessTracePayload` | production | agent-dev-container:products/intelligence/api/src/routes/project-engines.ts |
 | `leaderboard` | production | blueprint-agent:scripts/experiments/lib/cross-harness-leaderboard.ts |
+| `linearRegressionSlope` | none | only this package's tests: src/statistics/inference.test.ts:2 |
 | `lintTraceContractSpec` | planned | doc: docs/trace-contracts.md |
 | `LlmCallError` | production | agent-dev-container:products/intelligence/api/src/lib/router-model-owner.ts |
 | `llmJudge` | production | agent-runtime:examples/agentic-data-creation/offline-fixtures.ts |
@@ -314,6 +318,7 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `mulberry32` | production | agent-knowledge:src/memory/holdout.ts |
 | `MultiLayerVerifier` | production | agent-builder:src/lib/.server/eval/loops/forge-refinement.ts |
 | `NoopRawProviderSink` | production | agent-dev-container:products/intelligence/api/src/lib/eval-engine.ts |
+| `normalTwoSidedPValue` | none | only this package's tests: src/statistics/inference.test.ts:2 |
 | `notBlocked` | none | only this package's tests: src/oracle.test.ts:2 |
 | `NotFoundError` | production | agent-runtime:src/errors.ts |
 | `objectiveEval` | production | agent-knowledge:src/research-loop.ts |
@@ -398,6 +403,7 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `runSemanticConceptJudge` | production | agent-dev-container:products/intelligence/api/src/lib/optimization-engine.ts |
 | `runsForScenario` | production | phony:products/builder/api/src/eval/gate.ts |
 | `runTaskScore` | production | discovery-lab:tools/strict-screen.mjs |
+| `sampleMedian` | planned | doc: docs/concepts.md |
 | `scoreKnowledgeReadiness` | production | agent-builder:src/lib/.server/runtime/agent-task-readiness.ts |
 | `scorePolicyEditReadiness` | none | — |
 | `scoreRedTeamOutput` | production | agent-dev-container:products/intelligence/api/src/lib/optimization-gate.ts |
@@ -409,12 +415,15 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `ServedCrossFamilyError` | none | only this package's tests: src/integrity/served-model.test.ts:2 |
 | `servedModelAcceptable` | production | this package: src/integrity/preflight.ts:33 |
 | `shareAllowed` | planned | doc: docs/redaction.md |
+| `spanRecordsToTraceAnalysisStore` | planned | doc: docs/trace-analysis.md |
 | `spearmanR` | production | blueprint-agent:scripts/experiments/lib/benchmark-validity-report.mjs |
 | `stripFencedJson` | production | blueprint-agent:scripts/experiments/compare-analysis-reports.ts |
+| `studentTTwoSidedPValue` | production | this package: src/baseline.ts:19 |
 | `subjectiveEval` | production | creative-agent:eval/control/creative-onboarding.ts |
 | `summarizeBackendIntegrity` | production | agent-dev-container:products/intelligence/api/src/lib/run-provenance.ts |
 | `summarizeNumberSeries` | production | this package: src/campaign/cell-aggregates.ts:12 |
 | `summarizePreferenceMemory` | production | agent-builder:src/lib/.server/eval/loops/auto-research-runner.ts |
+| `summarizeSample` | planned | doc: docs/concepts.md |
 | `summaryTable` | production | blueprint-agent:scripts/experiments/vb-runrecord-analysis.ts |
 | `textInSnapshot` | none | only this package's tests: src/oracle.test.ts:2 |
 | `toAgentProfileJson` | production | blueprint-agent:scripts/experiments/lib/agent-profile-cell.ts |
@@ -425,6 +434,7 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `traceContract` | production | creative-agent:eval/lib/trace-contracts.ts |
 | `TraceEmitter` | production | agent-builder:src/lib/.server/runtime/trace-runtime.ts |
 | `transientDispatchFailure` | planned | doc: docs/eval-surface-map.md |
+| `twoByTwoPValue` | planned | doc: docs/concepts.md |
 | `UNKNOWN_MODEL` | production | this package: src/campaign/presets/run-profile-matrix.ts:54 |
 | `urlContains` | none | only this package's tests: src/oracle.test.ts:2 |
 | `userQuestionsForKnowledgeGaps` | production | agent-dev-container:products/intelligence/api/src/routes/project-engines.ts |
@@ -441,6 +451,7 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `viteDeployRunner` | production | blueprint-agent:scripts/experiments/lib/deploy-runner.ts |
 | `weightedComposite` | production | agent-app:src/eval/index.ts |
 | `weightedMean` | production | blueprint-agent:scripts/experiments/lib/synthesis-quality-judge.ts |
+| `welchsTTest` | production | this package: src/contract/analyze-runs.ts:25 |
 | `WILCOXON_EXACT_MAX_N` | none | — |
 | `wilcoxonSignedRank` | production | agent-builder:src/lib/.server/eval/loops/differential-eval.ts |
 | `wilson` | production | agent-dev-container:products/intelligence/api/src/lib/project-analysis-coverage-recommendations.ts |
@@ -489,7 +500,7 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `AnalystRegistry` | production | blueprint-agent:scripts/experiments/vb-multi-analyst.ts |
 | `analystUsageReceiptFromPrimeUsage` | production | this package: src/analyst/benchmark-runner-prime.ts:19 |
 | `appendVerificationArtifactsToOtlp` | production | this package: src/analyst/benchmark-public-data.ts:32 |
-| `askTraceQuestions` | planned | doc: docs/jev-behavior-review.md |
+| `askTraceQuestions` | planned | doc: docs/systemone-behavior-review.md |
 | `assertProposalFindings` | production | agent-runtime:src/improvement/code-execution.ts |
 | `behavioralAnalyst` | production | agent-dev-container:products/intelligence/api/src/lib/deep-trace-analyst.ts |
 | `bindAnalyst` | planned | doc: docs/trace-analysis.md |
@@ -644,7 +655,7 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 
 ### `./campaign`
 
-166 value exports — 117 production, 22 planned, 27 none.
+168 value exports — 119 production, 22 planned, 27 none.
 
 | symbol | consumer | evidence |
 | --- | --- | --- |
@@ -665,6 +676,7 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `buildTraceAnalystSurfaceDispatch` | none | only this package's tests: src/campaign/analyst-surface.test.ts:4 |
 | `campaignBreakdown` | production | this package: src/campaign/external-text-evaluation.ts:11 |
 | `campaignCellSearchResult` | production | this package: src/campaign/presets/run-optimization.ts:51 |
+| `campaignCoverage` | production | this package: src/campaign/presets/run-optimization.ts:32 |
 | `campaignMeanComposite` | production | agent-dev-container:products/intelligence/api/src/lib/optimization-engine.ts |
 | `campaignMeasurementDigest` | production | this package: src/contract/self-improve-method.ts:14 |
 | `campaignScenarioIdentity` | production | agent-runtime:src/improvement/method-identity.ts |
@@ -681,6 +693,7 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `costFromLedgerSummary` | production | supervisor-lab:bench/drain/seat.ts |
 | `createProfileMatrixPlan` | production | discovery-lab:tools/run-profile-confirmation.mjs |
 | `createReferenceEquivalenceJudge` | none | only this package's tests: src/reference-equivalence-judge.test.ts:13 |
+| `createRetainedDispatch` | production | this package: scripts/prove-retained-dispatch-integrity.ts:9 |
 | `createRunCostLedger` | production | agent-dev-container:products/intelligence/api/src/lib/eval-engine.ts |
 | `createSearchHistoryReceipt` | production | this package: src/campaign/search-ledger-recording.ts:27 |
 | `crowdedFrontierParent` | production | this package: scripts/search-sim.ts:146 |
@@ -713,13 +726,13 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `inMemoryCampaignStorage` | production | agent-dev-container:products/sandbox/evals/src/auto-optimization.ts |
 | `isProposedCandidate` | production | supervisor-lab:bench/sequential-arena.ts |
 | `isTransientTransportFailure` | production | gtm-agent:eval/suites/persona.ts |
-| `LabeledScenarioStoreError` | none | only this package's tests: tests/campaign/run-campaign.test.ts:6 |
+| `LabeledScenarioStoreError` | none | only this package's tests: tests/campaign/run-campaign.test.ts:7 |
 | `labelTrustRank` | production | this package: src/campaign/labeled-store/fs-adapter.ts:40 |
 | `llmJudge` | production | agent-runtime:examples/agentic-data-creation/offline-fixtures.ts |
 | `loadEvalFixture` | planned | doc: docs/eval-fixtures.md |
 | `loadEvalFixtureScenarios` | planned | example: examples/eval-fixtures-quickstart/index.ts:11 |
 | `loopProvenanceArgsFromResult` | production | this package: src/contract/self-improve.ts:36 |
-| `loopProvenanceSpans` | none | only this package's tests: tests/campaign/presets.test.ts:6 |
+| `loopProvenanceSpans` | none | only this package's tests: tests/campaign/presets.test.ts:7 |
 | `makePlaybackDispatch` | none | only this package's tests: src/campaign/presets/playback.test.ts:6 |
 | `makeProposalFinding` | production | agent-runtime:examples/improve/improve.ts |
 | `neutralizationGate` | planned | consumer tests: tax-agent:tests/eval/benchmarks/taxcalc/evolve.ts |
@@ -907,9 +920,9 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 
 | symbol | consumer | evidence |
 | --- | --- | --- |
-| `asAnalyst` | production | this package: src/jev.ts:5 |
-| `asJudge` | production | this package: src/jev.ts:5 |
-| `createEvaluator` | production | this package: src/jev.ts:5 |
+| `asAnalyst` | production | this package: src/systemone.ts:5 |
+| `asJudge` | production | this package: src/systemone.ts:5 |
+| `createEvaluator` | production | this package: src/systemone.ts:5 |
 
 ### `./experiment`
 
@@ -1061,40 +1074,6 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `TraceSpanEventSchema` | planned | example: examples/hosted-ingest-server/server.ts:33 |
 | `UnixNanoTimestampSchema` | none | — |
 
-### `./jev`
-
-15 value exports — 8 production, 6 planned, 1 none.
-
-| symbol | consumer | evidence |
-| --- | --- | --- |
-| `asAnalyst` | production | this package: src/jev.ts:5 |
-| `asJudge` | production | this package: src/jev.ts:5 |
-| `assessJevReview` | planned | doc: docs/changes/jev-evidence-review.md |
-| `createEvaluator` | production | this package: src/jev.ts:5 |
-| `jevAnalyst` | planned | example: examples/jev-behavior-review.ts:3 |
-| `jevEvaluator` | planned | doc: docs/jev-behavior-review.md |
-| `jevJudge` | planned | example: examples/jev-product-integrity.ts:4 |
-| `JevResponseError` | none | — |
-| `jevReviewFindings` | planned | example: examples/jev-behavior-review.ts:3 |
-| `jevUsage` | production | this package: src/jev.ts:7 |
-| `normalizedJevScore` | planned | doc: docs/jev.md |
-| `parseJevQuestions` | production | this package: src/jev.ts:7 |
-| `parseJevRequest` | production | this package: src/jev-review.ts:9 |
-| `parseJevResult` | production | this package: src/jev-review.ts:9 |
-| `prepareJevReview` | production | this package: src/rl/claim-integrity.ts:39 |
-
-### `./jev/protocol`
-
-5 value exports — 4 production, 0 planned, 1 none.
-
-| symbol | consumer | evidence |
-| --- | --- | --- |
-| `JevResponseError` | none | — |
-| `jevUsage` | production | this package: src/jev.ts:7 |
-| `parseJevQuestions` | production | this package: src/jev.ts:7 |
-| `parseJevRequest` | production | this package: src/jev-review.ts:9 |
-| `parseJevResult` | production | this package: src/jev-review.ts:9 |
-
 ### `./ledger-core`
 
 20 value exports — 19 production, 0 planned, 1 none.
@@ -1116,7 +1095,7 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `readTrustedHeadFile` | production | this package: src/ledger-core/journal.ts:55 |
 | `replayLedgerText` | production | this package: src/campaign/search-ledger.ts:23 |
 | `trustedHeadPathFor` | production | this package: src/ledger-core/journal.ts:55 |
-| `tryAcquireAtomicFileLock` | production | this package: src/campaign/single-run-lock.ts:17 |
+| `tryAcquireAtomicFileLock` | production | this package: scripts/prove-retained-dispatch.ts:10 |
 | `tryWithLedgerFileLock` | production | this package: src/campaign/search-ledger-file.ts:3 |
 | `verifyEntriesAgainstTrustedHead` | production | this package: src/ledger-core/journal.ts:55 |
 | `withLedgerFileLock` | production | this package: src/analyst/benchmark-response-cache.ts:6 |
@@ -1134,12 +1113,13 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 
 ### `./meta-eval`
 
-28 value exports — 13 production, 14 planned, 1 none.
+32 value exports — 13 production, 18 planned, 1 none.
 
 | symbol | consumer | evidence |
 | --- | --- | --- |
 | `auditEvaluator` | planned | example: examples/evaluation-integrity/index.ts:15 |
 | `auditProbabilityPolicy` | planned | doc: docs/changes/probability-policy-audit.md |
+| `brierScore` | planned | doc: docs/systemone.md |
 | `calibrateJudge` | production | creative-agent:eval/calibrate-judges.ts |
 | `calibrateJudgeContinuous` | planned | consumer tests: tax-agent:tests/eval/lib/judge-sentinel.ts |
 | `calibrationCurve` | planned | doc: docs/eval-surface-map.md |
@@ -1149,6 +1129,8 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `correlationStudy` | production | phony:products/builder/api/src/eval/outcomes.ts |
 | `definePlant` | planned | doc: docs/eval-surface-map.md |
 | `evalHealthStamp` | production | creative-agent:eval/lib/judge-sentinel.ts |
+| `evaluatorAdmissionPolicySchema` | planned | doc: docs/evaluator-construction.md |
+| `evaluatorAuditObservationSchema` | planned | doc: docs/evaluator-construction.md |
 | `fileSentinelStore` | production | creative-agent:eval/lib/judge-sentinel.ts |
 | `FileSystemOutcomeStore` | production | phony:products/builder/api/src/eval/outcomes.ts |
 | `InMemoryOutcomeStore` | planned | consumer tests: phony:products/builder/api/src/eval/outcomes-correlate.test.ts |
@@ -1156,6 +1138,7 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `judgeSentinelReport` | production | creative-agent:eval/lib/judge-sentinel.ts |
 | `OutcomeStoreError` | planned | doc: docs/outcome-validity.md |
 | `perturbEvidence` | planned | doc: docs/plants.md |
+| `planEvaluatorAudit` | planned | doc: docs/evaluator-construction.md |
 | `plantByPerturbation` | planned | doc: docs/plants.md |
 | `positionalBias` | planned | doc: docs/concepts.md |
 | `rubricPredictiveValidity` | production | physim:apps/server/src/scripts/agent-eval/validate-rubrics.ts |
@@ -1294,14 +1277,14 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `buildRlDataset` | production | this package: src/rl/corpus.ts:24 |
 | `buildVerifiedFindingRow` | none | only this package's tests: src/rl/verified-findings-dataset.test.ts:5 |
 | `campaignToRunRecords` | production | agent-builder:src/lib/.server/eval/loops/auto-research-runner.ts |
-| `checkerReadSignal` | planned | doc: docs/jev-behavior-review.md |
+| `checkerReadSignal` | planned | doc: docs/systemone-behavior-review.md |
 | `CLAIM_INTEGRITY_VERSION` | none | — |
 | `claimBrief` | none | — |
 | `claimIntegrityDigest` | none | only this package's tests: src/rl/claim-integrity.test.ts:4 |
-| `claimIntegrityJevQuestions` | none | — |
-| `claimIntegrityJevReview` | planned | doc: docs/jev-behavior-review.md |
-| `claimIntegrityTraceQuestions` | planned | doc: docs/jev-behavior-review.md |
-| `claimIntegrityVerdict` | planned | doc: docs/jev-behavior-review.md |
+| `claimIntegritySystemOneQuestions` | none | — |
+| `claimIntegritySystemOneReview` | planned | doc: docs/systemone-behavior-review.md |
+| `claimIntegrityTraceQuestions` | planned | doc: docs/systemone-behavior-review.md |
+| `claimIntegrityVerdict` | planned | doc: docs/systemone-behavior-review.md |
 | `compareAdaptationCurves` | planned | doc: docs/design/statistics-decisions.md |
 | `datasheetToMarkdown` | none | only this package's tests: src/rl/dataset.test.ts:7 |
 | `defaultBehaviorFeatures` | none | only this package's tests: src/rl/sim-fidelity.test.ts:3 |
@@ -1312,29 +1295,28 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `extractStepRewards` | none | only this package's tests: tests/rl-process-reward.test.ts:3 |
 | `extractVerifiableReward` | planned | named in gtm-agent (bind not in the import graph) |
 | `extractVerifiableRewardsFromRecords` | production | creative-agent:eval/canonical-export.ts |
-| `farmingSignal` | planned | doc: docs/jev-behavior-review.md |
+| `farmingSignal` | planned | doc: docs/systemone-behavior-review.md |
 | `FileSystemOutcomeStore` | production | phony:products/builder/api/src/eval/outcomes.ts |
 | `filterDeterministicallyRewarded` | production | this package: src/rl/reward-hacking.ts:43 |
 | `firstPassK` | planned | doc: docs/statistical-evidence.md |
 | `fitBradleyTerry` | none | only this package's tests: tests/rl-tournament.test.ts:2 |
 | `GRADER_REFERENCE_PHRASES` | none | — |
-| `graderReferenceSignal` | planned | doc: docs/jev-behavior-review.md |
+| `graderReferenceSignal` | planned | doc: docs/systemone-behavior-review.md |
 | `injectIrrelevantClause` | none | only this package's tests: tests/rl-contamination.test.ts:2 |
 | `InMemoryOutcomeStore` | planned | consumer tests: phony:products/builder/api/src/eval/outcomes-correlate.test.ts |
 | `inverseProbabilityWeighting` | planned | consumer tests: agent-knowledge:tests/memory-holdout.test.ts |
-| `jevSignal` | planned | doc: docs/jev-behavior-review.md |
 | `jsDivergence` | none | only this package's tests: src/rl/sim-fidelity.test.ts:3 |
 | `loadVerifiedFindingsDataset` | planned | doc: docs/verified-labels-flywheel.md |
-| `monitorSignal` | planned | doc: docs/jev-behavior-review.md |
+| `monitorSignal` | planned | doc: docs/systemone-behavior-review.md |
 | `offPolicyEstimateAll` | none | only this package's tests: tests/rl-off-policy.test.ts:3 |
 | `paretoFrontier` | production | agent-dev-container:products/sandbox/evals/run.ts |
-| `parseSourceScopeReferee` | planned | doc: docs/jev-behavior-review.md |
+| `parseSourceScopeReferee` | planned | doc: docs/systemone-behavior-review.md |
 | `PredictiveValidityResearcher` | planned | doc: docs/design/mlbenchmarks-book-review.md |
 | `prmTrainingPairs` | planned | doc: examples/fine-tune-with-prime-rl/README.md |
 | `quantileEdges` | none | only this package's tests: src/rl/sim-fidelity.test.ts:3 |
 | `readCorpus` | planned | named in agent-runtime (bind not in the import graph) |
 | `REFEREE_VERDICTS` | none | — |
-| `refereeSignal` | planned | doc: docs/jev-behavior-review.md |
+| `refereeSignal` | planned | doc: docs/systemone-behavior-review.md |
 | `renameVariables` | none | only this package's tests: tests/rl-contamination.test.ts:2 |
 | `ruleHolds` | none | only this package's tests: src/rl/claim-integrity.test.ts:4 |
 | `runAdaptationCurve` | planned | doc: docs/statistical-evidence.md |
@@ -1343,14 +1325,15 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `runEvalCampaign` | production | agent-builder:scripts/eval.ts |
 | `runRLCampaign` | planned | doc: docs/outcome-validity.md |
 | `runwiseStepRewardSummary` | none | only this package's tests: tests/rl-process-reward.test.ts:3 |
-| `scopeSignals` | planned | doc: docs/jev-behavior-review.md |
+| `scopeSignals` | planned | doc: docs/systemone-behavior-review.md |
 | `selfConsistency` | none | only this package's tests: tests/rl-adversarial-and-compute.test.ts:2 |
 | `selfNormalizedImportanceWeighting` | planned | consumer tests: agent-knowledge:tests/memory-holdout.test.ts |
 | `shuffleOrder` | none | — |
 | `simFidelityReport` | none | only this package's tests: src/rl/sim-fidelity.test.ts:3 |
-| `sourceScopeRefereeMessages` | planned | doc: docs/jev-behavior-review.md |
+| `sourceScopeRefereeMessages` | planned | doc: docs/systemone-behavior-review.md |
 | `stepRewardsToJsonl` | none | only this package's tests: src/rollout/gate-properties.test.ts:38 |
 | `summarizeVerifiedFindings` | none | only this package's tests: src/rl/verified-findings-dataset.test.ts:5 |
+| `systemOneSignal` | planned | doc: docs/systemone-behavior-review.md |
 | `toAnthropicFormat` | production | blueprint-agent:scripts/experiments/lib/steering-dataset.ts |
 | `toDpoJsonl` | production | this package: src/rl/dataset.ts:27 |
 | `toDpoRows` | production | this package: src/rl/dataset.ts:27 |
@@ -1369,7 +1352,7 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 
 ### `./rollout`
 
-67 value exports — 42 production, 15 planned, 10 none.
+68 value exports — 42 production, 16 planned, 10 none.
 
 | symbol | consumer | evidence |
 | --- | --- | --- |
@@ -1401,13 +1384,14 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `gateGamedOutcome` | none | only this package's tests: src/rollout/gate-properties.test.ts:56 |
 | `HARBOR_IMPORT_GAP` | none | only this package's tests: src/rollout/interchange/harbor.test.ts:6 |
 | `isRealnessGated` | production | starter-foundry:src/lib/held-out-gate.ts |
-| `isRolloutLine` | production | this package: src/supervisor-run/rollout-nodes.ts:26 |
+| `isRolloutLine` | production | this package: src/supervisor-run/rollout-nodes.ts:27 |
 | `isTrainableSplit` | production | this package: src/rollout/release/hf-dataset.ts:24 |
 | `measureFormatGate` | production | this package: src/rollout/release/hf-dataset.ts:26 |
 | `mintRolloutRows` | production | agent-builder:src/lib/.server/eval/loops/auto-research-runner.ts |
 | `observedScore` | production | this package: src/rl/reward-hacking.ts:40 |
 | `observedSplitScore` | production | starter-foundry:src/lib/held-out-gate.ts |
 | `openOpencodeDb` | production | discovery-lab:dashboard/lib/transcript.mjs |
+| `parseClaudeTranscript` | planned | doc: docs/rollout.md |
 | `parseRolloutReleaseArgs` | none | only this package's tests: src/rollout/release/hf-dataset.test.ts:9 |
 | `planPushCommand` | none | only this package's tests: src/rollout/release/hf-dataset.test.ts:9 |
 | `readClaudeTranscript` | planned | doc: docs/rollout.md |
@@ -1430,7 +1414,7 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `toHarborTrajectory` | planned | doc: docs/rollout.md |
 | `toJsonl` | production | supervisor-lab:bench/vertical-rollout-package.ts |
 | `toRewardRows` | planned | doc: docs/rollout.md |
-| `toRftItem` | none | only this package's tests: src/rollout/reward-invariant.test.ts:33 |
+| `toRftItem` | none | only this package's tests: src/rollout/mint.test.ts:9 |
 | `toRftItems` | production | this package: src/rollout/release/hf-dataset.ts:22 |
 | `toSftRows` | production | this package: src/rl/dataset.ts:27 |
 | `toVerifiersRolloutOutput` | none | only this package's tests: src/rollout/exporters.test.ts:2 |
@@ -1539,6 +1523,46 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `writeSupervisorRunReport` | planned | named in agent-runtime (bind not in the import graph) |
 | `writeSupervisorRunReportSafe` | planned | named in agent-runtime (bind not in the import graph) |
 
+### `./systemone`
+
+18 value exports — 8 production, 10 planned, 0 none.
+
+| symbol | consumer | evidence |
+| --- | --- | --- |
+| `asAnalyst` | production | this package: src/systemone.ts:5 |
+| `asJudge` | production | this package: src/systemone.ts:5 |
+| `assessSystemOneReview` | planned | doc: docs/systemone-behavior-review.md |
+| `createEvaluator` | production | this package: src/systemone.ts:5 |
+| `normalizedSystemOneScore` | planned | doc: docs/systemone.md |
+| `normalizeSystemOneResult` | planned | doc: docs/systemone.md |
+| `parseSystemOneQuestions` | production | this package: src/systemone.ts:14 |
+| `parseSystemOneRecordedRequest` | planned | example: examples/systemone-replay.ts:7 |
+| `parseSystemOneRequest` | production | this package: src/systemone-review.ts:11 |
+| `parseSystemOneResult` | production | this package: src/systemone-review.ts:11 |
+| `prepareSystemOneReview` | production | this package: src/rl/claim-integrity.ts:47 |
+| `systemOneAnalyst` | planned | example: examples/systemone-behavior-review.ts:3 |
+| `systemOneConfidence` | planned | doc: docs/systemone.md |
+| `systemOneEvaluator` | planned | doc: docs/systemone-behavior-review.md |
+| `systemOneJudge` | planned | example: examples/systemone-product-integrity.ts:4 |
+| `SystemOneResponseError` | planned | doc: docs/systemone.md |
+| `systemOneReviewFindings` | planned | example: examples/systemone-behavior-review.ts:3 |
+| `systemOneUsage` | production | this package: src/systemone.ts:14 |
+
+### `./systemone/protocol`
+
+8 value exports — 4 production, 4 planned, 0 none.
+
+| symbol | consumer | evidence |
+| --- | --- | --- |
+| `normalizeSystemOneResult` | planned | doc: docs/systemone.md |
+| `parseSystemOneQuestions` | production | this package: src/systemone.ts:14 |
+| `parseSystemOneRecordedRequest` | planned | example: examples/systemone-replay.ts:7 |
+| `parseSystemOneRequest` | production | this package: src/systemone-review.ts:11 |
+| `parseSystemOneResult` | production | this package: src/systemone-review.ts:11 |
+| `systemOneConfidence` | planned | doc: docs/systemone.md |
+| `SystemOneResponseError` | planned | doc: docs/systemone.md |
+| `systemOneUsage` | production | this package: src/systemone.ts:14 |
+
 ### `./trace-attributes`
 
 28 value exports — 26 production, 1 planned, 1 none.
@@ -1679,7 +1703,7 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 
 ### `./traces`
 
-124 value exports — 96 production, 8 planned, 20 none.
+125 value exports — 96 production, 9 planned, 20 none.
 
 | symbol | consumer | evidence |
 | --- | --- | --- |
@@ -1702,7 +1726,7 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `convertTraceStoresToOtlp` | planned | consumer tests: legal-agent:tests/eval/lib/traces-to-otlp.ts |
 | `createBoundedTraceAnalysisStore` | production | braid:src/adapters/analysis/trace-store.ts |
 | `DEFAULT_TRACE_ANALYST_BUDGETS` | production | agent-builder:src/lib/.server/eval/stores/d1-trace-analysis-store-adapter.ts |
-| `defaultProviderRedactor` | production | this package: src/llm-client.ts:53 |
+| `defaultProviderRedactor` | production | this package: src/llm-client.ts:54 |
 | `defaultTraceInsightPanel` | none | only this package's tests: src/trace-analyst/insights.test.ts:3 |
 | `describeTraceInsightScope` | production | blueprint-agent:scripts/experiments/lib/analyze-vb-run/report-html.ts |
 | `detectCredential` | production | this package: scripts/redaction-corpus.ts:9 |
@@ -1763,7 +1787,7 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `OUTPUT_VALUE` | production | agent-runtime:src/runtime/supervise-surface.ts |
 | `planTraceInsightQuestions` | none | only this package's tests: src/trace-analyst/insights.test.ts:3 |
 | `projectOtlpFlatLine` | production | this package: src/trace-analyst/otlp-to-run-records.ts:60 |
-| `providerFromBaseUrl` | production | this package: src/llm-client.ts:53 |
+| `providerFromBaseUrl` | production | this package: src/llm-client.ts:54 |
 | `redact` | production | this package: scripts/redaction-corpus.ts:9 |
 | `redactForShare` | production | this package: scripts/redaction-corpus.ts:9 |
 | `REDACTION_PROFILES` | production | this package: scripts/redaction-corpus.ts:9 |
@@ -1779,6 +1803,7 @@ The `none` set was reviewed symbol by symbol on 2026-08-21. Four rules decided m
 | `shareAllowed` | planned | doc: docs/redaction.md |
 | `SPAN_KIND_ATTR_KEYS` | production | traces:src/run-span-tree.ts |
 | `SpanNotFoundError` | production | this package: src/trace-analyst/store-boundary.ts:1 |
+| `spanRecordsToTraceAnalysisStore` | planned | doc: docs/trace-analysis.md |
 | `stringField` | planned | named in agent-dev-container (bind not in the import graph) |
 | `throwIfRunIncomplete` | none | only this package's tests: tests/run-integrity.test.ts:3 |
 | `tokenizeDomainWords` | planned | consumer tests: blueprint-agent:scripts/experiments/__tests__/analyze-vb-run.test.ts |

package/docs/{jev-behavior-review.md → systemone-behavior-review.md}

@@ -4,17 +4,17 @@ Use native evaluations to assess bounded hypotheses about recorded behavior. Do
 classifier whether an agent is "malicious" or a whole run is "safe". Observable policy breaches,
 model-supported hypotheses, missing evidence and unusual behavior are different results.
 
-The optional `/jev` helpers are ordinary functions:
+The optional `/systemone` helpers are ordinary functions:
 
-- `prepareJevReview`: validate and freeze caller-defined native questions, evidence references,
+- `prepareSystemOneReview`: validate and freeze caller-defined native questions, evidence references,
   coverage and decision thresholds before spending. The digest binds the definition, input and
   authored question/option order. Optional undefined object fields are absent on the wire.
-- `assessJevReview`: validate native answers and interpret designated alternative masses as
+- `assessSystemOneReview`: validate native answers and interpret designated alternative masses as
   `supported`, `refuted` or `unresolved`. Re-scoring makes no network call.
-- `jevReviewFindings`: map that report into the existing `AnalystFinding` contract. It keeps
+- `systemOneReviewFindings`: map that report into the existing `AnalystFinding` contract. It keeps
   unresolved checks visible and labels supported claims as model-supported hypotheses.
 
-These helpers do not constrain `jevEvaluator`, `jevJudge`, `jevAnalyst`, or the provider-independent
+These helpers do not constrain `systemOneEvaluator`, `systemOneJudge`, `systemOneAnalyst`, or the provider-independent
 `createEvaluator`/`asJudge`/`asAnalyst` APIs. Use an ordinary custom mapper for different policies.
 No model, taxonomy, severity, operating threshold, global safety score or storage backend is chosen
 by the generic implementation.
@@ -26,16 +26,16 @@ Supply the native request on each invocation. This example's `question`, `eviden
 
 ```ts
 import {
-  prepareJevReview, assessJevReview, jevEvaluator,
-} from '@tangle-network/agent-eval/jev'
+  prepareSystemOneReview, assessSystemOneReview, systemOneEvaluator,
+} from '@tangle-network/agent-eval/systemone'
 
-const evaluate = jevEvaluator({
+const evaluate = systemOneEvaluator({
   evaluate: nativeTransport, // One Router /v1/systemone attempt; preserve signal/idempotency key.
   receipt: readRouterReceipt,
   maximumCharge: enforcedMaximumCharge,
 })
 
-const review = prepareJevReview({
+const review = prepareSystemOneReview({
   version: policy.version,
   request: {
     model: selectedModel,
@@ -59,7 +59,7 @@ const review = prepareJevReview({
 })
 const result = await evaluate(review.request, executionContext)
 await persistAuthorizedObservation({ review, result })
-const report = assessJevReview(review, result.value)
+const report = assessSystemOneReview(review, result.value)
 ```
 
 The example assumes a native Choice question with `supported`, `refuted` and `insufficient`
@@ -71,8 +71,8 @@ not a requirement imposed on other evaluators.
 
 ## Optional behavior recipe
 
-`examples/jev-behavior-review.ts` provides replaceable question definitions and
-`behaviorReviewAnalyst`, which composes the existing `jevAnalyst`. Copy/adapt the recipe in an
+`examples/systemone-behavior-review.ts` provides replaceable question definitions and
+`behaviorReviewAnalyst`, which composes the existing `systemOneAnalyst`. Copy/adapt the recipe in an
 application; it is not a default policy installed by importing the package. Select only relevant
 checks and provide thresholds, severity, exact policy and evidence for every selected check.
 
@@ -126,7 +126,7 @@ probability with its frozen binary threshold rule; it must not silently drop the
 `/rl` already has `detectRewardHacking` for reward divergence, distribution shift, reward
 disagreement and judge drift. Those diagnostic scores are not probabilities. Keep the raw,
 ungated proxy and independently measured task outcome so a previous rejection does not hide the
-disagreement. Do not substitute the Jev monitor as the only independent outcome.
+disagreement. Do not substitute the System One monitor as the only independent outcome.
 
 The analyst system already has deterministic control-integrity and behavioral/efficiency
 checks, plus deep trace analyst definitions such as intent divergence and knowledge poisoning.
@@ -134,7 +134,7 @@ Run exact checks for facts code can establish; use a bounded semantic review for
 escalate selected cases to the existing deeper analyst. None is replaced by this recipe.
 
 A prepared review can be a custom registry input or built from a filtered trace store. Existing
-`asAnalyst`/`jevAnalyst` forward the shared paid-call account, cancellation and deadline. The
+`asAnalyst`/`systemOneAnalyst` forward the shared paid-call account, cancellation and deadline. The
 recipe verifies persisted definition identity before inference, records the native observation
 before mapping, and returns ordinary findings. Graphs consume these through their existing
 analyst-registry adapter. No new supervisor, scheduler, ledger, or trace database is introduced.
@@ -158,7 +158,7 @@ It decides `pass`, `hack` or `unknown` for one claim from signals the caller com
 - `farmingSignal`: one value of a parameter the scope declares, repeated across a lane's sweep of claims.
 - `graderReferenceSignal`: the claim page justifies itself by the checker or by the printed letter.
 - `refereeSignal`: a model reads the source and the claim's parameters, never the claimant's prose (`sourceScopeRefereeMessages`, `parseSourceScopeReferee`).
-- `jevSignal`: typed questions over the same brief (`claimIntegrityJevReview`). The caller supplies calibrated thresholds.
+- `systemOneSignal`: typed questions over the same brief (`claimIntegritySystemOneReview`). The caller supplies calibrated thresholds.
 - `monitorSignal`: `askTraceQuestions` with `claimIntegrityTraceQuestions`, on escalation only.
 - `checkerReadSignal`: context for a reader; it never moves a verdict.
 
@@ -172,7 +172,7 @@ Keep verdicts, scopes and questions away from the agent that filed the claim: an
 ## Security and validation boundaries
 
 TypeSafe's Jev 1.13 documentation explicitly warns that adversarial state content can move the
-answer. Delimiters and explicit criteria are useful instructions, not a proof of injection
+answer; treat every System One model the same way. Delimiters and explicit criteria are useful instructions, not a proof of injection
 resistance. These monitors are advisory evidence, not the sole authorization/safety barrier.
 Trusted policy and references must come from outside the evaluated agent; tenant scope, redaction,
 credential handling and controls remain with existing application/security owners.
@@ -192,7 +192,7 @@ attacks, not only the worker under a fixed monitor.
 
 The new tests exercise pure mapping, canonical identity, metadata isolation and real
 Eval/AnalystRegistry accounting composition with an injected model boundary. They do not prove
-that Jev detects these behaviors correctly or is robust against adversarial prompts. No live
+that a System One model detects these behaviors correctly or is robust against adversarial prompts. No live
 provider, product UI rollout, runtime hook installation or production enforcement is added here.
 
 Sources reviewed: https://docs.typesafe.ai/confidence and

package/docs/{jev-integrations.md → systemone-integrations.md}

@@ -1,7 +1,7 @@
 # Ten useful integrations, one evaluator
 
-Jev evaluates bounded questions. Applications supply state, questions and the policy interpreting
-answers. jevEvaluator handles the native model; createEvaluator supports other classifiers.
+A System One model evaluates bounded questions. Applications supply state, questions and the policy interpreting
+answers. systemOneEvaluator handles the native model; createEvaluator supports other classifiers.
 asJudge and asAnalyst adapt the same operation to existing consumers.
 
 | Use | Outcome | Existing execution path |
@@ -10,12 +10,12 @@ asJudge and asAnalyst adapt the same operation to existing consumers.
 | Recommendation triage | Prioritize evidence-linked Intelligence recommendations | Product triage selects existing engine findings and preserves their identities. |
 | Trace diagnosis | Narrow likely failures before deeper investigation | AnalystRegistry, scoped evidence, and the existing deep analyst. |
 | Behavioral review | Flag observable policy concerns, not hidden intent | Optional behavior-review recipe plus deterministic reward/control checks. |
-| Rubric judging | Compare product or agent outputs | jevJudge/asJudge with caller mappings and independent final assessment. |
+| Rubric judging | Compare product or agent outputs | systemOneJudge/asJudge with caller mappings and independent final assessment. |
 | Context selection | Select useful retrieved evidence | Native assessments, then deterministic packing with mandatory context protected. |
 | Skill selection | Select from authorized capabilities | Caller alternatives, including no-selection; existing runtime executes actions. |
 | Next verification | Choose a test, retrieval or review that resolves uncertainty | Existing graph checkpoints and awaited boundaries, not new permissions. |
 | Candidate selection | Choose among generated plans or artifacts | Existing matrix/graph, exact checks first, semantic assessment second. |
-| Configuration experiments | Compare questions, context, models and thresholds | examples/jev-decision-benchmark.ts using runAgentMatrix and cost receipts. |
+| Configuration experiments | Compare questions, context, models and thresholds | examples/systemone-decision-benchmark.ts using runAgentMatrix and cost receipts. |
 
 These are compositions, not ten new frameworks. Product triage and the benchmark recipe are concrete
 adoption work; other rows reuse existing APIs or identify application policies to configure. An API

package/docs/{jev-replay.md → systemone-replay.md}

@@ -1,4 +1,4 @@
-# Inspect recorded Jev evidence without new inference
+# Inspect recorded System One evidence without new inference
 
 The historical decoder from #897 is already the protocol owner. This example
 uses that implementation; it adds no parser, SDK, ledger, database or replay
@@ -7,7 +7,7 @@ policy.
 From a repository checkout with dependencies installed:
 
 ```sh
-pnpm exec tsx examples/jev-replay.ts request.json response.json
+pnpm exec tsx examples/systemone-replay.ts request.json response.json
 ```
 
 Supply the original native request and response as separate JSON files. The
@@ -18,8 +18,8 @@ metadata such as generation IDs and decimal receipt strings. Neither input file
 is rewritten. This is offline inspection, not replay of a paid provider request.
 
 Historical null state and null score levels remain readable through
-`parseJevRecordedRequest`. They are still invalid for new paid calls through
-`parseJevRequest`. Do not cast a historical request to the live type, or change
+`parseSystemOneRecordedRequest`. They are still invalid for new paid calls through
+`parseSystemOneRequest`. Do not cast a historical request to the live type, or change
 its stored rubric just to satisfy today's admission schema.
 
 A mismatch, unreadable file or malformed record returns exit code 1 with a
@@ -30,9 +30,9 @@ terminal or output destination and retain its existing access controls.
 The same helper is importable from the example:
 
 ```ts
-import { inspectJevFiles } from './examples/jev-replay'
+import { inspectSystemOneFiles } from './examples/systemone-replay'
 
-const observation = await inspectJevFiles('request.json', 'response.json')
+const observation = await inspectSystemOneFiles('request.json', 'response.json')
 // Apply an application-owned deterministic mapping to the retained answers.
 ```