@tangle-network/agent-eval 0.150.1 → 0.161.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +178 -1
- package/README.md +7 -3
- package/dist/{active-curriculum-C4mk67HP.js → active-curriculum-CD5TU2yW.js} +3 -13
- package/dist/active-curriculum-CD5TU2yW.js.map +1 -0
- package/dist/{agent-profile-cell-BkcRDikH.d.ts → agent-profile-cell-CTOZJUuE.d.ts} +4 -2
- package/dist/agent-profile-cell-CTOZJUuE.d.ts.map +1 -0
- package/dist/analyst/index.d.ts +19 -36
- package/dist/analyst/index.d.ts.map +1 -1
- package/dist/analyst/index.js +8 -8
- package/dist/analyst/index.js.map +1 -1
- package/dist/{backend-integrity-DOCa_QrR.d.ts → backend-integrity-DxuQCu_A.d.ts} +4 -3
- package/dist/backend-integrity-DxuQCu_A.d.ts.map +1 -0
- package/dist/{benchmark-BtAWA8nT.d.ts → benchmark-CGPp-kDC.d.ts} +3 -3
- package/dist/{benchmark-BtAWA8nT.d.ts.map → benchmark-CGPp-kDC.d.ts.map} +1 -1
- package/dist/{benchmark-command-CAFwbH0L.js → benchmark-command-BVtaq_ve.js} +26 -31
- package/dist/benchmark-command-BVtaq_ve.js.map +1 -0
- package/dist/benchmarks/index.d.ts +6 -19
- package/dist/benchmarks/index.d.ts.map +1 -1
- package/dist/benchmarks/index.js +4 -4
- package/dist/benchmarks/index.js.map +1 -1
- package/dist/builder-eval/index.d.ts +3 -3
- package/dist/builder-eval/index.js +3 -3
- package/dist/campaign/index.d.ts +9 -9
- package/dist/campaign/index.js +7 -7
- package/dist/{campaign-CN_7xJdV.js → campaign-BSmOwskD.js} +77 -795
- package/dist/campaign-BSmOwskD.js.map +1 -0
- package/dist/{canonical-D-XsTQ6_.js → canonical-IL-Bu-14.js} +26 -2
- package/dist/canonical-IL-Bu-14.js.map +1 -0
- package/dist/{capture-fetch-BBVFzhkk.d.ts → capture-fetch-CqwsJkkG.d.ts} +3 -3
- package/dist/{capture-fetch-BBVFzhkk.d.ts.map → capture-fetch-CqwsJkkG.d.ts.map} +1 -1
- package/dist/{chat-client-2bVfrzhN.js → chat-client-DlMlAeYI.js} +5 -55
- package/dist/{chat-client-2bVfrzhN.js.map → chat-client-DlMlAeYI.js.map} +1 -1
- package/dist/chat-json-call-6g5sJobJ.js +53 -0
- package/dist/chat-json-call-6g5sJobJ.js.map +1 -0
- package/dist/cli.js +54 -19
- package/dist/cli.js.map +1 -1
- package/dist/{client-LIuo-KPv.js → client-CX7KqIdB.js} +3 -3
- package/dist/client-CX7KqIdB.js.map +1 -0
- package/dist/{client-kPQYT_56.d.ts → client-L9VVPkim.d.ts} +4 -4
- package/dist/{client-kPQYT_56.d.ts.map → client-L9VVPkim.d.ts.map} +1 -1
- package/dist/contract/index.d.ts +13 -27
- package/dist/contract/index.d.ts.map +1 -1
- package/dist/contract/index.js +14 -17
- package/dist/contract/index.js.map +1 -1
- package/dist/{counterfactual--bpysZF0.d.ts → counterfactual-BaFUWK3H.d.ts} +4 -4
- package/dist/{counterfactual--bpysZF0.d.ts.map → counterfactual-BaFUWK3H.d.ts.map} +1 -1
- package/dist/{counterfactual-lDfCx0Uz.js → counterfactual-D_VWavVm.js} +2 -2
- package/dist/{counterfactual-lDfCx0Uz.js.map → counterfactual-D_VWavVm.js.map} +1 -1
- package/dist/{dataset-CJjKqQfA.d.ts → dataset-DQqhOCPt.d.ts} +5 -4
- package/dist/{dataset-CJjKqQfA.d.ts.map → dataset-DQqhOCPt.d.ts.map} +1 -1
- package/dist/{default-registry-Cw0Ohdoj.d.ts → default-registry-G9CKMNkc.d.ts} +7 -8
- package/dist/{default-registry-Cw0Ohdoj.d.ts.map → default-registry-G9CKMNkc.d.ts.map} +1 -1
- package/dist/{define-agent-eval-CEQWL9Hy.d.ts → define-agent-eval-Dx1JnPEa.d.ts} +26 -6
- package/dist/define-agent-eval-Dx1JnPEa.d.ts.map +1 -0
- package/dist/{define-agent-eval-rqNyVhVV.js → define-agent-eval-h-s-sI-v.js} +17 -11
- package/dist/define-agent-eval-h-s-sI-v.js.map +1 -0
- package/dist/{descriptive-B5MwKfbf.js → descriptive-jDOuI6mz.js} +22 -2
- package/dist/descriptive-jDOuI6mz.js.map +1 -0
- package/dist/{dspy-rlm-engine-DHI0WrUU.js → dspy-rlm-engine-DptEII26.js} +95 -15
- package/dist/dspy-rlm-engine-DptEII26.js.map +1 -0
- package/dist/{emitter-CPBAhxum.js → emitter-BpYFQPj4.js} +2 -18
- package/dist/emitter-BpYFQPj4.js.map +1 -0
- package/dist/{emitter-DGQGoLyj.d.ts → emitter-D_jYSGRd.d.ts} +4 -20
- package/dist/{emitter-DGQGoLyj.d.ts.map → emitter-D_jYSGRd.d.ts.map} +1 -1
- package/dist/{engine-BLzhNzoY.d.ts → engine-Cu5qD5Fc.d.ts} +9 -11
- package/dist/{engine-BLzhNzoY.d.ts.map → engine-Cu5qD5Fc.d.ts.map} +1 -1
- package/dist/{eval-campaign-C4jmuM-b.js → eval-campaign-BsXWL2-2.js} +17 -28
- package/dist/eval-campaign-BsXWL2-2.js.map +1 -0
- package/dist/{exact-types-ccQAyut1.d.ts → exact-types-qnexxJ1Z.d.ts} +2 -2
- package/dist/{exact-types-ccQAyut1.d.ts.map → exact-types-qnexxJ1Z.d.ts.map} +1 -1
- package/dist/{exec-y-DCLqK7.js → exec-D9WpA2p-.js} +2 -2
- package/dist/exec-D9WpA2p-.js.map +1 -0
- package/dist/experiment/index.d.ts +45 -11
- package/dist/experiment/index.d.ts.map +1 -1
- package/dist/experiment/index.js +30 -11
- package/dist/experiment/index.js.map +1 -1
- package/dist/{experiment-tracker-0MhuPArU.d.ts → experiment-tracker-DCO6Cz4s.d.ts} +2 -2
- package/dist/{experiment-tracker-0MhuPArU.d.ts.map → experiment-tracker-DCO6Cz4s.d.ts.map} +1 -1
- package/dist/{exporters-q9iL-2Jf.js → exporters-Df7TgHFv.js} +3 -3
- package/dist/exporters-Df7TgHFv.js.map +1 -0
- package/dist/{external-optimizer-contracts-DbLsm4Po.d.ts → external-optimizer-contracts-szBJ_1vh.d.ts} +2 -2
- package/dist/{external-optimizer-contracts-DbLsm4Po.d.ts.map → external-optimizer-contracts-szBJ_1vh.d.ts.map} +1 -1
- package/dist/{external-optimizer-process-Bhmzngf-.js → external-optimizer-process-WosTBChy.js} +4 -4
- package/dist/{external-optimizer-process-Bhmzngf-.js.map → external-optimizer-process-WosTBChy.js.map} +1 -1
- package/dist/{external-optimizer-subprocess-Dn90UqN2.js → external-optimizer-subprocess-BIWbHpgD.js} +10 -6
- package/dist/external-optimizer-subprocess-BIWbHpgD.js.map +1 -0
- package/dist/{failure-cluster-BLURuWG4.d.ts → failure-cluster-CXL8NbEw.d.ts} +3 -3
- package/dist/{failure-cluster-BLURuWG4.d.ts.map → failure-cluster-CXL8NbEw.d.ts.map} +1 -1
- package/dist/{feedback-trajectory-DpTTjo0q.d.ts → feedback-trajectory-B3ZHaHV_.d.ts} +7 -7
- package/dist/{feedback-trajectory-DpTTjo0q.d.ts.map → feedback-trajectory-B3ZHaHV_.d.ts.map} +1 -1
- package/dist/fuzz.js +1 -1
- package/dist/{hf-dataset-XggBupCr.js → hf-dataset-D8_RNIis.js} +4 -4
- package/dist/{hf-dataset-XggBupCr.js.map → hf-dataset-D8_RNIis.js.map} +1 -1
- package/dist/hosted/index.d.ts +3 -14
- package/dist/hosted/index.d.ts.map +1 -1
- package/dist/hosted/index.js +2 -2
- package/dist/{index-CTKpu9ry.d.ts → index-CGtH1piv.d.ts} +48 -26
- package/dist/index-CGtH1piv.d.ts.map +1 -0
- package/dist/{index-BNPtkBPf.d.ts → index-D-IiQIBB.d.ts} +5 -10
- package/dist/index-D-IiQIBB.d.ts.map +1 -0
- package/dist/{index-IQccV3Ou.d.ts → index-D-V8gCs_.d.ts} +13 -90
- package/dist/index-D-V8gCs_.d.ts.map +1 -0
- package/dist/{index-B8Ui1mr1.d.ts → index-lfaSeKSD.d.ts} +18 -2
- package/dist/index-lfaSeKSD.d.ts.map +1 -0
- package/dist/index-vrJugRal.d.ts +1 -0
- package/dist/index.d.ts +68 -125
- package/dist/index.d.ts.map +1 -1
- package/dist/index.js +61 -112
- package/dist/index.js.map +1 -1
- package/dist/{insight-report-BeT8KCgI.d.ts → insight-report-DRe8LB6d.d.ts} +4 -4
- package/dist/{insight-report-BeT8KCgI.d.ts.map → insight-report-DRe8LB6d.d.ts.map} +1 -1
- package/dist/{integrity-DysDBWDu.js → integrity-CyWSSoQS.js} +17 -4
- package/dist/integrity-CyWSSoQS.js.map +1 -0
- package/dist/{integrity-B0dZ96EO.d.ts → integrity-DUNX9Fao.d.ts} +3 -3
- package/dist/{integrity-B0dZ96EO.d.ts.map → integrity-DUNX9Fao.d.ts.map} +1 -1
- package/dist/{judge-calibration-DZkWrm5H.js → judge-calibration-zZjLz8hr.js} +2 -2
- package/dist/{judge-calibration-DZkWrm5H.js.map → judge-calibration-zZjLz8hr.js.map} +1 -1
- package/dist/{kind-factory-DmAa0h3K.js → kind-factory-DY8FdoXf.js} +3 -71
- package/dist/kind-factory-DY8FdoXf.js.map +1 -0
- package/dist/ledger-core/index.d.ts +2 -2
- package/dist/ledger-core/index.js +3 -3
- package/dist/{ledger-core-DTae9rv_.js → ledger-core-BOzlRygb.js} +2 -2
- package/dist/{ledger-core-DTae9rv_.js.map → ledger-core-BOzlRygb.js.map} +1 -1
- package/dist/{llm-client-Bg32RW0j.js → llm-client-hgDieDNN.js} +53 -98
- package/dist/llm-client-hgDieDNN.js.map +1 -0
- package/dist/{llm-judge-CVq33oz1.js → llm-judge-BhasIPFT.js} +1136 -78
- package/dist/llm-judge-BhasIPFT.js.map +1 -0
- package/dist/{matrix-DrVnRp4G.d.ts → matrix-eXKRMHnL.d.ts} +74 -72
- package/dist/matrix-eXKRMHnL.d.ts.map +1 -0
- package/dist/meta-eval/index.d.ts +8 -6
- package/dist/meta-eval/index.d.ts.map +1 -1
- package/dist/meta-eval/index.js +9 -7
- package/dist/meta-eval/index.js.map +1 -1
- package/dist/{mint-BV6tLVWl.js → mint-DfODW1KW.js} +3 -3
- package/dist/{mint-BV6tLVWl.js.map → mint-DfODW1KW.js.map} +1 -1
- package/dist/multishot/golden/index.d.ts +2 -8
- package/dist/multishot/golden/index.d.ts.map +1 -1
- package/dist/multishot/golden/index.js +56 -86
- package/dist/multishot/golden/index.js.map +1 -1
- package/dist/multishot/index.d.ts +11 -46
- package/dist/multishot/index.d.ts.map +1 -1
- package/dist/multishot/index.js +30 -83
- package/dist/multishot/index.js.map +1 -1
- package/dist/openapi.json +1 -1
- package/dist/{opencode-sqlite-DJWAXLms.js → opencode-sqlite-eK6HW6dr.js} +2 -6
- package/dist/{opencode-sqlite-DJWAXLms.js.map → opencode-sqlite-eK6HW6dr.js.map} +1 -1
- package/dist/pipelines/index.d.ts +5 -5
- package/dist/pipelines/index.js +3 -3
- package/dist/{pre-registration-zFSLEiFU.d.ts → pre-registration-CzFCcwYk.d.ts} +55 -40
- package/dist/pre-registration-CzFCcwYk.d.ts.map +1 -0
- package/dist/pre-registration-KN9jkh58.js +110 -0
- package/dist/pre-registration-KN9jkh58.js.map +1 -0
- package/dist/{produced-state-jfk8Du3b.js → produced-state-DZ89riy5.js} +8 -8
- package/dist/produced-state-DZ89riy5.js.map +1 -0
- package/dist/profile-cell.d.ts +1 -1
- package/dist/profile-cell.js +31 -5
- package/dist/profile-cell.js.map +1 -1
- package/dist/{promotion-policy-DLOUkYhI.d.ts → promotion-policy-DtnOIZvk.d.ts} +2 -2
- package/dist/{promotion-policy-DLOUkYhI.d.ts.map → promotion-policy-DtnOIZvk.d.ts.map} +1 -1
- package/dist/{query-Di7eEQ79.js → query-CHmMP42p.js} +20 -11
- package/dist/query-CHmMP42p.js.map +1 -0
- package/dist/{query-CJ_DX8vl.d.ts → query-DxPYqpmT.d.ts} +10 -4
- package/dist/query-DxPYqpmT.d.ts.map +1 -0
- package/dist/raw-provider-sink-BU29Sh8h.d.ts +134 -0
- package/dist/raw-provider-sink-BU29Sh8h.d.ts.map +1 -0
- package/dist/{registry-BQwrSYpC.d.ts → registry-8You7OK1.d.ts} +5 -7
- package/dist/{registry-BQwrSYpC.d.ts.map → registry-8You7OK1.d.ts.map} +1 -1
- package/dist/{release-confidence-BknrpBnO.js → release-confidence-DKfD2RYU.js} +28 -14
- package/dist/release-confidence-DKfD2RYU.js.map +1 -0
- package/dist/{release-confidence-4XrqlpFD.d.ts → release-confidence-Dqt0NFep.d.ts} +7 -6
- package/dist/release-confidence-Dqt0NFep.d.ts.map +1 -0
- package/dist/reporting.d.ts +3 -3
- package/dist/reporting.js +3 -3
- package/dist/{researcher-DJnoUE8c.d.ts → researcher-Cz565b7D.d.ts} +34 -21
- package/dist/researcher-Cz565b7D.d.ts.map +1 -0
- package/dist/{reward-hacking-62tojkQd.d.ts → reward-hacking-MBf7qpSB.d.ts} +2 -2
- package/dist/{reward-hacking-62tojkQd.d.ts.map → reward-hacking-MBf7qpSB.d.ts.map} +1 -1
- package/dist/{reward-hacking-DKI9T52l.js → reward-hacking-t4lB1yt8.js} +3 -3
- package/dist/{reward-hacking-DKI9T52l.js.map → reward-hacking-t4lB1yt8.js.map} +1 -1
- package/dist/rl.d.ts +11 -42
- package/dist/rl.d.ts.map +1 -1
- package/dist/rl.js +41 -24
- package/dist/rl.js.map +1 -1
- package/dist/rollout/index.d.ts +3 -3
- package/dist/rollout/index.js +7 -7
- package/dist/{rollout-ytVQ7WT8.js → rollout-Dm2tSdiQ.js} +6 -6
- package/dist/{rollout-ytVQ7WT8.js.map → rollout-Dm2tSdiQ.js.map} +1 -1
- package/dist/{rubric-predictive-validity-Cwwyd7ah.js → rubric-predictive-validity-CK8SCOg-.js} +6 -16
- package/dist/rubric-predictive-validity-CK8SCOg-.js.map +1 -0
- package/dist/{rubric-predictive-validity-C7LnNvF2.d.ts → rubric-predictive-validity-CxycqzX5.d.ts} +4 -3
- package/dist/rubric-predictive-validity-CxycqzX5.d.ts.map +1 -0
- package/dist/{run-record-D2lDdSAz.js → run-record-BC0ebuRP.js} +2 -2
- package/dist/{run-record-D2lDdSAz.js.map → run-record-BC0ebuRP.js.map} +1 -1
- package/dist/{run-record-DVV82Gwh.d.ts → run-record-VVy4T9OW.d.ts} +3 -3
- package/dist/{run-record-DVV82Gwh.d.ts.map → run-record-VVy4T9OW.d.ts.map} +1 -1
- package/dist/{schema-BtVldJ3T.d.ts → schema-Bjgdsn73.d.ts} +2 -4
- package/dist/{schema-BtVldJ3T.d.ts.map → schema-Bjgdsn73.d.ts.map} +1 -1
- package/dist/{schema-Cef2cFmb.d.ts → schema-BzWDXhOR.d.ts} +2 -5
- package/dist/schema-BzWDXhOR.d.ts.map +1 -0
- package/dist/{schema-C6DW4ZHR.js → schema-C1aaAxTf.js} +2 -2
- package/dist/schema-C1aaAxTf.js.map +1 -0
- package/dist/{schema-CRhEY1SO.js → schema-k6ZBftVv.js} +2 -8
- package/dist/{schema-CRhEY1SO.js.map → schema-k6ZBftVv.js.map} +1 -1
- package/dist/{semantic-concept-judge-laMCnTLn.js → semantic-concept-judge-BSkKKHeq.js} +14 -38
- package/dist/semantic-concept-judge-BSkKKHeq.js.map +1 -0
- package/dist/{sequential-eprocess-CbUt2htw.js → sequential-eprocess-D1jKoihe.js} +49 -2
- package/dist/sequential-eprocess-D1jKoihe.js.map +1 -0
- package/dist/{sequential-C458DXNf.js → sequential-rYW-Ophm.js} +41 -16
- package/dist/sequential-rYW-Ophm.js.map +1 -0
- package/dist/{series-convergence-BxKEgBwA.d.ts → series-convergence-D9WgpXGi.d.ts} +2 -2
- package/dist/{series-convergence-BxKEgBwA.d.ts.map → series-convergence-D9WgpXGi.d.ts.map} +1 -1
- package/dist/{server-dIWwF3j_.js → server-BtFd4uzB.js} +19 -42
- package/dist/server-BtFd4uzB.js.map +1 -0
- package/dist/{skillopt-optimization-method-DLeUcK-K.js → skillopt-optimization-method-DbaekMcn.js} +794 -8
- package/dist/skillopt-optimization-method-DbaekMcn.js.map +1 -0
- package/dist/{skillopt-optimization-method-BDD_o1xE.d.ts → skillopt-optimization-method-x7TTF23P.d.ts} +20 -7
- package/dist/{skillopt-optimization-method-BDD_o1xE.d.ts.map → skillopt-optimization-method-x7TTF23P.d.ts.map} +1 -1
- package/dist/{statistical-heldout-_woZ9q9j.d.ts → statistical-heldout-Cy3EhjlC.d.ts} +21 -8
- package/dist/statistical-heldout-Cy3EhjlC.d.ts.map +1 -0
- package/dist/{steps-AmkT-GIM.d.ts → steps-CiNVJry_.d.ts} +2 -17
- package/dist/steps-CiNVJry_.d.ts.map +1 -0
- package/dist/{store-CT9YIIve.d.ts → store-B06JdC56.d.ts} +2 -2
- package/dist/{store-CT9YIIve.d.ts.map → store-B06JdC56.d.ts.map} +1 -1
- package/dist/{store-otlp-CDYWW_8N.js → store-otlp-C_Rq5I4D.js} +2 -2
- package/dist/{store-otlp-CDYWW_8N.js.map → store-otlp-C_Rq5I4D.js.map} +1 -1
- package/dist/{store-tool-spans-Br2_IUhm.d.ts → store-tool-spans-DPUG7UUY.d.ts} +6 -6
- package/dist/{store-tool-spans-Br2_IUhm.d.ts.map → store-tool-spans-DPUG7UUY.d.ts.map} +1 -1
- package/dist/{store-tool-spans-CykkbOlv.js → store-tool-spans-Dlh9vkFK.js} +3 -3
- package/dist/{store-tool-spans-CykkbOlv.js.map → store-tool-spans-Dlh9vkFK.js.map} +1 -1
- package/dist/storyboard/index.d.ts +1 -1
- package/dist/{summary-report-Blysd6Z2.js → summary-report-BI5hUtvK.js} +7 -17
- package/dist/summary-report-BI5hUtvK.js.map +1 -0
- package/dist/{summary-report-B__Y5ub3.d.ts → summary-report-CC07PhEL.d.ts} +6 -5
- package/dist/summary-report-CC07PhEL.d.ts.map +1 -0
- package/dist/supervisor-run/index.d.ts +27 -18
- package/dist/supervisor-run/index.d.ts.map +1 -1
- package/dist/supervisor-run/index.js +104 -26
- package/dist/supervisor-run/index.js.map +1 -1
- package/dist/{task-failure-attributes--ZTP3tYO.js → task-failure-attributes-DTl-7-Kw.js} +3 -3
- package/dist/{task-failure-attributes--ZTP3tYO.js.map → task-failure-attributes-DTl-7-Kw.js.map} +1 -1
- package/dist/{tool-groups-B4tqh8jB.d.ts → tool-groups-Ci8i9ErB.d.ts} +3 -3
- package/dist/tool-groups-Ci8i9ErB.d.ts.map +1 -0
- package/dist/{tool-waste-BDdBZG1F.js → tool-waste-BqzmVdJk.js} +4 -4
- package/dist/{tool-waste-BDdBZG1F.js.map → tool-waste-BqzmVdJk.js.map} +1 -1
- package/dist/{tool-waste-DjRDEsuI.d.ts → tool-waste-Dro0gJi3.d.ts} +4 -4
- package/dist/{tool-waste-DjRDEsuI.d.ts.map → tool-waste-Dro0gJi3.d.ts.map} +1 -1
- package/dist/trace-repair/index.d.ts +4 -77
- package/dist/trace-repair/index.d.ts.map +1 -1
- package/dist/trace-repair/index.js +5 -15
- package/dist/trace-repair/index.js.map +1 -1
- package/dist/traces.d.ts +13 -23
- package/dist/traces.d.ts.map +1 -1
- package/dist/traces.js +9 -20
- package/dist/traces.js.map +1 -1
- package/dist/{trajectory-YC15QDYQ.d.ts → trajectory-Bi157Gun.d.ts} +3 -3
- package/dist/{trajectory-YC15QDYQ.d.ts.map → trajectory-Bi157Gun.d.ts.map} +1 -1
- package/dist/trajectory-replay/index.d.ts +5 -5
- package/dist/trajectory-replay/index.js +5 -5
- package/dist/{provenance-oA4-zUqm.d.ts → transient-failure-DKF5Mofa.d.ts} +468 -13
- package/dist/transient-failure-DKF5Mofa.d.ts.map +1 -0
- package/dist/types-B3jzCp0p.js.map +1 -1
- package/dist/{types-CLAwnY-L.d.ts → types-BPb2Kf_C2.d.ts} +3 -3
- package/dist/types-BPb2Kf_C2.d.ts.map +1 -0
- package/dist/types-Bfk0uxRj.d.ts +443 -0
- package/dist/types-Bfk0uxRj.d.ts.map +1 -0
- package/dist/{types-DdFNuyxQ.d.ts → types-D4s7Z6nq.d.ts} +30 -6
- package/dist/types-D4s7Z6nq.d.ts.map +1 -0
- package/dist/{types-B2NsbrNy.d.ts → types-D9ssmxKL.d.ts} +3 -3
- package/dist/{types-B2NsbrNy.d.ts.map → types-D9ssmxKL.d.ts.map} +1 -1
- package/dist/{types-yLK8gXE9.d.ts → types-DeIUdzNd.d.ts} +160 -10
- package/dist/types-DeIUdzNd.d.ts.map +1 -0
- package/dist/{verdict-BndeTAh_.js → verdict-B0xltqu6.js} +2 -2
- package/dist/{verdict-BndeTAh_.js.map → verdict-B0xltqu6.js.map} +1 -1
- package/dist/verdict-cache-CdVVTVmn.js +88 -0
- package/dist/verdict-cache-CdVVTVmn.js.map +1 -0
- package/dist/wire/index.d.ts +21 -111
- package/dist/wire/index.d.ts.map +1 -1
- package/dist/wire/index.js +2 -2
- package/docs/adapters-observability.md +9 -23
- package/docs/building-doctrine.md +3 -3
- package/docs/campaign-proposers.md +54 -15
- package/docs/concepts.md +3 -4
- package/docs/design/statistics-decisions.md +89 -1
- package/docs/eval-surface-map.md +14 -0
- package/docs/experiment.md +19 -2
- package/docs/feedback-trajectories.md +1 -1
- package/docs/multishot-golden-records.md +4 -4
- package/docs/public-api.md +1616 -0
- package/docs/research-report-methodology.md +1 -1
- package/docs/search-history-receipts.md +39 -1
- package/docs/trace-analysis.md +1 -1
- package/docs/trace-repair-admission.md +1 -1
- package/docs/trace-repair-continuation.md +1 -1
- package/docs/verdicts.md +24 -0
- package/docs/wire-protocol.md +1 -1
- package/package.json +6 -2
- package/dist/active-curriculum-C4mk67HP.js.map +0 -1
- package/dist/agent-profile-cell-BkcRDikH.d.ts.map +0 -1
- package/dist/backend-integrity-DOCa_QrR.d.ts.map +0 -1
- package/dist/benchmark-command-CAFwbH0L.js.map +0 -1
- package/dist/campaign-CN_7xJdV.js.map +0 -1
- package/dist/canonical-D-XsTQ6_.js.map +0 -1
- package/dist/client-LIuo-KPv.js.map +0 -1
- package/dist/define-agent-eval-CEQWL9Hy.d.ts.map +0 -1
- package/dist/define-agent-eval-rqNyVhVV.js.map +0 -1
- package/dist/descriptive-B5MwKfbf.js.map +0 -1
- package/dist/dspy-rlm-engine-DHI0WrUU.js.map +0 -1
- package/dist/emitter-CPBAhxum.js.map +0 -1
- package/dist/eval-campaign-C4jmuM-b.js.map +0 -1
- package/dist/exec-y-DCLqK7.js.map +0 -1
- package/dist/exporters-q9iL-2Jf.js.map +0 -1
- package/dist/external-optimizer-subprocess-Dn90UqN2.js.map +0 -1
- package/dist/index-B8Ui1mr1.d.ts.map +0 -1
- package/dist/index-BNPtkBPf.d.ts.map +0 -1
- package/dist/index-C1ravkGA.d.ts +0 -1
- package/dist/index-CTKpu9ry.d.ts.map +0 -1
- package/dist/index-IQccV3Ou.d.ts.map +0 -1
- package/dist/integrity-DysDBWDu.js.map +0 -1
- package/dist/kind-factory-DmAa0h3K.js.map +0 -1
- package/dist/llm-client-Bg32RW0j.js.map +0 -1
- package/dist/llm-judge-CVq33oz1.js.map +0 -1
- package/dist/matrix-DrVnRp4G.d.ts.map +0 -1
- package/dist/pre-registration-DakwTRXk.js +0 -96
- package/dist/pre-registration-DakwTRXk.js.map +0 -1
- package/dist/pre-registration-zFSLEiFU.d.ts.map +0 -1
- package/dist/produced-state-jfk8Du3b.js.map +0 -1
- package/dist/provenance-oA4-zUqm.d.ts.map +0 -1
- package/dist/query-CJ_DX8vl.d.ts.map +0 -1
- package/dist/query-Di7eEQ79.js.map +0 -1
- package/dist/release-confidence-4XrqlpFD.d.ts.map +0 -1
- package/dist/release-confidence-BknrpBnO.js.map +0 -1
- package/dist/researcher-DJnoUE8c.d.ts.map +0 -1
- package/dist/rubric-predictive-validity-C7LnNvF2.d.ts.map +0 -1
- package/dist/rubric-predictive-validity-Cwwyd7ah.js.map +0 -1
- package/dist/schema-C6DW4ZHR.js.map +0 -1
- package/dist/schema-Cef2cFmb.d.ts.map +0 -1
- package/dist/semantic-concept-judge-laMCnTLn.js.map +0 -1
- package/dist/sequential-C458DXNf.js.map +0 -1
- package/dist/sequential-eprocess-CbUt2htw.js.map +0 -1
- package/dist/server-dIWwF3j_.js.map +0 -1
- package/dist/skillopt-optimization-method-DLeUcK-K.js.map +0 -1
- package/dist/statistical-heldout-_woZ9q9j.d.ts.map +0 -1
- package/dist/steps-AmkT-GIM.d.ts.map +0 -1
- package/dist/summary-report-B__Y5ub3.d.ts.map +0 -1
- package/dist/summary-report-Blysd6Z2.js.map +0 -1
- package/dist/tool-groups-B4tqh8jB.d.ts.map +0 -1
- package/dist/types-CLAwnY-L.d.ts.map +0 -1
- package/dist/types-DdFNuyxQ.d.ts.map +0 -1
- package/dist/types-jUBXJ7Iz.d.ts +0 -884
- package/dist/types-jUBXJ7Iz.d.ts.map +0 -1
- package/dist/types-yLK8gXE9.d.ts.map +0 -1
- package/dist/verdict-cache-mZf5FEiY.js +0 -107
- package/dist/verdict-cache-mZf5FEiY.js.map +0 -1
|
@@ -1,22 +1,45 @@
|
|
|
1
1
|
import { c as ValidationError, r as CaptureIntegrityError } from "../errors-DEE6u6ot.js";
|
|
2
2
|
import { n as bonferroni, r as holm, t as benjaminiHochberg } from "../multiplicity-DIWHvysC.js";
|
|
3
|
-
import { C as
|
|
3
|
+
import { A as ProportionInterval, B as wilson, C as EProcess, D as eProcess, E as EProcessStep, I as pairedRiskDifference, L as pairedRiskDifferenceExact, P as mcnemar, R as pairedRiskDifferenceScore, S as pairRunRecords, T as EProcessState, a as evaluateHypothesis, b as comparePairedArms, c as signManifest, d as MatchedPair, g as PairedArmRow, i as SignedManifestAlgo, l as verifyManifest, m as PairArmsResult, n as HypothesisResult, o as hashJson, p as PairArmsOptions, r as SignedManifest, s as manifestContentDigest, t as HypothesisManifest, w as EProcessOptions, x as pairArms } from "../pre-registration-CzFCcwYk.js";
|
|
4
4
|
import { c as BOOTSTRAP_GATE_MIN_N, d as PairedBootstrapResult, h as pairedBootstrap, u as PairedBootstrapOptions } from "../paired-promotion-decision-CGzg0cI_.js";
|
|
5
|
-
import { _ as requiredSampleSize, a as ExperimentVerdict, d as inMemoryExperimentStore, f as mulberry32, g as requiredPairedSampleSize, h as pairedMde, i as ExperimentTracker, l as fileExperimentStore, m as mcnemarRequiredN, n as ExperimentRep, p as mcnemarPower, r as ExperimentStats, t as Experiment } from "../experiment-tracker-
|
|
5
|
+
import { _ as requiredSampleSize, a as ExperimentVerdict, d as inMemoryExperimentStore, f as mulberry32, g as requiredPairedSampleSize, h as pairedMde, i as ExperimentTracker, l as fileExperimentStore, m as mcnemarRequiredN, n as ExperimentRep, p as mcnemarPower, r as ExperimentStats, t as Experiment } from "../experiment-tracker-DCO6Cz4s.js";
|
|
6
6
|
import { c as powerPreflight, o as PowerPreflight, s as PowerPreflightOptions } from "../pareto-BqNW3LJR.js";
|
|
7
|
-
import { c as PromotionPolicy, d as paretoSignificanceGate, i as EvidenceVector, l as buildEvidenceVector, u as paretoPolicy } from "../promotion-policy-
|
|
8
|
-
import { c as pairHoldout, d as SequentialDecision, f as SequentialObservation, g as
|
|
7
|
+
import { c as PromotionPolicy, d as paretoSignificanceGate, i as EvidenceVector, l as buildEvidenceVector, u as paretoPolicy } from "../promotion-policy-DtnOIZvk.js";
|
|
8
|
+
import { _ as sequentialPairedGate, c as pairHoldout, d as SequentialDecision, f as SequentialObservation, g as sequentialDecide, i as PairedHoldout, m as SequentialPairedGateOptions, n as HeldoutSignificance, p as SequentialPairedGate, r as HeldoutSignificanceOptions, s as heldoutSignificance } from "../statistical-heldout-Cy3EhjlC.js";
|
|
9
9
|
import { a as PairedEvalueStep, d as sequentialCrossingHorizon, i as PairedEvalueSequence, o as SequentialCrossingHorizon, r as PairedEvalueOptions, s as SequentialCrossingHorizonOptions, u as pairedEvalueSequence } from "../sequential-BhsrMupG.js";
|
|
10
10
|
import { z } from "zod";
|
|
11
11
|
//#region src/experiment/ast.d.ts
|
|
12
|
+
/**
|
|
13
|
+
* The registered-rule AST: every rule an experiment registers is DATA.
|
|
14
|
+
*
|
|
15
|
+
* A closure cannot be canonicalized or hashed; a node tree can. `sealExperiment`
|
|
16
|
+
* hashes the whole tree, and every interpreter in this file takes only a node
|
|
17
|
+
* plus evidence records — no parameter for alpha, threshold, metric, or
|
|
18
|
+
* stopping rule exists on any executable surface. The registered object and
|
|
19
|
+
* the executed object are therefore the same object, and registered-vs-ran
|
|
20
|
+
* drift is unrepresentable rather than checked.
|
|
21
|
+
*
|
|
22
|
+
* Node families:
|
|
23
|
+
* Predicate closed-key comparisons — the only leaf
|
|
24
|
+
* AdmissionRule monotone funnel stages with registered waivers
|
|
25
|
+
* SelectionRule deterministic subsets over a closed field set
|
|
26
|
+
* Estimand what the experiment measures
|
|
27
|
+
* IntervalSpec how uncertainty is computed, seed included
|
|
28
|
+
* Condition decision guards over named derived quantities
|
|
29
|
+
* DecisionRule ordered verdict table, or a registered absence of one
|
|
30
|
+
* Obligation a control that must exist before a verdict class is read
|
|
31
|
+
* ValidityGate pre-spend design checks
|
|
32
|
+
* HaltRule gates as prerequisites — failure refuses the spend
|
|
33
|
+
* BudgetRule spend schedules with a named ledger
|
|
34
|
+
* MatchedBudgetRule arm budget matching as a refusal
|
|
35
|
+
* ReissuePolicy carrier faults are reissued; model outcomes stand
|
|
36
|
+
*/
|
|
12
37
|
/** JSON-serializable value — everything a sealed node may carry. */
|
|
13
38
|
type JsonValue = string | number | boolean | null | JsonValue[] | {
|
|
14
39
|
[k: string]: JsonValue;
|
|
15
40
|
};
|
|
16
41
|
/** Evidence row shape. Fields are addressed by dot-separated paths. */
|
|
17
42
|
type EvidenceRecord = Record<string, unknown>;
|
|
18
|
-
/** A decision rule's branches did not cover the evidence. */
|
|
19
|
-
declare class DecisionTableNotTotalError extends ValidationError {}
|
|
20
43
|
/**
|
|
21
44
|
* Closed-key comparison over a declared record schema. The only leaf node.
|
|
22
45
|
* `field` is a dot-separated path into an evidence record.
|
|
@@ -630,11 +653,22 @@ interface SealAmendment {
|
|
|
630
653
|
/** Digest of the spec after this amendment. */
|
|
631
654
|
digest: string;
|
|
632
655
|
}
|
|
656
|
+
/**
|
|
657
|
+
* Digest scheme of a sealed experiment. Both are sha256 hex over the
|
|
658
|
+
* serialized spec and differ only in the serialization: `'sha256-rfc8785'` is
|
|
659
|
+
* RFC 8785 canonical JSON, `'sha256-content'` is key-sorted `JSON.stringify`.
|
|
660
|
+
*/
|
|
661
|
+
type SealAlgo = 'sha256-content' | 'sha256-rfc8785';
|
|
633
662
|
interface SealedExperiment {
|
|
634
663
|
spec: ExperimentSpec;
|
|
635
|
-
/** sha256
|
|
664
|
+
/** sha256 over the serialized spec, under the scheme `algo` names. */
|
|
636
665
|
digest: string;
|
|
637
|
-
|
|
666
|
+
/**
|
|
667
|
+
* Digest scheme of `digest`. `'sha256-rfc8785'` is what {@link sealExperiment}
|
|
668
|
+
* emits; `'sha256-content'` is read-only, carried by seals from an earlier
|
|
669
|
+
* release, and still verifies.
|
|
670
|
+
*/
|
|
671
|
+
algo: SealAlgo;
|
|
638
672
|
sealedAt: string;
|
|
639
673
|
/** Digest of the original registration, before any amendment. */
|
|
640
674
|
initialDigest: string;
|
|
@@ -655,7 +689,8 @@ declare function amendExperiment(sealed: SealedExperiment, amendment: {
|
|
|
655
689
|
blind: string[];
|
|
656
690
|
at?: string;
|
|
657
691
|
}): Promise<SealedExperiment>;
|
|
658
|
-
/** True when the sealed digest still matches the spec it carries
|
|
692
|
+
/** True when the sealed digest still matches the spec it carries, under the
|
|
693
|
+
* scheme the seal declares. */
|
|
659
694
|
declare function verifySealedExperiment(sealed: SealedExperiment): Promise<boolean>;
|
|
660
695
|
/** Evidence a gate executor consumes, discriminated to match the gate kind. */
|
|
661
696
|
type GateEvidence = {
|
|
@@ -896,7 +931,6 @@ declare const evidenceRegistryRecordSchema: z.ZodObject<{
|
|
|
896
931
|
type EvidenceRegistryRecord = z.infer<typeof evidenceRegistryRecordSchema>;
|
|
897
932
|
/** Parse one record; throws with the zod issue list on any violation. */
|
|
898
933
|
declare function parseEvidenceRegistryRecord(raw: unknown): EvidenceRegistryRecord;
|
|
899
|
-
declare class EvidenceRegistryError extends Error {}
|
|
900
934
|
/**
|
|
901
935
|
* Validate a set of records as one registry: every record parses, ids are
|
|
902
936
|
* unique, and every `supersedes` target exists. Returns records sorted in
|
|
@@ -990,5 +1024,5 @@ declare function clusteredPower(options: ClusteredPowerOptions): ClusteredPowerR
|
|
|
990
1024
|
/** Throw the refusal for callers that want configuration-time failure. */
|
|
991
1025
|
declare function assertDesignAdequate(result: ClusteredPowerResult): void;
|
|
992
1026
|
//#endregion
|
|
993
|
-
export { type AdmissionExecution, type AdmissionPartition, type AdmissionRule, type AdmissionStage, type ArmRealizedBudget, type ArmSpec, BOOTSTRAP_GATE_MIN_N, type BudgetRule, type ClusteredPowerOptions, type ClusteredPowerPoint, type ClusteredPowerRefusal, type ClusteredPowerResult, type ComputedInterval, type Condition, type CreateEvidenceReceiptInput, type DecisionBranch, type DecisionOutcome, type DecisionRule,
|
|
1027
|
+
export { type AdmissionExecution, type AdmissionPartition, type AdmissionRule, type AdmissionStage, type ArmRealizedBudget, type ArmSpec, BOOTSTRAP_GATE_MIN_N, type BudgetRule, type ClusteredPowerOptions, type ClusteredPowerPoint, type ClusteredPowerRefusal, type ClusteredPowerResult, type ComputedInterval, type Condition, type CreateEvidenceReceiptInput, type DecisionBranch, type DecisionOutcome, type DecisionRule, type DerivedQuantities, DesignRefusalError, type EProcess, type EProcessOptions, type EProcessState, type EProcessStep, EVIDENCE_AUTHORITY_KINDS, EVIDENCE_RECEIPT_VERSION, EVIDENCE_STATES, type Estimand, type EstimandResult, type EvidenceAuthority, type EvidenceAuthorityKind, type EvidenceBinding, type EvidenceDenominator, type EvidenceReceipt, type EvidenceReceiptVerification, type EvidenceRecord, type EvidenceRegistryRecord, type EvidenceState, type EvidenceVector, type Experiment, type ExperimentFunnel, type ExperimentRep, type ExperimentSpec, type ExperimentStats, ExperimentTracker, type ExperimentVerdict, FunnelIntegrityError, type FunnelPartitionCount, type FunnelStageCount, type FunnelStageInput, type GateEvidence, type GateResult, type HaltOutcome, type HaltRule, type HeldoutSignificance, type HeldoutSignificanceOptions, type HypothesisManifest, type HypothesisResult, INDEPENDENT_EVIDENCE_AUTHORITY_KINDS, type IntervalSpec, type JsonValue, MatchedBudgetError, type MatchedBudgetRule, type MatchedBudgetVerdict, type MatchedPair, type NLadderProjection, type Obligation, type OutcomeSpec, type PairArmsOptions, type PairArmsResult, type PairedArmRow, type PairedBootstrapOptions, type PairedBootstrapResult, type PairedEvalueOptions, type PairedEvalueSequence, type PairedEvalueStep, type PairedHoldout, type PowerPreflight, type PowerPreflightOptions, type Predicate, type PromotionPolicy, type ProportionInterval, type RegisteredExperiment, type ReissuePolicy, type ReissueVerdict, type SealAmendment, SealIntegrityError, type SealedExperiment, type SeedDerivation, type SelectionRule, type SequentialCrossingHorizon, type SequentialCrossingHorizonOptions, type SequentialDecision, type SequentialObservation, type SequentialPairedGate, type SequentialPairedGateOptions, type SetExpr, type SignFlipFloor, type SignedManifest, type SignedManifestAlgo, type UniformPassDecision, type UniformPassSchedule, type ValidityGate, amendExperiment, assertDesignAdequate, assertFunnelReconciles, assertMatchedBudgets, benjaminiHochberg, bonferroni, buildEvidenceVector, buildFunnel, classifyReissue, clusteredPower, comparePairedArms, composeFunnels, computeEstimand, computeInterval, createEvidenceReceipt, defineExperiment, eProcess, evaluateCondition, evaluateHaltRule, evaluateHypothesis, evaluateIdentityGate, evaluateOracleDeterminismGate, evaluatePopulationReproducibilityGate, evaluatePowerFloorGate, evaluatePredicate, evaluateProvenanceGate, evidenceRegistryRecordSchema, executeAdmissionRule, executeDecisionRule, fileExperimentStore, hashJson, heldoutSignificance, holm, inMemoryExperimentStore, isIndependentEvidence, manifestContentDigest, mcnemar, mcnemarPower, mcnemarRequiredN, mulberry32, openSealedExperiment, pairArms, pairHoldout, pairRunRecords, pairedBootstrap, pairedEvalueSequence, pairedMde, pairedRiskDifference, pairedRiskDifferenceExact, pairedRiskDifferenceScore, paretoPolicy, paretoSignificanceGate, parseEvidenceRegistryRecord, powerPreflight, projectNLadderBudget, readField, renderEvidenceIndex, renderFunnelTable, requiredPairedSampleSize, requiredSampleSize, runSelectionRule, runUniformPassBudget, sealExperiment, sequentialCrossingHorizon, sequentialDecide, sequentialPairedGate, signManifest, validateEvidenceRegistry, verifyEvidenceReceipt, verifyManifest, verifyMatchedBudgets, verifySealedExperiment, wilson };
|
|
994
1028
|
//# sourceMappingURL=index.d.ts.map
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"index.d.ts","names":[],"sources":["../../src/experiment/ast.ts","../../src/experiment/budget.ts","../../src/experiment/funnel.ts","../../src/experiment/define.ts","../../src/attestation.ts","../../src/experiment/evidence-receipt.ts","../../src/experiment/evidence-record.ts","../../src/experiment/power.ts"],"mappings":"
|
|
1
|
+
{"version":3,"file":"index.d.ts","names":[],"sources":["../../src/experiment/ast.ts","../../src/experiment/budget.ts","../../src/experiment/funnel.ts","../../src/experiment/define.ts","../../src/attestation.ts","../../src/experiment/evidence-receipt.ts","../../src/experiment/evidence-record.ts","../../src/experiment/power.ts"],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;KA8BY,+CAA+C;GAAiB,YAAY;;;KAG5E,iBAAiB;;;;;KAWjB;EAEN;EACA;EACA;EACA,OAAO;;EAEP;EAAY;EAAe,QAAQ;;EACnC;EAAa,IAAI;;EACjB;EAAa,IAAI;;EACjB;EAAa,IAAI;;;iBAGP,UAAU,QAAQ,gBAAgB;;iBAUlC,kBAAkB,WAAW,WAAW,QAAQ;;;;;UA4C/C;EACf;EACA,MAAM;EACN;;;UAIe;EACf;;EAEA;EACA,MAAM;EACN;;;UAIe;EACf;EACA,QAAQ;EACR,aAAa;;;;;;;KAUH;EAEN;EACA;EACA;EACA;IAAe;IAAe;;EAC9B;EACA;;EAGA;;EAEA;EACA,MAAM;EACN;IAAS;IAAe;;;;;;;;;;;;;iBAad,iBACd,MAAM,eACN,kBAAkB,kBAClB;;EAEE;;EAEA,QAAQ;;EAER,YAAY,eAAe;;;KA0DnB;EACN;EAAoB;EAAa,OAAO;;EACxC;EAAmB,IAAI;;;;;;KAMjB;EACN;EAAc,OAAO;EAAW;;EAChC;EAA4B,OAAO;EAAW;;EAE9C;EACA;EACA;EACA;EACA;;;;;EAKA;;EAEA;;EAGA;EACA;EACA;EACA,WAAW;EACX,aAAa;;UAGF;EACf;EACA;EACA;;;iBA8Cc,gBACd,UAAU,UACV,eAAe,mBACd;;KA2ES;EAEN;EACA;EACA;EACA;EACA;EACA;;EAEA;EAAyB;;UAEd;EACf;EACA;EACA;;;;;;;;;;iBAWc,gBACd,MAAM,cACN;EACM;EAAc,eAAe;EAAkB;;EAC/C;EAAkB;EAAmB;IAC1C;;KA8HS;EACN;EAAgC;EAAkB;;EAClD;EAAgC;;EAChC;EAA4B;EAAkB;EAAiC;;EAC/E;EAAwB;;EACxB;EAAa,IAAI;;EACjB;EAAa,IAAI;;EACjB;EAAa,IAAI;;;UAGN;EACf,WAAW;IAAiB;IAAe;;EAC3C,YAAY;EACZ,gBAAgB;;iBAGF,kBAAkB,WAAW,WAAW,UAAU;UAyCjD;EACf,MAAM;EACN;EACA;;;;;;;;;KAUU;EACN;EAAe,UAAU;;EAEzB;EACA;EACA;EACA;EACA;IAAmB;IAAuB;;;UAG/B;EACf;EACA;;iBAGc,oBACd,MAAM,cACN,UAAU,oBACT;;UAec;EACf;EACA;EACA;;;KAMU;EAEN;EACA;EACA;EACA;;EAGA;EACA;EACA;EACA;;EAEA;EAA8B,OAAO;;EAErC;EACA;EACA;EACA;IAAO;IAAgB;IAAmB;;;EAE1C;EAAkB;EAAuB;EAAmB;;UAEjD;EACf;EACA;EACA,UAAU;;;;;;iBAOI,8BACd,YACA,MAAM,QAAQ;EAAgB;IAC9B,aAAa,qCACZ;;;;;;iBAiBa,sCACd,YACA,MAAM,QAAQ;EAAgB;IAC9B;EAAe,eAAe;EAAkB,gBAAgB;IAC/D;;iBAkBa,uBACd,YACA,MAAM,QAAQ;EAAgB;IAC9B,YAAY,iBACX;;iBAMa,qBACd,YACA,OAAO,QAAQ;EAAgB;IAC/B;EAAc;EAAgB;IAC7B;;;;;;iBAea,uBACd,YACA,MAAM,QAAQ;EAAgB;IAC9B;EAAkB;EAAgB;MACjC;;UAsBc;EACf;IAAQ;IAAyB;;EACjC;EACA;;UAGe;EACf;EACA;EACA;;iBAGc,iBAAiB,MAAM,UAAU,gBAAgB,eAAe;;;;;KAoBpE;EAEN;EACA;EACA;EACA;EACA;EACA;IAAU;IAAY;;EACtB;;EAGA;EACA;EACA;EACA;EACA;;UAGW;EACf;EACA;EACA;EACA;;UAGe;EACf,WAAW;EACX;;;;;;;;iBASc,qBACd,MAAM,QAAQ;EAAc;IAC5B,uCACC;UAuBc;EACf;EACA;IAAe;IAAW;IAAsB;;EAChD;IAAkB;IAAuC;;;;;;;;iBAQ3C,qBACd,MAAM,QAAQ;EAAc;IAC5B;EAAY;EAAqB;IAChC;;UAoBc;EACf;EACA;EACA;;;UAIe;EACf;EACA;EACA;;KAGU;;;;;;iBAOI,gBACd,QAAQ,eACR,eACA,sBACC;;;;cCn6BU,2BAA2B;UAEvB;EACf;;EAEA;;UAGe;EACf,MAAM;EACN,MAAM;;EAEN;;EAEA;EACA;;EAEA;IAAkB;IAA2B;;;;;;;;iBAQ/B,qBACd,MAAM,mBACN,eAAe,sBACd;;iBAsCa,qBACd,MAAM,mBACN,eAAe,sBACd;;;;cC/DU,6BAA6B;UAEzB;EACf;EACA;EACA;EACA;;EAEA,aAAa;;EAEb;;UAGe;EACf;;EAEA;EACA;;EAEA;;UAGe;EACf;EACA;EACA,QAAQ;EACR;EACA,YAAY;;UAGG;EACf;;EAEA;EACA,aAAa;EACb;;;;;;;;;;;iBAYc,YAAY;EAC1B;EACA;EACA,QAAQ;EACR;IAAe;IAAY;IAAc;;IACvC;;iBAsEY,uBAAuB,QAAQ;UAwB9B;EACf,QAAQ;EACR,WAAW;;EAEX,eAAe,eAAe;;;;;;;;iBAShB,qBACd,MAAM,eACN,kBAAkB,mBACjB;;;;;;iBA8Ca,eACd,OAAO,kBACP,QAAQ,mBACP;;;;;;iBA4Ba,kBAAkB,QAAQ;;;;cCtM7B,2BAA2B;UAIvB;EACf;EACA;;EAEA;;EAEA;;EAEA,OAAO,eAAe;;KAGZ;EAEN;;EAEA;EACA;;EAEA;;EAEA;;EAGA;EACA;EACA;EACA;;;;;;UAOW;EACf;;UAGe;EACf;;EAEA;EACA,MAAM;EACN,SAAS;EACT,YAAY;;EAEZ,aAAa,eAAe;;;;;EAK5B,gBAAgB;EAChB,YAAY,eAAe;EAC3B,YAAY,eAAe;EAC3B,UAAU;EACV,cAAc;;EAEd,QAAQ,eAAe;EACvB,OAAO;EACP,SAAS;EACT,gBAAgB;EAChB,UAAU;EACV,iBAAiB;EACjB;;;;;;;;;;iBAwCc,iBAAiB,MAAM,iBAAiB;UAkHvC;;EAEf;EACA;;EAEA;;EAEA;;;;;;;KAQU;UAEK;EACf,MAAM;;EAEN;;;;;;EAMA,MAAM;EACN;;EAEA;EACA,YAAY;;;iBAIQ,eACpB,MAAM,gBACN;EAAW;IACV,QAAQ;;;;;;iBAkBW,gBACpB,QAAQ,kBACR;EAAa,MAAM;EAAgB;EAAgB;EAAiB;IACnE,QAAQ;;;iBAwBW,uBAAuB,QAAQ,mBAAmB;;KAoC5D;EACN;EAA4B,aAAa;;EAEzC;EACA,eAAe;EACf,gBAAgB;;EAEhB;EAA8B,YAAY;;EAC1C;EAAkB;EAAgB;;EAClC;EAAqB;IAAkB;IAAgB;;;;;;;UAM5C;WACN,QAAQ;;EAEjB,OAAO,UAAU,oBAAoB;;EAErC,MAAM,kBAAkB,mBAAmB;;EAE3C,OAAO,cAAc,kBAAkB,kBAAkB;IAAW;;;EAEpE,KAAK,cAAc,UAAU,eAAe;;EAE5C,KAAK,gBAAgB,eAAe;;EAEpC,qBAAqB,uCAAuC;;EAE5D,qBAAqB;IAAY;IAAqB;MAAiB;;EAEvE,eAAe,eAAe,sBAAsB;;EAEpD,SAAS,cAAc,eAAe,mBAAmB;;EAEzD,SACE,cACA;IACM;IAAc,eAAe;IAAkB;;IAC/C;IAAkB;IAAmB;MAC1C;;;;;;;iBAQiB,qBACpB,QAAQ,mBACP,QAAQ;;;;;;;;;;;;;;;;;;;;;;;cC1aE;UAEI;;EAEf,eAAe;;EAEf;;;EAGA;;EAEA;;EAEA;;;EAGA;;UAGe;;EAEf;EACA,YAAY;EACZ,kBAAkB;;;;;;EAMlB;;;;cChCW;;cAGA;KAQD,gCAAgC;cAE/B;UAOI;WACN,MAAM;;WAEN;;UAGM;WACN,sBAAsB;;WAEtB;;WAEA;;WAEA;;WAEA;;WAEA;;WAEA;;WAEA;;WAEA;WACA,WAAW;;WAEX;;WAEA;;UAGM;WACN,SAAS;WACT,aAAa;;UAGP;WACN;WACA;;UAGM,mCAAmC,KAAK;;;;;;;iBAQzC,sBACd,OAAO,4BACP,YAAY,wBACX;;;;;;;iBA8Ba,sBAAsB,SAAS,kBAAkB;;;;;;iBA6CjD,sBAAsB,SAAS;;;;;;;;;;;;;;;cClIlC;KAQD,wBAAwB;;cAO9B,2BAAyB,EAAA;;;;GAM7B,EAAA,KAAA;KAEU,sBAAsB,EAAE,aAAa;cAEpC,8BAA4B,EAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAsCvC,EAAA,KAAA;KAEU,yBAAyB,EAAE,aAAa;;iBAGpC,4BAA4B,eAAe;;;;;;iBAW3C,yBAAyB,2BAA2B;;;;;;iBA4CpD,oBAAoB;;;;cCpIvB,2BAA2B;UAEvB;;EAEf;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA;;EAEA;;;;;;EAMA;IAAkB;IAAe;;;UAGlB;EACf;EACA;EACA;;UAGe;;EAEf;;EAEA;EACA;;EAEA;;EAEA;;UAGe;EACf;EACA;EACA;;UAGe;EACf;EACA;EACA;EACA;EACA;EACA;EACA;EACA,OAAO;;EAEP;EACA,eAAe;;EAEf;;EAEA,SAAS;;;;;;iBAOK,eAAe,SAAS,wBAAwB;;iBAsGhD,qBAAqB,QAAQ"}
|
package/dist/experiment/index.js
CHANGED
|
@@ -1,16 +1,18 @@
|
|
|
1
1
|
import { n as CaptureIntegrityError, s as ValidationError } from "../errors-Dngq5h35.js";
|
|
2
|
-
import {
|
|
2
|
+
import { r as canonicalString } from "../canonical-IL-Bu-14.js";
|
|
3
|
+
import { a as verifyManifest, i as signManifest, n as hashJson, r as manifestContentDigest, t as evaluateHypothesis } from "../pre-registration-KN9jkh58.js";
|
|
3
4
|
import { c as mulberry32 } from "../internal-BDHPCnjk.js";
|
|
4
5
|
import { a as requiredSampleSize, c as holm, i as requiredPairedSampleSize, n as mcnemarRequiredN, o as benjaminiHochberg, r as pairedMde, s as bonferroni, t as mcnemarPower } from "../power-and-mde-CHIrXJll.js";
|
|
5
6
|
import { f as mcnemar, g as pairedRiskDifferenceScore, h as pairedRiskDifferenceExact, m as pairedRiskDifference, n as pairArms, r as pairRunRecords, t as comparePairedArms, v as wilson } from "../paired-arms-D-XRF_fy.js";
|
|
6
7
|
import { r as pairedBootstrap, t as BOOTSTRAP_GATE_MIN_N } from "../paired-tests-BHIhYVdu.js";
|
|
7
|
-
import { t as eProcess } from "../sequential-eprocess-
|
|
8
|
-
import { n as contentHash } from "../verdict-cache-
|
|
8
|
+
import { t as eProcess } from "../sequential-eprocess-D1jKoihe.js";
|
|
9
|
+
import { n as contentHash } from "../verdict-cache-CdVVTVmn.js";
|
|
9
10
|
import { a as heldoutSignificance, o as pairHoldout, t as powerPreflight } from "../power-preflight-DEw-uC7q.js";
|
|
10
11
|
import { a as inMemoryExperimentStore, r as fileExperimentStore, t as ExperimentTracker } from "../experiment-tracker-Ym6rEQT1.js";
|
|
11
12
|
import { n as pairedEvalueSequence, r as sequentialCrossingHorizon } from "../sequential-CzK5DarL.js";
|
|
12
13
|
import { n as paretoPolicy, r as paretoSignificanceGate, t as buildEvidenceVector } from "../promotion-policy-xzA40Evo.js";
|
|
13
|
-
import { n as sequentialPairedGate, t as sequentialDecide } from "../sequential-
|
|
14
|
+
import { n as sequentialPairedGate, t as sequentialDecide } from "../sequential-rYW-Ophm.js";
|
|
15
|
+
import { createHash } from "node:crypto";
|
|
14
16
|
import { z } from "zod";
|
|
15
17
|
//#region src/experiment/ast.ts
|
|
16
18
|
/**
|
|
@@ -864,11 +866,11 @@ function deepFreeze(value) {
|
|
|
864
866
|
/** Validate, canonicalize, and hash a spec into its registration. */
|
|
865
867
|
async function sealExperiment(spec, options = {}) {
|
|
866
868
|
const validated = defineExperiment(spec);
|
|
867
|
-
const digest =
|
|
869
|
+
const digest = specDigest(validated, "sha256-rfc8785");
|
|
868
870
|
return {
|
|
869
871
|
spec: validated,
|
|
870
872
|
digest,
|
|
871
|
-
algo: "sha256-
|
|
873
|
+
algo: "sha256-rfc8785",
|
|
872
874
|
sealedAt: options.sealedAt ?? (/* @__PURE__ */ new Date()).toISOString(),
|
|
873
875
|
initialDigest: digest,
|
|
874
876
|
amendments: []
|
|
@@ -882,11 +884,11 @@ async function sealExperiment(spec, options = {}) {
|
|
|
882
884
|
async function amendExperiment(sealed, amendment) {
|
|
883
885
|
await assertSealIntact(sealed);
|
|
884
886
|
const validated = defineExperiment(amendment.spec);
|
|
885
|
-
const digest =
|
|
887
|
+
const digest = specDigest(validated, "sha256-rfc8785");
|
|
886
888
|
return {
|
|
887
889
|
spec: validated,
|
|
888
890
|
digest,
|
|
889
|
-
algo: "sha256-
|
|
891
|
+
algo: "sha256-rfc8785",
|
|
890
892
|
sealedAt: sealed.sealedAt,
|
|
891
893
|
initialDigest: sealed.initialDigest,
|
|
892
894
|
amendments: [...sealed.amendments, {
|
|
@@ -897,9 +899,26 @@ async function amendExperiment(sealed, amendment) {
|
|
|
897
899
|
}]
|
|
898
900
|
};
|
|
899
901
|
}
|
|
900
|
-
/** True when the sealed digest still matches the spec it carries
|
|
902
|
+
/** True when the sealed digest still matches the spec it carries, under the
|
|
903
|
+
* scheme the seal declares. */
|
|
901
904
|
async function verifySealedExperiment(sealed) {
|
|
902
|
-
return
|
|
905
|
+
return specDigest(sealed.spec, sealed.algo) === sealed.digest;
|
|
906
|
+
}
|
|
907
|
+
/**
|
|
908
|
+
* Serialize a spec under `algo` and digest it. `'sha256-content'` is read-only
|
|
909
|
+
* — it exists so a seal written by an earlier release still verifies, and no
|
|
910
|
+
* path that WRITES a digest may pass it.
|
|
911
|
+
*/
|
|
912
|
+
function specDigest(spec, algo) {
|
|
913
|
+
const serialized = algo === "sha256-rfc8785" ? canonicalString(spec) : JSON.stringify(sortKeysDeep(spec));
|
|
914
|
+
return createHash("sha256").update(serialized, "utf8").digest("hex");
|
|
915
|
+
}
|
|
916
|
+
function sortKeysDeep(value) {
|
|
917
|
+
if (value === null || typeof value !== "object") return value;
|
|
918
|
+
if (Array.isArray(value)) return value.map(sortKeysDeep);
|
|
919
|
+
const out = {};
|
|
920
|
+
for (const key of Object.keys(value).sort()) out[key] = sortKeysDeep(value[key]);
|
|
921
|
+
return out;
|
|
903
922
|
}
|
|
904
923
|
async function assertSealIntact(sealed) {
|
|
905
924
|
if (!await verifySealedExperiment(sealed)) throw new SealIntegrityError(`sealed experiment '${sealed.spec.id}' digest ${sealed.digest} does not match its spec — the registration was tampered with`);
|
|
@@ -1510,6 +1529,6 @@ function mixSeed(seed, effect) {
|
|
|
1510
1529
|
return (seed ^ Math.round(effect * 1000003) * 2654435769) >>> 0 | 0;
|
|
1511
1530
|
}
|
|
1512
1531
|
//#endregion
|
|
1513
|
-
export { BOOTSTRAP_GATE_MIN_N,
|
|
1532
|
+
export { BOOTSTRAP_GATE_MIN_N, DesignRefusalError, EVIDENCE_AUTHORITY_KINDS, EVIDENCE_RECEIPT_VERSION, EVIDENCE_STATES, ExperimentTracker, FunnelIntegrityError, INDEPENDENT_EVIDENCE_AUTHORITY_KINDS, MatchedBudgetError, SealIntegrityError, amendExperiment, assertDesignAdequate, assertFunnelReconciles, assertMatchedBudgets, benjaminiHochberg, bonferroni, buildEvidenceVector, buildFunnel, classifyReissue, clusteredPower, comparePairedArms, composeFunnels, computeEstimand, computeInterval, createEvidenceReceipt, defineExperiment, eProcess, evaluateCondition, evaluateHaltRule, evaluateHypothesis, evaluateIdentityGate, evaluateOracleDeterminismGate, evaluatePopulationReproducibilityGate, evaluatePowerFloorGate, evaluatePredicate, evaluateProvenanceGate, evidenceRegistryRecordSchema, executeAdmissionRule, executeDecisionRule, fileExperimentStore, hashJson, heldoutSignificance, holm, inMemoryExperimentStore, isIndependentEvidence, manifestContentDigest, mcnemar, mcnemarPower, mcnemarRequiredN, mulberry32, openSealedExperiment, pairArms, pairHoldout, pairRunRecords, pairedBootstrap, pairedEvalueSequence, pairedMde, pairedRiskDifference, pairedRiskDifferenceExact, pairedRiskDifferenceScore, paretoPolicy, paretoSignificanceGate, parseEvidenceRegistryRecord, powerPreflight, projectNLadderBudget, readField, renderEvidenceIndex, renderFunnelTable, requiredPairedSampleSize, requiredSampleSize, runSelectionRule, runUniformPassBudget, sealExperiment, sequentialCrossingHorizon, sequentialDecide, sequentialPairedGate, signManifest, validateEvidenceRegistry, verifyEvidenceReceipt, verifyManifest, verifyMatchedBudgets, verifySealedExperiment, wilson };
|
|
1514
1533
|
|
|
1515
1534
|
//# sourceMappingURL=index.js.map
|