@tangle-network/agent-eval 0.163.2 → 0.171.0
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 +127 -0
- package/README.md +2 -0
- package/dist/adapters/http.d.ts +108 -0
- package/dist/adapters/http.d.ts.map +1 -0
- package/dist/adapters/http.js +208 -0
- package/dist/adapters/http.js.map +1 -0
- package/dist/analyst/index.d.ts +40 -70
- package/dist/analyst/index.d.ts.map +1 -1
- package/dist/analyst/index.js +18 -311
- package/dist/analyst/index.js.map +1 -1
- package/dist/{backend-integrity-DxuQCu_A.d.ts → backend-integrity-e79K3UPD.d.ts} +3 -3
- package/dist/{backend-integrity-DxuQCu_A.d.ts.map → backend-integrity-e79K3UPD.d.ts.map} +1 -1
- package/dist/{benchmark-BhT16ep9.js → benchmark-C4wk_Sjr.js} +10 -3
- package/dist/benchmark-C4wk_Sjr.js.map +1 -0
- package/dist/{benchmark-command-CF-4GEWZ.js → benchmark-command-D8k3Gf0J.js} +236 -251
- package/dist/benchmark-command-D8k3Gf0J.js.map +1 -0
- package/dist/{benchmark-CGPp-kDC.d.ts → benchmark-h-h4bfqj.d.ts} +3 -3
- package/dist/{benchmark-CGPp-kDC.d.ts.map → benchmark-h-h4bfqj.d.ts.map} +1 -1
- package/dist/benchmarks/index.d.ts +5 -5
- package/dist/benchmarks/index.js +3 -3
- package/dist/builder-eval/index.d.ts +3 -3
- package/dist/builder-eval/index.js +1 -1
- package/dist/campaign/index.d.ts +8 -8
- package/dist/campaign/index.js +7 -7
- package/dist/{campaign-DQZmc2Dq.js → campaign-B72njjHj.js} +15 -14
- package/dist/campaign-B72njjHj.js.map +1 -0
- package/dist/{canonical-IL-Bu-14.js → canonical-DPyQ_rpt.js} +22 -2
- package/dist/{canonical-IL-Bu-14.js.map → canonical-DPyQ_rpt.js.map} +1 -1
- package/dist/{chat-client-DlMlAeYI.js → chat-client-DEtybj5i.js} +5 -5
- package/dist/{chat-client-DlMlAeYI.js.map → chat-client-DEtybj5i.js.map} +1 -1
- package/dist/cli.js +2 -2
- package/dist/{client-CX7KqIdB.js → client-BvwNkIRN.js} +2 -2
- package/dist/{client-CX7KqIdB.js.map → client-BvwNkIRN.js.map} +1 -1
- package/dist/{client-L9VVPkim.d.ts → client-CDtcZ3p9.d.ts} +4 -4
- package/dist/{client-L9VVPkim.d.ts.map → client-CDtcZ3p9.d.ts.map} +1 -1
- package/dist/contract/index.d.ts +12 -703
- package/dist/contract/index.js +11 -11
- package/dist/{counterfactual-BaFUWK3H.d.ts → counterfactual-Bee5_BIn.d.ts} +4 -4
- package/dist/{counterfactual-BaFUWK3H.d.ts.map → counterfactual-Bee5_BIn.d.ts.map} +1 -1
- package/dist/{default-registry-G9CKMNkc.d.ts → default-registry-B0s2zU-s.d.ts} +6 -6
- package/dist/{default-registry-G9CKMNkc.d.ts.map → default-registry-B0s2zU-s.d.ts.map} +1 -1
- package/dist/{define-agent-eval-Dx1JnPEa.d.ts → define-agent-eval-Cjy2yhqP.d.ts} +7 -7
- package/dist/{define-agent-eval-Dx1JnPEa.d.ts.map → define-agent-eval-Cjy2yhqP.d.ts.map} +1 -1
- package/dist/{define-agent-eval-D08pWIJb.js → define-agent-eval-Dy8QgxAI.js} +20 -8
- package/dist/{define-agent-eval-D08pWIJb.js.map → define-agent-eval-Dy8QgxAI.js.map} +1 -1
- package/dist/{dspy-rlm-engine-DhA9qKIm.js → dspy-rlm-engine-CS3qcCEk.js} +3 -10
- package/dist/dspy-rlm-engine-CS3qcCEk.js.map +1 -0
- package/dist/{emitter-D_jYSGRd.d.ts → emitter-Bvnu0VzL.d.ts} +3 -3
- package/dist/{emitter-D_jYSGRd.d.ts.map → emitter-Bvnu0VzL.d.ts.map} +1 -1
- package/dist/{engine-Cu5qD5Fc.d.ts → engine-CAmTUk52.d.ts} +7 -7
- package/dist/{engine-Cu5qD5Fc.d.ts.map → engine-CAmTUk52.d.ts.map} +1 -1
- package/dist/{eval-campaign-BfohKmzx.js → eval-campaign-JDTeE6Pl.js} +4 -4
- package/dist/{eval-campaign-BfohKmzx.js.map → eval-campaign-JDTeE6Pl.js.map} +1 -1
- package/dist/{exact-types-qnexxJ1Z.d.ts → exact-types-BEecmnWm.d.ts} +2 -2
- package/dist/{exact-types-qnexxJ1Z.d.ts.map → exact-types-BEecmnWm.d.ts.map} +1 -1
- package/dist/experiment/index.d.ts +5 -5
- package/dist/experiment/index.js +4 -4
- package/dist/{experiment-tracker-DCO6Cz4s.d.ts → experiment-tracker-Dm8yQMqb.d.ts} +2 -2
- package/dist/{experiment-tracker-DCO6Cz4s.d.ts.map → experiment-tracker-Dm8yQMqb.d.ts.map} +1 -1
- package/dist/{external-optimizer-process-BFmh36vW.js → external-optimizer-process-CQxylYeG.js} +4 -11
- package/dist/external-optimizer-process-CQxylYeG.js.map +1 -0
- package/dist/{external-optimizer-subprocess-CqLMW3nh.js → external-optimizer-subprocess-Cex8Da2i.js} +25 -11
- package/dist/external-optimizer-subprocess-Cex8Da2i.js.map +1 -0
- package/dist/{failure-cluster-CXL8NbEw.d.ts → failure-cluster-6YSvsKlp.d.ts} +3 -3
- package/dist/{failure-cluster-CXL8NbEw.d.ts.map → failure-cluster-6YSvsKlp.d.ts.map} +1 -1
- package/dist/{feedback-trajectory-B3ZHaHV_.d.ts → feedback-trajectory-DIqpCyF0.d.ts} +6 -6
- package/dist/{feedback-trajectory-B3ZHaHV_.d.ts.map → feedback-trajectory-DIqpCyF0.d.ts.map} +1 -1
- package/dist/fuzz.d.ts +1 -1
- package/dist/{skillopt-optimization-method-x7TTF23P.d.ts → heldout-gate-Dh2b62w8.d.ts} +111 -111
- package/dist/heldout-gate-Dh2b62w8.d.ts.map +1 -0
- package/dist/hosted/index.d.ts +2 -2
- package/dist/hosted/index.js +1 -1
- package/dist/{index-D-V8gCs_.d.ts → index-8VIogTyS.d.ts} +30 -22
- package/dist/index-8VIogTyS.d.ts.map +1 -0
- package/dist/{index-D-IiQIBB.d.ts → index-DMoxLG8P.d.ts} +3 -3
- package/dist/{index-D-IiQIBB.d.ts.map → index-DMoxLG8P.d.ts.map} +1 -1
- package/dist/{index-D_P7Ye43.d.ts → index-DNgf5gyG.d.ts} +2 -2
- package/dist/{index-D_P7Ye43.d.ts.map → index-DNgf5gyG.d.ts.map} +1 -1
- package/dist/{index-CGtH1piv.d.ts → index-DT73JraI.d.ts} +7 -38
- package/dist/index-DT73JraI.d.ts.map +1 -0
- package/dist/index-fNXZMCzX.d.ts +704 -0
- package/dist/index-fNXZMCzX.d.ts.map +1 -0
- package/dist/index.d.ts +197 -38
- package/dist/index.d.ts.map +1 -1
- package/dist/index.js +736 -28
- package/dist/index.js.map +1 -1
- package/dist/{insight-report-DRe8LB6d.d.ts → insight-report-08F022xN.d.ts} +4 -4
- package/dist/{insight-report-DRe8LB6d.d.ts.map → insight-report-08F022xN.d.ts.map} +1 -1
- package/dist/{integrity-DUNX9Fao.d.ts → integrity-B_EDELom.d.ts} +2 -2
- package/dist/{integrity-DUNX9Fao.d.ts.map → integrity-B_EDELom.d.ts.map} +1 -1
- package/dist/internal-BMFSR8Ns.js.map +1 -1
- package/dist/{kind-factory-DY8FdoXf.js → kind-factory-DMeEoMQZ.js} +3 -10
- package/dist/kind-factory-DMeEoMQZ.js.map +1 -0
- package/dist/ledger-core/index.js +2 -2
- package/dist/{ledger-core-BOzlRygb.js → ledger-core-PIfjCbKn.js} +2 -2
- package/dist/{ledger-core-BOzlRygb.js.map → ledger-core-PIfjCbKn.js.map} +1 -1
- package/dist/{llm-judge-Du7WQPh7.js → llm-judge-BtJ2Sfk_.js} +101 -20
- package/dist/llm-judge-BtJ2Sfk_.js.map +1 -0
- package/dist/matrix/index.d.ts +2 -2
- package/dist/{matrix-eXKRMHnL.d.ts → matrix-DiHmUobV.d.ts} +3 -3
- package/dist/{matrix-eXKRMHnL.d.ts.map → matrix-DiHmUobV.d.ts.map} +1 -1
- package/dist/meta-eval/index.d.ts +3 -3
- package/dist/meta-eval/index.js +1 -1
- package/dist/{mint-DfODW1KW.js → mint-DjfDUMHr.js} +2 -2
- package/dist/{mint-DfODW1KW.js.map → mint-DjfDUMHr.js.map} +1 -1
- package/dist/multishot/golden/index.d.ts +1 -1
- package/dist/multishot/index.d.ts +2 -2
- package/dist/openapi.json +4 -4
- package/dist/pipelines/index.d.ts +5 -5
- package/dist/pipelines/index.js +3 -3
- package/dist/{pareto-BqNW3LJR.d.ts → power-preflight-Ptse_Kq7.d.ts} +43 -43
- package/dist/power-preflight-Ptse_Kq7.d.ts.map +1 -0
- package/dist/{pre-registration-KN9jkh58.js → pre-registration-D94b7Of5.js} +2 -2
- package/dist/{pre-registration-KN9jkh58.js.map → pre-registration-D94b7Of5.js.map} +1 -1
- package/dist/{pre-registration-CzFCcwYk.d.ts → pre-registration-DHz6P_6f.d.ts} +2 -2
- package/dist/{pre-registration-CzFCcwYk.d.ts.map → pre-registration-DHz6P_6f.d.ts.map} +1 -1
- package/dist/{produced-state-Be0BK3RN.js → produced-state-Cm6DU_Ao.js} +4 -4
- package/dist/{produced-state-Be0BK3RN.js.map → produced-state-Cm6DU_Ao.js.map} +1 -1
- package/dist/profile-cell.js +1 -1
- package/dist/{promotion-policy-DtnOIZvk.d.ts → promotion-policy-WSXtBgBb.d.ts} +3 -3
- package/dist/{promotion-policy-DtnOIZvk.d.ts.map → promotion-policy-WSXtBgBb.d.ts.map} +1 -1
- package/dist/{run-score-lDzV0X8j.js → proposal-findings-bko3GGy-.js} +2 -31
- package/dist/proposal-findings-bko3GGy-.js.map +1 -0
- package/dist/{transient-failure-DKF5Mofa.d.ts → provenance-CafMdZKM.d.ts} +895 -891
- package/dist/provenance-CafMdZKM.d.ts.map +1 -0
- package/dist/{query-_5g6re3_.js → query-BPGMVlbM.js} +3 -3
- package/dist/{query-_5g6re3_.js.map → query-BPGMVlbM.js.map} +1 -1
- package/dist/{query-CwnHlu5p.d.ts → query-Na5gEIGd.d.ts} +3 -3
- package/dist/{query-CwnHlu5p.d.ts.map → query-Na5gEIGd.d.ts.map} +1 -1
- package/dist/{registry-8You7OK1.d.ts → registry-xEb_xfns.d.ts} +3 -3
- package/dist/{registry-8You7OK1.d.ts.map → registry-xEb_xfns.d.ts.map} +1 -1
- package/dist/{release-confidence-nGDJiiwc.js → release-confidence-CzUHc4z4.js} +3 -3
- package/dist/{release-confidence-nGDJiiwc.js.map → release-confidence-CzUHc4z4.js.map} +1 -1
- package/dist/{release-confidence-Dqt0NFep.d.ts → release-confidence-D6lQw_o7.d.ts} +4 -4
- package/dist/{release-confidence-Dqt0NFep.d.ts.map → release-confidence-D6lQw_o7.d.ts.map} +1 -1
- package/dist/reporting.d.ts +3 -3
- package/dist/reporting.js +2 -2
- package/dist/{researcher-Cz565b7D.d.ts → researcher-CMUTQXD7.d.ts} +6 -6
- package/dist/{researcher-Cz565b7D.d.ts.map → researcher-CMUTQXD7.d.ts.map} +1 -1
- package/dist/{reward-hacking-MBf7qpSB.d.ts → reward-hacking-CgPRUesA.d.ts} +2 -2
- package/dist/{reward-hacking-MBf7qpSB.d.ts.map → reward-hacking-CgPRUesA.d.ts.map} +1 -1
- package/dist/{reward-hacking-O5zKDANP.js → reward-hacking-SkxYgT0x.js} +2 -2
- package/dist/{reward-hacking-O5zKDANP.js.map → reward-hacking-SkxYgT0x.js.map} +1 -1
- package/dist/rl.d.ts +8 -8
- package/dist/rl.d.ts.map +1 -1
- package/dist/rl.js +6 -5
- package/dist/rl.js.map +1 -1
- package/dist/rollout/index.d.ts +1 -1
- package/dist/rollout/index.js +2 -2
- package/dist/{rollout-Dm2tSdiQ.js → rollout-Crypdx8s.js} +2 -2
- package/dist/{rollout-Dm2tSdiQ.js.map → rollout-Crypdx8s.js.map} +1 -1
- package/dist/{rubric-predictive-validity-CxycqzX5.d.ts → rubric-predictive-validity-DluJLCKQ.d.ts} +2 -2
- package/dist/{rubric-predictive-validity-CxycqzX5.d.ts.map → rubric-predictive-validity-DluJLCKQ.d.ts.map} +1 -1
- package/dist/{run-record-BC0ebuRP.js → run-record-DLORoL7t.js} +2 -2
- package/dist/{run-record-BC0ebuRP.js.map → run-record-DLORoL7t.js.map} +1 -1
- package/dist/{run-record-VVy4T9OW.d.ts → run-record-DQjRcYwA.d.ts} +3 -3
- package/dist/{run-record-VVy4T9OW.d.ts.map → run-record-DQjRcYwA.d.ts.map} +1 -1
- package/dist/{schema-k6ZBftVv.js → schema-CdIX2aHu.js} +5 -1
- package/dist/{schema-k6ZBftVv.js.map → schema-CdIX2aHu.js.map} +1 -1
- package/dist/{schema-Bjgdsn73.d.ts → schema-DID1Cqct.d.ts} +7 -3
- package/dist/{schema-Bjgdsn73.d.ts.map → schema-DID1Cqct.d.ts.map} +1 -1
- package/dist/{semantic-concept-judge-BsDMOwJr.js → semantic-concept-judge-I36eejJx.js} +2 -2
- package/dist/{semantic-concept-judge-BsDMOwJr.js.map → semantic-concept-judge-I36eejJx.js.map} +1 -1
- package/dist/{sequential-BLMbdrD7.js → sequential-B51qAYE4.js} +2 -2
- package/dist/{sequential-BLMbdrD7.js.map → sequential-B51qAYE4.js.map} +1 -1
- package/dist/{server-BjYiJHoJ.js → server-CCEnywOR.js} +26 -18
- package/dist/server-CCEnywOR.js.map +1 -0
- package/dist/{skillopt-optimization-method-UArRo-nr.js → skillopt-optimization-method-DzlF2RM7.js} +9 -9
- package/dist/{skillopt-optimization-method-UArRo-nr.js.map → skillopt-optimization-method-DzlF2RM7.js.map} +1 -1
- package/dist/{statistical-heldout-Cy3EhjlC.d.ts → statistical-heldout-UhiexnjU.d.ts} +3 -3
- package/dist/{statistical-heldout-Cy3EhjlC.d.ts.map → statistical-heldout-UhiexnjU.d.ts.map} +1 -1
- package/dist/{store-B06JdC56.d.ts → store-Cq9oOrI1.d.ts} +2 -2
- package/dist/{store-B06JdC56.d.ts.map → store-Cq9oOrI1.d.ts.map} +1 -1
- package/dist/{store-otlp-C_Rq5I4D.js → store-otlp-CHjBvWQY.js} +2 -2
- package/dist/{store-otlp-C_Rq5I4D.js.map → store-otlp-CHjBvWQY.js.map} +1 -1
- package/dist/{store-tool-spans-BVga3c37.js → store-tool-spans-B9o6tU8f.js} +3 -3
- package/dist/{store-tool-spans-BVga3c37.js.map → store-tool-spans-B9o6tU8f.js.map} +1 -1
- package/dist/{store-tool-spans-DPUG7UUY.d.ts → store-tool-spans-D_qMl2__.d.ts} +102 -36
- package/dist/store-tool-spans-D_qMl2__.d.ts.map +1 -0
- package/dist/storyboard/index.d.ts +1 -1
- package/dist/{summary-report-BXeQ5Ues.js → summary-report-Bgh8CpNK.js} +2 -2
- package/dist/{summary-report-BXeQ5Ues.js.map → summary-report-Bgh8CpNK.js.map} +1 -1
- package/dist/{summary-report-CC07PhEL.d.ts → summary-report-DRstQNBX.d.ts} +3 -3
- package/dist/{summary-report-CC07PhEL.d.ts.map → summary-report-DRstQNBX.d.ts.map} +1 -1
- package/dist/{task-failure-attributes-DTl-7-Kw.js → task-failure-attributes-CBGtLS_H.js} +3 -3
- package/dist/{task-failure-attributes-DTl-7-Kw.js.map → task-failure-attributes-CBGtLS_H.js.map} +1 -1
- package/dist/{tool-groups-Ci8i9ErB.d.ts → tool-groups-RGYfVWpc.d.ts} +3 -3
- package/dist/tool-groups-RGYfVWpc.d.ts.map +1 -0
- package/dist/{tool-waste-CKc7bYIg.d.ts → tool-waste-BrmLKxMw.d.ts} +4 -4
- package/dist/{tool-waste-CKc7bYIg.d.ts.map → tool-waste-BrmLKxMw.d.ts.map} +1 -1
- package/dist/{tool-waste-8BQiUc8K.js → tool-waste-CwGHzBzX.js} +2 -2
- package/dist/{tool-waste-8BQiUc8K.js.map → tool-waste-CwGHzBzX.js.map} +1 -1
- package/dist/trace-repair/index.d.ts +3 -3
- package/dist/trace-repair/index.d.ts.map +1 -1
- package/dist/trace-repair/index.js +4 -3
- package/dist/trace-repair/index.js.map +1 -1
- package/dist/traces.d.ts +59 -61
- package/dist/traces.d.ts.map +1 -1
- package/dist/traces.js +7 -7
- package/dist/{trajectory-Bi157Gun.d.ts → trajectory-r1bQqvBQ.d.ts} +3 -3
- package/dist/{trajectory-Bi157Gun.d.ts.map → trajectory-r1bQqvBQ.d.ts.map} +1 -1
- package/dist/trajectory-replay/index.d.ts +3 -3
- package/dist/trajectory-replay/index.js +1 -1
- package/dist/{types-BPb2Kf_C.d.ts → types-CMyW4GnH.d.ts} +3 -3
- package/dist/{types-BPb2Kf_C.d.ts.map → types-CMyW4GnH.d.ts.map} +1 -1
- package/dist/{types-BI4fT3HN.js → types-CiWITkGo.js} +11 -2
- package/dist/types-CiWITkGo.js.map +1 -0
- package/dist/{types-D9ssmxKL.d.ts → types-DMoNFDWi.d.ts} +6 -3
- package/dist/{types-D9ssmxKL.d.ts.map → types-DMoNFDWi.d.ts.map} +1 -1
- package/dist/{types-D4s7Z6nq.d.ts → types-JHMOqZI4.d.ts} +13 -3
- package/dist/types-JHMOqZI4.d.ts.map +1 -0
- package/dist/{verdict-B0xltqu6.js → verdict-BQ3pCFf8.js} +2 -2
- package/dist/{verdict-B0xltqu6.js.map → verdict-BQ3pCFf8.js.map} +1 -1
- package/dist/{verdict-cache-CdVVTVmn.js → verdict-cache-B3eCVQtY.js} +2 -2
- package/dist/{verdict-cache-CdVVTVmn.js.map → verdict-cache-B3eCVQtY.js.map} +1 -1
- package/dist/wire/index.d.ts +26 -11
- package/dist/wire/index.d.ts.map +1 -1
- package/dist/wire/index.js +2 -2
- package/docs/campaign-proposers.md +4 -0
- package/docs/code-agent-intake.md +64 -0
- package/docs/concepts.md +1 -1
- package/docs/design/statistics-decisions.md +1 -1
- package/docs/distributed-driver.md +3 -6
- package/docs/public-api.md +122 -106
- package/docs/wire-protocol.md +5 -3
- package/package.json +27 -19
- package/dist/benchmark-BhT16ep9.js.map +0 -1
- package/dist/benchmark-command-CF-4GEWZ.js.map +0 -1
- package/dist/campaign-DQZmc2Dq.js.map +0 -1
- package/dist/capture-fetch-CqwsJkkG.d.ts +0 -68
- package/dist/capture-fetch-CqwsJkkG.d.ts.map +0 -1
- package/dist/contract/index.d.ts.map +0 -1
- package/dist/dspy-rlm-engine-DhA9qKIm.js.map +0 -1
- package/dist/external-optimizer-process-BFmh36vW.js.map +0 -1
- package/dist/external-optimizer-subprocess-CqLMW3nh.js.map +0 -1
- package/dist/index-CGtH1piv.d.ts.map +0 -1
- package/dist/index-D-V8gCs_.d.ts.map +0 -1
- package/dist/index-vrJugRal.d.ts +0 -1
- package/dist/kind-factory-DY8FdoXf.js.map +0 -1
- package/dist/llm-judge-Du7WQPh7.js.map +0 -1
- package/dist/pareto-BqNW3LJR.d.ts.map +0 -1
- package/dist/run-score-lDzV0X8j.js.map +0 -1
- package/dist/server-BjYiJHoJ.js.map +0 -1
- package/dist/skillopt-optimization-method-x7TTF23P.d.ts.map +0 -1
- package/dist/store-tool-spans-DPUG7UUY.d.ts.map +0 -1
- package/dist/tool-groups-Ci8i9ErB.d.ts.map +0 -1
- package/dist/transient-failure-DKF5Mofa.d.ts.map +0 -1
- package/dist/types-BI4fT3HN.js.map +0 -1
- package/dist/types-D4s7Z6nq.d.ts.map +0 -1
package/dist/contract/index.js
CHANGED
|
@@ -1,22 +1,22 @@
|
|
|
1
1
|
import { s as ValidationError } from "../errors-Dngq5h35.js";
|
|
2
|
-
import {
|
|
2
|
+
import { a as hashCanonical } from "../canonical-DPyQ_rpt.js";
|
|
3
3
|
import { r as pairedBootstrap } from "../paired-tests-C8iCsioC.js";
|
|
4
|
-
import { a as summarizeExecution, i as analyzeRuns, n as SelfImproveRunError, r as selfImprove, t as defineAgentEval } from "../define-agent-eval-
|
|
5
|
-
import { a as parseRunRecordSafe } from "../run-record-
|
|
6
|
-
import {
|
|
4
|
+
import { a as summarizeExecution, i as analyzeRuns, n as SelfImproveRunError, r as selfImprove, t as defineAgentEval } from "../define-agent-eval-Dy8QgxAI.js";
|
|
5
|
+
import { a as parseRunRecordSafe } from "../run-record-DLORoL7t.js";
|
|
6
|
+
import { O as compareOptimizationMethods, Z as defaultProductionGate, g as runImprovementLoop, i as transientDispatchFailure, it as runCampaign, mt as campaignSplitDigest, rt as runEval, t as llmJudge } from "../llm-judge-BtJ2Sfk_.js";
|
|
7
7
|
import { i as isModelPriced, n as estimateCost } from "../metrics-Qv-cpptD.js";
|
|
8
8
|
import { i as CostLedger } from "../cost-ledger-B1qx30B4.js";
|
|
9
|
-
import { d as mapConcurrentRange } from "../ledger-core-
|
|
10
|
-
import {
|
|
9
|
+
import { d as mapConcurrentRange } from "../ledger-core-PIfjCbKn.js";
|
|
10
|
+
import { C as inMemoryCampaignStorage, S as fsCampaignStorage } from "../external-optimizer-subprocess-Cex8Da2i.js";
|
|
11
11
|
import { a as heldoutSignificance, s as decidePairedPromotion } from "../power-preflight-CFXm0Vjo.js";
|
|
12
|
-
import {
|
|
13
|
-
import { G as isOtlpModelCall, W as classifyOtlpSpanRole } from "../kind-factory-
|
|
14
|
-
import { n as buildDefaultAnalystRegistry, t as createChatClient } from "../chat-client-
|
|
12
|
+
import { i as makeProposalFinding } from "../types-CiWITkGo.js";
|
|
13
|
+
import { G as isOtlpModelCall, W as classifyOtlpSpanRole } from "../kind-factory-DMeEoMQZ.js";
|
|
14
|
+
import { n as buildDefaultAnalystRegistry, t as createChatClient } from "../chat-client-DEtybj5i.js";
|
|
15
15
|
import { LLM_MODEL_ATTR_KEYS, SPAN_KIND_ATTR_KEYS } from "../trace-attributes.js";
|
|
16
16
|
import { t as extractUsage } from "../extract-usage-BrQ8mCLX.js";
|
|
17
17
|
import { n as InMemoryOutcomeStore, t as FileSystemOutcomeStore } from "../outcome-store-ChBKlTd_.js";
|
|
18
|
-
import { i as summarizeTraceErrors, n as recordAggregateMeasurements, r as summarizeExecutionMeasurements, t as readTaskFailureLabels } from "../task-failure-attributes-
|
|
19
|
-
import { c as externalTextOptimizationMethod, d as REFERENCE_EQUIVALENCE_INPUT_LIMITS, f as REFERENCE_EQUIVALENCE_JUDGE_VERSION, m as runReferenceEquivalenceJudge, n as gepaOptimizationMethod, o as heldOutGate, p as createReferenceEquivalenceJudge, s as composeGate, t as skillOptOptimizationMethod } from "../skillopt-optimization-method-
|
|
18
|
+
import { i as summarizeTraceErrors, n as recordAggregateMeasurements, r as summarizeExecutionMeasurements, t as readTaskFailureLabels } from "../task-failure-attributes-CBGtLS_H.js";
|
|
19
|
+
import { c as externalTextOptimizationMethod, d as REFERENCE_EQUIVALENCE_INPUT_LIMITS, f as REFERENCE_EQUIVALENCE_JUDGE_VERSION, m as runReferenceEquivalenceJudge, n as gepaOptimizationMethod, o as heldOutGate, p as createReferenceEquivalenceJudge, s as composeGate, t as skillOptOptimizationMethod } from "../skillopt-optimization-method-DzlF2RM7.js";
|
|
20
20
|
import { n as paretoPolicy, r as paretoSignificanceGate, t as buildEvidenceVector } from "../promotion-policy-LY9mVQ7W.js";
|
|
21
21
|
import { createHash } from "node:crypto";
|
|
22
22
|
import { agentCandidateBenchmarkSuiteSchema, agentCandidateBenchmarkTaskSchema, agentCandidateBundleSchema, agentCandidateEvaluationPolicySchema, agentCandidateExperimentSchema, agentImprovementMeasuredComparisonSchema, agentProfileImprovementExperimentSchema, agentProfileImprovementMeasuredComparisonSchema, agentProfileImprovementRunCellSchema, agentProfileImprovementRunReceiptSchema, agentProfileImprovementSuiteInputsSchema, agentProfileImprovementSuiteSchema, agentProfileImprovementTaskSchema, candidateExecutionEvidenceSchema, canonicalCandidateDigest, canonicalCandidateJson, numbersApproximatelyEqual, omitTopLevelDigest } from "@tangle-network/agent-interface";
|
|
@@ -1,6 +1,6 @@
|
|
|
1
|
-
import { s as TraceStore } from "./store-
|
|
2
|
-
import { i as TraceEmitter } from "./emitter-
|
|
3
|
-
import { n as TrajectoryStep, t as Trajectory } from "./trajectory-
|
|
1
|
+
import { s as TraceStore } from "./store-Cq9oOrI1.js";
|
|
2
|
+
import { i as TraceEmitter } from "./emitter-Bvnu0VzL.js";
|
|
3
|
+
import { n as TrajectoryStep, t as Trajectory } from "./trajectory-r1bQqvBQ.js";
|
|
4
4
|
//#region src/counterfactual.d.ts
|
|
5
5
|
type CounterfactualMutation = {
|
|
6
6
|
kind: 'swap-model';
|
|
@@ -57,4 +57,4 @@ interface CounterfactualRunner {
|
|
|
57
57
|
declare function runCounterfactual(store: TraceStore, originalRunId: string, mutation: CounterfactualMutation, runner: CounterfactualRunner): Promise<CounterfactualResult>;
|
|
58
58
|
//#endregion
|
|
59
59
|
export { runCounterfactual as a, CounterfactualRunner as i, CounterfactualMutation as n, CounterfactualResult as r, CounterfactualContext as t };
|
|
60
|
-
//# sourceMappingURL=counterfactual-
|
|
60
|
+
//# sourceMappingURL=counterfactual-Bee5_BIn.d.ts.map
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"counterfactual-
|
|
1
|
+
{"version":3,"file":"counterfactual-Bee5_BIn.d.ts","names":[],"sources":["../src/counterfactual.ts"],"mappings":";;;;KAoBY;EACN;EAAoB;EAAY;;EAChC;EAA0B;EAAY;;EACtC;EAAwB;;EACxB;EAA+B;EAAY;;EAE3C;EACA;EACA;EACA,QAAQ,MAAM,mBAAmB;;UAGtB;EACf;EACA,oBAAoB;;;EAGpB,QAAQ;EACR,UAAU;;;;EAIV,aAAa;;UAGE;EACf;EACA;EACA,UAAU;;EAEV;IACE;IACA;IACA;;;UAIa;;;;;;EAMf,cAAc,KAAK,uBAAuB,SAAS,iBAAiB;;iBAGhD,kBACpB,OAAO,YACP,uBACA,UAAU,wBACV,QAAQ,uBACP,QAAQ"}
|
|
@@ -1,8 +1,8 @@
|
|
|
1
|
-
import { i as AnalystFinding, n as AnalystContext, w as TraceAnalysisStore } from "./types-
|
|
2
|
-
import { l as ExactCapableAnalyst } from "./exact-types-
|
|
3
|
-
import { n as AnalystRegistry, r as AnalystRegistryOptions } from "./registry-
|
|
4
|
-
import { c as RawAnalystFinding, i as TraceAnalystLimits, r as TraceAnalysisEngineResult, t as TraceAnalysisEngine } from "./engine-
|
|
5
|
-
import { t as TraceToolGroupName } from "./tool-groups-
|
|
1
|
+
import { i as AnalystFinding, n as AnalystContext, w as TraceAnalysisStore } from "./types-DMoNFDWi.js";
|
|
2
|
+
import { l as ExactCapableAnalyst } from "./exact-types-BEecmnWm.js";
|
|
3
|
+
import { n as AnalystRegistry, r as AnalystRegistryOptions } from "./registry-xEb_xfns.js";
|
|
4
|
+
import { c as RawAnalystFinding, i as TraceAnalystLimits, r as TraceAnalysisEngineResult, t as TraceAnalysisEngine } from "./engine-CAmTUk52.js";
|
|
5
|
+
import { t as TraceToolGroupName } from "./tool-groups-RGYfVWpc.js";
|
|
6
6
|
//#region src/trace-analyst/behavioral-metrics.d.ts
|
|
7
7
|
type SuboptimalCode = 'monotonic-input-growth' | 'output-length-decay' | 'single-tool-dependency' | 'no-self-verification';
|
|
8
8
|
interface SuboptimalSignal {
|
|
@@ -110,4 +110,4 @@ interface DefaultAnalystRegistryOptions {
|
|
|
110
110
|
declare function buildDefaultAnalystRegistry(options?: DefaultAnalystRegistryOptions): AnalystRegistry;
|
|
111
111
|
//#endregion
|
|
112
112
|
export { createTraceAnalyst as a, BehavioralAnalystOptions as c, TraceAnalystDefinition as i, behavioralAnalyst as l, buildDefaultAnalystRegistry as n, renderPriorFindings as o, CreateTraceAnalystOptions as r, runTraceAnalyst as s, DefaultAnalystRegistryOptions as t, deriveEfficiencyFindings as u };
|
|
113
|
-
//# sourceMappingURL=default-registry-
|
|
113
|
+
//# sourceMappingURL=default-registry-B0s2zU-s.d.ts.map
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"default-registry-
|
|
1
|
+
{"version":3,"file":"default-registry-B0s2zU-s.d.ts","names":[],"sources":["../src/trace-analyst/behavioral-metrics.ts","../src/analyst/behavioral-analyst.ts","../src/analyst/kind-factory.ts","../src/analyst/default-registry.ts"],"mappings":";;;;;;KAoBY;UAMK;EACf,MAAM;EACN;;EAEA;;EAEA,UAAU;;UAGK;;EAEf;EACA;;EAEA,gBAAgB;;EAEhB;EACA;EACA,eAAe;EACf;EACA;;EAEA;EACA;EACA,SAAS;;UAGM;EACf;EACA;EACA,sBAAsB;EACtB,uBAAuB;;;;UC1BR;;EAEf;;EAEA;;;;;;iBA+Dc,yBACd,SAAS,mBACT;EAAQ;EAAoB;IAC3B;;iBAkCa,kBACd,UAAS,2BACR,oBAAoB;;;;UCrHN;EACf;EACA;EACA;EACA;;EAEA,sBAAsB,SAAS;;EAE/B;;EAEA,WAAW;;EAEX,kBACE,OAAO,oBACP,SAAS,wCACe;EAC1B,SAAS,QAAQ;EACjB,eAAe,KAAK,mBAAmB,SAAS,mBAAmB;EACnE;EACA;;UAGe;EACf,QAAQ;;EAER;EACA;;;iBAIoB,gBAAgB;EACpC,YAAY;EACZ,QAAQ;EACR,OAAO;EACP,SAAS;EACT;IACE,QAAQ;;iBAgFI,mBACd,YAAY,wBACZ,SAAS,4BACR,oBAAoB;iBA+QP,oBAAoB,OAAO;;;UCpZ1B;;;;;;EAMf,SAAS;EACT,uBAAuB;;EAEvB;EACA,aAAa;EACb,WAAW;;iBAGG,4BACd,UAAS,gCACR"}
|
|
@@ -1,10 +1,10 @@
|
|
|
1
1
|
import { c as CostLedgerHandle, f as CostLedgerSummary, m as CostReceipt } from "./cost-ledger-DbQdN3nO.js";
|
|
2
|
-
import {
|
|
3
|
-
import {
|
|
4
|
-
import { H as SurfaceProposer, R as Scenario, S as JudgeConfig, a as CampaignResult, d as DispatchContext, j as MutableSurface, k as LabeledScenarioStore, p as Gate } from "./types-
|
|
5
|
-
import {
|
|
6
|
-
import { o as
|
|
7
|
-
import { n as HostedTenant } from "./client-
|
|
2
|
+
import { _ as ProposalFinding } from "./types-DMoNFDWi.js";
|
|
3
|
+
import { B as OptimizationMethodResult, F as OptimizationMethod, Mn as CampaignCellRetryPolicy, Un as CampaignStorage, jt as SearchHistoryReceipt, kn as RunEvalOptions, l as LoopProvenanceRecord, lt as PremeasuredOptimizationBaseline, st as RunImprovementLoopResult, ut as RunOptimizationOptions, z as OptimizationMethodProvenance } from "./provenance-CafMdZKM.js";
|
|
4
|
+
import { H as SurfaceProposer, R as Scenario, S as JudgeConfig, a as CampaignResult, d as DispatchContext, j as MutableSurface, k as LabeledScenarioStore, p as Gate } from "./types-JHMOqZI4.js";
|
|
5
|
+
import { t as PowerPreflight } from "./power-preflight-Ptse_Kq7.js";
|
|
6
|
+
import { o as InsightReport } from "./insight-report-08F022xN.js";
|
|
7
|
+
import { n as HostedTenant } from "./client-CDtcZ3p9.js";
|
|
8
8
|
//#region src/contract/self-improve.d.ts
|
|
9
9
|
interface SelfImproveBudget {
|
|
10
10
|
/** Hard spend cap across the full run. Each paid call reserves its enforced
|
|
@@ -404,4 +404,4 @@ interface DefinedAgentEval<TScenario extends Scenario, TArtifact> {
|
|
|
404
404
|
declare function defineAgentEval<TScenario extends Scenario, TArtifact>(defaults: DefineAgentEvalOptions<TScenario, TArtifact>): DefinedAgentEval<TScenario, TArtifact>;
|
|
405
405
|
//#endregion
|
|
406
406
|
export { DefinedAgentEval as a, SelfImproveOptions as c, SelfImproveRunError as d, selfImprove as f, DefineAgentEvalOptions as i, SelfImproveProgressEvent as l, AgentEvalEvaluateOptions as n, defineAgentEval as o, AgentEvalImproveOptions as r, SelfImproveBudget as s, AgentEvalAgent as t, SelfImproveResult as u };
|
|
407
|
-
//# sourceMappingURL=define-agent-eval-
|
|
407
|
+
//# sourceMappingURL=define-agent-eval-Cjy2yhqP.d.ts.map
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"define-agent-eval-
|
|
1
|
+
{"version":3,"file":"define-agent-eval-Cjy2yhqP.d.ts","names":[],"sources":["../src/contract/self-improve.ts","../src/contract/define-agent-eval.ts"],"mappings":";;;;;;;;UAiEiB;;;EAGf;;;;EAIA;;EAEA;;EAEA;;;EAGA;;;EAGA;;;;EAIA;;EAEA,mBAAmB;;;;;;;;;EASnB;;EAEA;;;;;EAKA;;KAGU;EACN;EAA0B;;EAC1B;EAA4B;EAAuB;;EACnD;EAA4B;EAAe;;EAC3C;EAA8B;EAAe;EAAuB;;EAGpE;EAAsB;EAAkB;;EACxC;EAAyB;EAAW;EAAY;EAAa;;UAElD,mBAAmB,kBAAkB,UAAU;;;;;;;;;;;;;;EAc9D,QAAQ,SAAS,gBAAgB,UAAU,WAAW,KAAK,oBAAoB,QAAQ;;;;;;;EAQvF;;;EAIA,WAAW;;EAGX,OAAO,YAAY,WAAW;;;EAI9B,iBAAiB;;EAGjB,SAAS;;;;;;;;;;;;EAaT,sBAAsB,gCAAgC,WAAW;;;;;EAMjE,WAAW,gBAAgB;;;;;;EAO3B,SAAS,mBAAmB,WAAW;;;EAIvC,qBAAqB;;;EAIrB,OAAO,KAAK,WAAW;;;;;;;;EASvB,cAAc,eAAe,gBAAgB,iBAAiB,mBAAmB;;;;EAKjF,UAAU;;;;EAKV;;;EAIA,gBAAgB,QAAQ;;;;;EAMxB,iBAAiB;IACf,UAAU;IACV;IACA;;;;EAKF;;;;;;;EAQA,YAAY;;;EAIZ,cAAc,OAAO;;;;EAKrB;EACA;EACA;;;;;;;;;;;;;;EAeA,eAAe;;;EAIf,eAAe;;;;EAKf,eAAe;;EAGf;;;;;;;;EASA;;;;;;;;;EAUA,oBAAoB,uBAAuB,WAAW;;;EAItD,WAAW;;;;;;;EAQX,mBAAmB,uBAAuB,WAAW;;;;;;EAOrD,eAAe,uBAAuB,WAAW;;;;EAKjD,eAAe,uBAAuB,WAAW;;UAGlC,kBAAkB,kBAAkB,UAAU;;;;EAI7D;IACE;IACA,aAAa;;;;;EAKf;IACE;IACA,aAAa;IACb,SAAS;;;IAGT;;;;IAIA;;;;;;EAMF;;;EAGA;;;;EAIA,YAAY;;EAEZ;;;EAGA;;EAEA;;EAEA;;EAEA,MAAM;;;EAGN,UAAU;;;EAGV,gBAAgB;;EAEhB;IACE;IACA,MAAM;IACN;IACA,aAAa;;;;;;;;EAQf,SAAS;;;;;EAKT,QAAQ;;;;;;EAMR,KAAK,yBAAyB,WAAW;;;cAI9B,4BAA4B;WAC9B,MAAM;WACN,UAAU;EAEnB,YAAY,gBAAgB,QAAQ;;;;;;;;;;;;;;;;;;;;;;;;;;;;iBAuMhB,YAAY,kBAAkB,UAAU,WAC5D,MAAM,mBAAmB,WAAW,aACnC,QAAQ,kBAAkB,WAAW;;;KCvkB5B,eAAe,kBAAkB,UAAU,cACrD,SAAS,gBACT,UAAU,WACV,KAAK,oBACF,QAAQ;KAED,uBAAuB,kBAAkB,UAAU,aAAa,mBAC1E,WACA;UAGe,yBAAyB,kBAAkB,UAAU,mBAC5D,KACN,eAAe,WAAW;;EAI5B,YAAY;;EAEZ,UAAU;;EAEV,QAAQ,eAAe,WAAW;;EAElC,QAAQ,YAAY,WAAW;;EAE/B,SAAS,YAAY,WAAW;;EAEhC;;KAGU,wBAAwB,kBAAkB,UAAU,aAAa,KAC3E,QAAQ,mBAAmB,WAAW;EAGtC,SAAS,QAAQ;EACjB,eAAe,QAAQ;;UAGR,iBAAiB,kBAAkB,UAAU;;WAEnD,oBAAoB;;WAEpB,iBAAiB;;;;;EAK1B,SACE,OAAO,yBAAyB,WAAW,aAC1C,QAAQ,eAAe,WAAW;;;;;;;EAOrC,QACE,OAAO,wBAAwB,WAAW,aACzC,QAAQ,kBAAkB,WAAW;;;;;;;;;iBAU1B,gBAAgB,kBAAkB,UAAU,WAC1D,UAAU,uBAAuB,WAAW,aAC3C,iBAAiB,WAAW"}
|
|
@@ -6,15 +6,15 @@ import { i as requiredPairedSampleSize, r as pairedMde } from "./power-and-mde-B
|
|
|
6
6
|
import { r as pairRunRecords } from "./paired-arms-D4aeIHUy.js";
|
|
7
7
|
import { o as pairedTTest, r as pairedBootstrap } from "./paired-tests-C8iCsioC.js";
|
|
8
8
|
import { r as welchsTTest } from "./baseline-BC-eBZ7U.js";
|
|
9
|
-
import "./query-
|
|
9
|
+
import "./query-BPGMVlbM.js";
|
|
10
10
|
import { r as observedSplitScore } from "./reward-nw2xZGZG.js";
|
|
11
|
-
import { c as validateRunRecord, i as modelHasSnapshot } from "./run-record-
|
|
12
|
-
import { r as paretoChart } from "./summary-report-
|
|
13
|
-
import {
|
|
14
|
-
import { c as campaignCellTaskScore, l as campaignCellToRunRecord, o as campaignCellExecutionEvidence, s as campaignCellJudgeDimensions } from "./reward-hacking-
|
|
15
|
-
import {
|
|
11
|
+
import { c as validateRunRecord, i as modelHasSnapshot } from "./run-record-DLORoL7t.js";
|
|
12
|
+
import { r as paretoChart } from "./summary-report-Bgh8CpNK.js";
|
|
13
|
+
import { E as assertOptimizationResult, F as surfaceContentHash, I as surfaceHash, Z as defaultProductionGate, d as loopProvenanceArgsFromResult, g as runImprovementLoop, ot as resolveRunDir, rt as runEval, u as emitLoopProvenance } from "./llm-judge-BtJ2Sfk_.js";
|
|
14
|
+
import { c as campaignCellTaskScore, l as campaignCellToRunRecord, o as campaignCellExecutionEvidence, s as campaignCellJudgeDimensions } from "./reward-hacking-SkxYgT0x.js";
|
|
15
|
+
import { C as inMemoryCampaignStorage, S as fsCampaignStorage, x as createRunCostLedger } from "./external-optimizer-subprocess-Cex8Da2i.js";
|
|
16
16
|
import { t as powerPreflight } from "./power-preflight-CFXm0Vjo.js";
|
|
17
|
-
import { t as createHostedClient } from "./client-
|
|
17
|
+
import { t as createHostedClient } from "./client-BvwNkIRN.js";
|
|
18
18
|
//#region src/contamination-guard.ts
|
|
19
19
|
function checkCanaries(output, scenarios) {
|
|
20
20
|
const leaks = [];
|
|
@@ -1023,6 +1023,13 @@ function splitMethodPartitions(searchScenarios, explicitSelection, fraction) {
|
|
|
1023
1023
|
function safeRunComponent(value) {
|
|
1024
1024
|
return value.replace(/[^a-zA-Z0-9._-]/g, "_");
|
|
1025
1025
|
}
|
|
1026
|
+
/** 32-bit FNV-1a over raw UTF-16 code units, read as an unsigned int and used
|
|
1027
|
+
* only to order scenarios deterministically.
|
|
1028
|
+
*
|
|
1029
|
+
* Frozen: the order it produces assigns scenarios to partitions, so a change
|
|
1030
|
+
* re-partitions decisions already recorded. It masks no byte, unlike
|
|
1031
|
+
* `fnv1a32` in `src/partition-held-out.ts`, so the two disagree on any
|
|
1032
|
+
* non-ASCII id and cannot be shared. */
|
|
1026
1033
|
function stableScenarioHash(value) {
|
|
1027
1034
|
let hash = 2166136261;
|
|
1028
1035
|
for (let index = 0; index < value.length; index++) {
|
|
@@ -1364,6 +1371,11 @@ function averageComposite(campaign) {
|
|
|
1364
1371
|
const aggs = Object.values(campaign.aggregates.byScenario);
|
|
1365
1372
|
return aggs.length === 0 ? 0 : aggs.reduce((s, a) => s + a.meanComposite, 0) / aggs.length;
|
|
1366
1373
|
}
|
|
1374
|
+
/** 32-bit FNV-1a over raw UTF-16 code units, rendered as hex for a cell key.
|
|
1375
|
+
*
|
|
1376
|
+
* Frozen: the key names a persisted cell, so a change orphans every cell
|
|
1377
|
+
* already written. Same loop as `stableScenarioHash` above but a different
|
|
1378
|
+
* return form; neither is a general-purpose hash. */
|
|
1367
1379
|
function hashString(s) {
|
|
1368
1380
|
let h = 2166136261;
|
|
1369
1381
|
for (let i = 0; i < s.length; i++) {
|
|
@@ -1497,4 +1509,4 @@ function requirePositiveInteger(value, field) {
|
|
|
1497
1509
|
//#endregion
|
|
1498
1510
|
export { summarizeExecution as a, analyzeRuns as i, SelfImproveRunError as n, checkCanaries as o, selfImprove as r, defineAgentEval as t };
|
|
1499
1511
|
|
|
1500
|
-
//# sourceMappingURL=define-agent-eval-
|
|
1512
|
+
//# sourceMappingURL=define-agent-eval-Dy8QgxAI.js.map
|