@tangle-network/agent-runtime 0.128.0 → 0.131.0
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/README.md +70 -20
- package/dist/{activation-DhWJ3p8N.js → activation-BCzMOTaV.js} +3 -3
- package/dist/{activation-DhWJ3p8N.js.map → activation-BCzMOTaV.js.map} +1 -1
- package/dist/agent.d.ts +2 -3
- package/dist/agent.js +4 -5
- package/dist/agent.js.map +1 -1
- package/dist/{analyst-loop-DvSciOfB.js → analyst-loop-BE8cDs5Q.js} +2 -2
- package/dist/{analyst-loop-DvSciOfB.js.map → analyst-loop-BE8cDs5Q.js.map} +1 -1
- package/dist/analyst-loop.d.ts +1 -1
- package/dist/analyst-loop.js +1 -1
- package/dist/authoring-CvHwo1oW.js +163 -0
- package/dist/authoring-CvHwo1oW.js.map +1 -0
- package/dist/candidate-execution/index.js +4 -4
- package/dist/{candidate-execution-BFpq-Xi6.js → candidate-execution-BNxKr-Bu.js} +5 -5
- package/dist/{candidate-execution-BFpq-Xi6.js.map → candidate-execution-BNxKr-Bu.js.map} +1 -1
- package/dist/{conversation-BpLQZGPH.js → conversation-DNtxaJ1Z.js} +113 -31
- package/dist/conversation-DNtxaJ1Z.js.map +1 -0
- package/dist/conversation.d.ts +2 -2
- package/dist/conversation.js +2 -2
- package/dist/environment-provider-CxvSd1W6.d.ts +86 -0
- package/dist/{environment-provider-DqFS6FSZ.js → environment-provider-Dyg8DtLK.js} +8 -4
- package/dist/environment-provider-Dyg8DtLK.js.map +1 -0
- package/dist/environment-provider.d.ts +1 -1
- package/dist/environment-provider.js +1 -1
- package/dist/graph-BJTxGOFB.js +471 -0
- package/dist/graph-BJTxGOFB.js.map +1 -0
- package/dist/{improvement-cycle-IJgCbKWQ.js → improvement-cycle-Csp38cWg.js} +133 -224
- package/dist/improvement-cycle-Csp38cWg.js.map +1 -0
- package/dist/{index-BhZhQw77.d.ts → index-CoO7atyo.d.ts} +556 -1278
- package/dist/{index-BhuzfG2r.d.ts → index-DwGtu9nc.d.ts} +7 -9
- package/dist/{index-Efjb3nrQ.d.ts → index-qYHpsmG2.d.ts} +36 -18
- package/dist/index.d.ts +353 -11
- package/dist/index.js +111 -354
- package/dist/index.js.map +1 -1
- package/dist/intelligence.d.ts +6 -6
- package/dist/intelligence.js +9 -8
- package/dist/intelligence.js.map +1 -1
- package/dist/kernel.d.ts +7 -5
- package/dist/kernel.js +13 -9
- package/dist/{knowledge-DF63xPr4.js → knowledge-ce0_uKCl.js} +19 -17
- package/dist/knowledge-ce0_uKCl.js.map +1 -0
- package/dist/knowledge.d.ts +1 -1
- package/dist/knowledge.js +1 -1
- package/dist/{loop-runner-bin-Ckp_9tmD.d.ts → loop-runner-bin-BwgZ8m_7.d.ts} +6 -3
- package/dist/{loop-runner-bin-CWqOpCEw.js → loop-runner-bin-DSbuDDqM.js} +5 -27
- package/dist/loop-runner-bin-DSbuDDqM.js.map +1 -0
- package/dist/loop-runner-bin.d.ts +1 -1
- package/dist/loop-runner-bin.js +1 -1
- package/dist/materialization-COJ1UYQ-.js +272 -0
- package/dist/materialization-COJ1UYQ-.js.map +1 -0
- package/dist/mcp/bin.js +39 -47
- package/dist/mcp/bin.js.map +1 -1
- package/dist/mcp/index.d.ts +24 -26
- package/dist/mcp/index.js +66 -83
- package/dist/mcp/index.js.map +1 -1
- package/dist/mcp/memory-bin.js +1 -1
- package/dist/{memory-server-DL6cE2Ag.js → memory-server-5HEJH672.js} +2 -2
- package/dist/{memory-server-DL6cE2Ag.js.map → memory-server-5HEJH672.js.map} +1 -1
- package/dist/model-policy-CqziaqS1.js +232 -0
- package/dist/model-policy-CqziaqS1.js.map +1 -0
- package/dist/{openai-tools-B68JaOCx.d.ts → openai-tools-D3oMyVGl.d.ts} +2 -2
- package/dist/{openai-tools-D3XfrrQ6.js → openai-tools-ru75mLjq.js} +2 -2
- package/dist/openai-tools-ru75mLjq.js.map +1 -0
- package/dist/{prepare--8EvLqCr.js → prepare-DYWjVcPx.js} +169 -153
- package/dist/prepare-DYWjVcPx.js.map +1 -0
- package/dist/primeintellect/index.d.ts +7 -6
- package/dist/primeintellect/index.js +9 -11
- package/dist/primeintellect/index.js.map +1 -1
- package/dist/profiles.d.ts +21 -174
- package/dist/profiles.js +67 -276
- package/dist/profiles.js.map +1 -1
- package/dist/{protected-model-port-CXVfOUu_.js → protected-model-port-48ALLGxT.js} +2 -2
- package/dist/{protected-model-port-CXVfOUu_.js.map → protected-model-port-48ALLGxT.js.map} +1 -1
- package/dist/{redact-PQzmE1Jn.d.ts → redact-9_Gf-8m3.d.ts} +58 -69
- package/dist/{researcher-CoVqNhfI.js → researcher-Skz5-Uc8.js} +50 -26
- package/dist/researcher-Skz5-Uc8.js.map +1 -0
- package/dist/{run-layout-C2FGmZ3v.js → run-layout-CeJsEAom.js} +2 -2
- package/dist/{run-layout-C2FGmZ3v.js.map → run-layout-CeJsEAom.js.map} +1 -1
- package/dist/runtime-D-QfLbSd.d.ts +893 -0
- package/dist/{runtime-5uDVVfER.js → runtime-hiAABiTk.js} +315 -1191
- package/dist/runtime-hiAABiTk.js.map +1 -0
- package/dist/{sandbox-events-Yhd1GYWl.js → sandbox-events-CRDwc5WN.js} +42 -5
- package/dist/sandbox-events-CRDwc5WN.js.map +1 -0
- package/dist/snapshot-CXiiuHhL.js +21 -0
- package/dist/snapshot-CXiiuHhL.js.map +1 -0
- package/dist/{spawn-journal-DsZKDqeh.js → spawn-journal-saHQzqYi.js} +15 -21
- package/dist/spawn-journal-saHQzqYi.js.map +1 -0
- package/dist/stream-agent-turn-C852AgMT.d.ts +128 -0
- package/dist/stream-agent-turn-rYgaOLO0.js +910 -0
- package/dist/stream-agent-turn-rYgaOLO0.js.map +1 -0
- package/dist/{structural-rollout-zY0oqQzO.js → structural-rollout-3uxVGcg2.js} +459 -235
- package/dist/structural-rollout-3uxVGcg2.js.map +1 -0
- package/dist/{supervise-CsTKbH9R.js → supervise-iPN27pO0.js} +864 -4785
- package/dist/supervise-iPN27pO0.js.map +1 -0
- package/dist/{supervisor-DpjO0Gmy.js → supervisor-CV6Jh28D.js} +7439 -2897
- package/dist/supervisor-CV6Jh28D.js.map +1 -0
- package/dist/testing.d.ts +3 -1
- package/dist/testing.js +271 -221
- package/dist/testing.js.map +1 -1
- package/dist/{top-app-5unxqovu.js → top-app-G4M18b6i.js} +3 -3
- package/dist/{top-app-5unxqovu.js.map → top-app-G4M18b6i.js.map} +1 -1
- package/dist/tui/bin.js +1 -1
- package/dist/tui/index.js +1 -1
- package/dist/{environment-provider-CUFsyymu.d.ts → types-C6Q-J0Dt.d.ts} +51 -114
- package/dist/{types-DnNGJ5Gz.d.ts → types-ebIY0dMG.d.ts} +556 -30
- package/dist/{util-MVgdwuIS.js → util-Bw6srryQ.js} +3 -2
- package/dist/{util-MVgdwuIS.js.map → util-Bw6srryQ.js.map} +1 -1
- package/dist/{workspace-archive-BQxvkypI.js → workspace-archive-aOfJ47ms.js} +3 -3
- package/dist/{workspace-archive-BQxvkypI.js.map → workspace-archive-aOfJ47ms.js.map} +1 -1
- package/package.json +12 -15
- package/skills/agent-graphs/IMPROVE.md +3 -3
- package/skills/agent-graphs/SKILL.md +4 -5
- package/skills/agent-graphs/cases/review-pipeline.json +1 -2
- package/skills/agent-graphs/cases/unmeasured-harness.json +2 -4
- package/dist/backends-CiOCyRHb.js +0 -743
- package/dist/backends-CiOCyRHb.js.map +0 -1
- package/dist/conversation-BpLQZGPH.js.map +0 -1
- package/dist/environment-provider-DqFS6FSZ.js.map +0 -1
- package/dist/improvement-cycle-IJgCbKWQ.js.map +0 -1
- package/dist/index-DLM0W1h1.d.ts +0 -545
- package/dist/knowledge-DF63xPr4.js.map +0 -1
- package/dist/local-harness-BIajef4A.d.ts +0 -465
- package/dist/loop-runner-bin-CWqOpCEw.js.map +0 -1
- package/dist/model-resolution-Btd9iIKV.js +0 -98
- package/dist/model-resolution-Btd9iIKV.js.map +0 -1
- package/dist/openai-tools-D3XfrrQ6.js.map +0 -1
- package/dist/prepare--8EvLqCr.js.map +0 -1
- package/dist/researcher-CoVqNhfI.js.map +0 -1
- package/dist/runtime-5uDVVfER.js.map +0 -1
- package/dist/sandbox-events-Yhd1GYWl.js.map +0 -1
- package/dist/spawn-journal-DsZKDqeh.js.map +0 -1
- package/dist/structural-rollout-zY0oqQzO.js.map +0 -1
- package/dist/supervise-CsTKbH9R.js.map +0 -1
- package/dist/supervisor-DpjO0Gmy.js.map +0 -1
- package/dist/types-C9j4qg6l.d.ts +0 -500
- package/skills/agent-graphs/cases/floor-trap-pi.json +0 -11
package/dist/testing.js
CHANGED
|
@@ -1,10 +1,12 @@
|
|
|
1
|
-
import { $ as parseExactAgentProfile, K as applyExactAgentProfileDiff, rt as canonicalCandidateDigest$1, st as immutableCandidateValue } from "./prepare
|
|
2
|
-
import { f as verifyAgentImprovementProposal } from "./improvement-cycle-
|
|
1
|
+
import { $ as parseExactAgentProfile, K as applyExactAgentProfileDiff, rt as canonicalCandidateDigest$1, st as immutableCandidateValue } from "./prepare-DYWjVcPx.js";
|
|
2
|
+
import { f as verifyAgentImprovementProposal } from "./improvement-cycle-Csp38cWg.js";
|
|
3
|
+
import { c as supervisorAgentWithTestBrain, dt as driverAgent, r as superviseWithTestBrain } from "./supervise-iPN27pO0.js";
|
|
4
|
+
import { i as runGraphWithTestBrain } from "./graph-BJTxGOFB.js";
|
|
3
5
|
import { SANDBOX_SIZE_PRESET_NAMES } from "@tangle-network/agent-interface";
|
|
4
6
|
//#region src/testing/fixtures/agent-improvement-proposal.json
|
|
5
7
|
var agent_improvement_proposal_default = {
|
|
6
8
|
changedSurfaces: ["prompt"],
|
|
7
|
-
digest: "sha256:
|
|
9
|
+
digest: "sha256:01a9d56883f0e99ee43e54a1ec921ee80542e1b67b78659ca6335aee267609d5",
|
|
8
10
|
evaluation: {
|
|
9
11
|
"decision": {
|
|
10
12
|
"contributingChecks": [
|
|
@@ -613,7 +615,7 @@ var agent_improvement_proposal_default = {
|
|
|
613
615
|
}
|
|
614
616
|
},
|
|
615
617
|
"candidate": {
|
|
616
|
-
"digest": "sha256:
|
|
618
|
+
"digest": "sha256:b357b2445d3c6c1ab2b859d66b2ced4c7e5770988c299d34f07b7e536c7a0784",
|
|
617
619
|
"kind": "agent-candidate-execution-evidence",
|
|
618
620
|
"materializationReceipt": {
|
|
619
621
|
"benchmark": {
|
|
@@ -648,16 +650,16 @@ var agent_improvement_proposal_default = {
|
|
|
648
650
|
},
|
|
649
651
|
"source": "evaluator-task-container"
|
|
650
652
|
},
|
|
651
|
-
"digest": "sha256:
|
|
653
|
+
"digest": "sha256:4ca0f8def5548e40000d714ec0d6dee07ead2d819113a787fdc7d7906f9644a5",
|
|
652
654
|
"digestAlgorithm": "rfc8785-sha256",
|
|
653
655
|
"executionPlan": {
|
|
654
656
|
"artifact": {
|
|
655
|
-
"byteLength":
|
|
656
|
-
"content": "
|
|
657
|
+
"byteLength": 2164,
|
|
658
|
+
"content": "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",
|
|
657
659
|
"encoding": "base64",
|
|
658
|
-
"sha256": "sha256:
|
|
660
|
+
"sha256": "sha256:ab47d989ef65673c99766817a2c193a835edc084828331c4dc6880895e57f78c"
|
|
659
661
|
},
|
|
660
|
-
"digest": "sha256:
|
|
662
|
+
"digest": "sha256:ab47d989ef65673c99766817a2c193a835edc084828331c4dc6880895e57f78c",
|
|
661
663
|
"kind": "agent-candidate-execution-plan",
|
|
662
664
|
"material": {
|
|
663
665
|
"codeKind": "disabled",
|
|
@@ -682,7 +684,7 @@ var agent_improvement_proposal_default = {
|
|
|
682
684
|
"value": "-c"
|
|
683
685
|
}, {
|
|
684
686
|
"kind": "public",
|
|
685
|
-
"value": "
|
|
687
|
+
"value": "model_instructions_file=.codex/system-prompt.md"
|
|
686
688
|
}],
|
|
687
689
|
"cwd": {
|
|
688
690
|
"path": ".",
|
|
@@ -727,8 +729,8 @@ var agent_improvement_proposal_default = {
|
|
|
727
729
|
},
|
|
728
730
|
"network": { "mode": "disabled" },
|
|
729
731
|
"profile": {
|
|
730
|
-
"mountPaths": ["AGENTS.md"],
|
|
731
|
-
"planDigest": "sha256:
|
|
732
|
+
"mountPaths": [".codex/system-prompt.md", "AGENTS.md"],
|
|
733
|
+
"planDigest": "sha256:9452cb632b127aae1dc94fa8396f7792f14247a8dd24bd97f6aec998e10e2626",
|
|
732
734
|
"targetWorkspace": "task"
|
|
733
735
|
},
|
|
734
736
|
"runCell": {
|
|
@@ -751,8 +753,12 @@ var agent_improvement_proposal_default = {
|
|
|
751
753
|
"harnessVersion": "1.2.3",
|
|
752
754
|
"kind": "agent-candidate-materialization",
|
|
753
755
|
"profileActivation": {
|
|
754
|
-
"digest": "sha256:
|
|
756
|
+
"digest": "sha256:41c3e5ee69665488a2aefdf82b746696422bfbcd528fdedc110bc8e267cb0284",
|
|
755
757
|
"files": [{
|
|
758
|
+
"content": "Return the exact measured answer.",
|
|
759
|
+
"mode": 420,
|
|
760
|
+
"path": ".codex/system-prompt.md"
|
|
761
|
+
}, {
|
|
756
762
|
"content": "Inspect the repository, implement the fix, and run tests.\n",
|
|
757
763
|
"mode": 420,
|
|
758
764
|
"path": "AGENTS.md"
|
|
@@ -760,16 +766,20 @@ var agent_improvement_proposal_default = {
|
|
|
760
766
|
"kind": "agent-candidate-profile-activation",
|
|
761
767
|
"profilePlan": {
|
|
762
768
|
"artifact": {
|
|
763
|
-
"byteLength":
|
|
764
|
-
"content": "
|
|
769
|
+
"byteLength": 532,
|
|
770
|
+
"content": "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",
|
|
765
771
|
"encoding": "base64",
|
|
766
|
-
"sha256": "sha256:
|
|
772
|
+
"sha256": "sha256:9452cb632b127aae1dc94fa8396f7792f14247a8dd24bd97f6aec998e10e2626"
|
|
767
773
|
},
|
|
768
|
-
"digest": "sha256:
|
|
774
|
+
"digest": "sha256:9452cb632b127aae1dc94fa8396f7792f14247a8dd24bd97f6aec998e10e2626",
|
|
769
775
|
"kind": "agent-profile-workspace-plan",
|
|
770
776
|
"material": {
|
|
771
777
|
"env": {},
|
|
772
778
|
"files": [{
|
|
779
|
+
"contentSha256": "sha256:7cd9c117d3377708275faa6a564af952f3f0c71352ae55833bf416377962d5b0",
|
|
780
|
+
"mode": 420,
|
|
781
|
+
"relPath": ".codex/system-prompt.md"
|
|
782
|
+
}, {
|
|
773
783
|
"contentSha256": "sha256:39adbb9b7f571f0fd586e7795023d25909adbe9586bb21ee134d3d7248bf1599",
|
|
774
784
|
"mode": 420,
|
|
775
785
|
"relPath": "AGENTS.md"
|
|
@@ -779,7 +789,7 @@ var agent_improvement_proposal_default = {
|
|
|
779
789
|
"value": "-c"
|
|
780
790
|
}, {
|
|
781
791
|
"kind": "public",
|
|
782
|
-
"value": "
|
|
792
|
+
"value": "model_instructions_file=.codex/system-prompt.md"
|
|
783
793
|
}],
|
|
784
794
|
"harness": "codex",
|
|
785
795
|
"sourceProfileDigest": "sha256:8b1e2f0f7b832958381c29baebe4a961b151af97a0ce9d611a15ad0741aeaa75",
|
|
@@ -801,12 +811,12 @@ var agent_improvement_proposal_default = {
|
|
|
801
811
|
"byteLength": 1149,
|
|
802
812
|
"locator": {
|
|
803
813
|
"bucket": "candidate-test-artifacts",
|
|
804
|
-
"key": "benchmark-result/
|
|
814
|
+
"key": "benchmark-result/068e40f78a8eb33666a28ee28a08b4bdfdc9f1caae1ecfbe827ab8d4f8fe3126",
|
|
805
815
|
"kind": "s3"
|
|
806
816
|
},
|
|
807
|
-
"sha256": "sha256:
|
|
817
|
+
"sha256": "sha256:068e40f78a8eb33666a28ee28a08b4bdfdc9f1caae1ecfbe827ab8d4f8fe3126"
|
|
808
818
|
},
|
|
809
|
-
"digest": "sha256:
|
|
819
|
+
"digest": "sha256:068e40f78a8eb33666a28ee28a08b4bdfdc9f1caae1ecfbe827ab8d4f8fe3126",
|
|
810
820
|
"kind": "agent-candidate-benchmark-result",
|
|
811
821
|
"material": {
|
|
812
822
|
"dimensions": [{
|
|
@@ -822,7 +832,7 @@ var agent_improvement_proposal_default = {
|
|
|
822
832
|
},
|
|
823
833
|
"sha256": "sha256:9b9b3a1471309177261cfe65ea9c298e0dd372e4b5d087f8d35d7b732485373d"
|
|
824
834
|
},
|
|
825
|
-
"executionPlanDigest": "sha256:
|
|
835
|
+
"executionPlanDigest": "sha256:ab47d989ef65673c99766817a2c193a835edc084828331c4dc6880895e57f78c",
|
|
826
836
|
"grader": {
|
|
827
837
|
"artifact": {
|
|
828
838
|
"byteLength": 25,
|
|
@@ -856,41 +866,41 @@ var agent_improvement_proposal_default = {
|
|
|
856
866
|
"kind": "agent-candidate-benchmark-result-material",
|
|
857
867
|
"passed": true,
|
|
858
868
|
"score": 1,
|
|
859
|
-
"taskOutcomeDigest": "sha256:
|
|
869
|
+
"taskOutcomeDigest": "sha256:3dd0532c46287c348018bf2a03d9e213bb2ced4adfc9b44021b84fa5313d623e"
|
|
860
870
|
}
|
|
861
871
|
},
|
|
862
872
|
"bundleDigest": "sha256:60fcbb1c728194bd51d7d19cb732d1c3f1881dce7e0a6266b41c8b98cfd65693",
|
|
863
|
-
"digest": "sha256:
|
|
873
|
+
"digest": "sha256:b944568cbf59671d4ead0fe2a66f0c420da4bb815f11da79c00442daeede5361",
|
|
864
874
|
"digestAlgorithm": "rfc8785-sha256",
|
|
865
|
-
"executionPlanDigest": "sha256:
|
|
875
|
+
"executionPlanDigest": "sha256:ab47d989ef65673c99766817a2c193a835edc084828331c4dc6880895e57f78c",
|
|
866
876
|
"executorCapture": {
|
|
867
877
|
"byteLength": 477,
|
|
868
878
|
"locator": {
|
|
869
879
|
"bucket": "candidate-test-artifacts",
|
|
870
|
-
"key": "executor-capture/
|
|
880
|
+
"key": "executor-capture/b6fad7b0238fd1192f347185d43ca28fb82159113f36d028f3d3a3abebf575a8",
|
|
871
881
|
"kind": "s3"
|
|
872
882
|
},
|
|
873
|
-
"sha256": "sha256:
|
|
883
|
+
"sha256": "sha256:b6fad7b0238fd1192f347185d43ca28fb82159113f36d028f3d3a3abebf575a8"
|
|
874
884
|
},
|
|
875
885
|
"kind": "agent-candidate-run",
|
|
876
|
-
"materializationReceiptDigest": "sha256:
|
|
886
|
+
"materializationReceiptDigest": "sha256:4ca0f8def5548e40000d714ec0d6dee07ead2d819113a787fdc7d7906f9644a5",
|
|
877
887
|
"memory": { "mode": "disabled" },
|
|
878
888
|
"modelSettlement": {
|
|
879
889
|
"artifact": {
|
|
880
890
|
"byteLength": 643,
|
|
881
891
|
"locator": {
|
|
882
892
|
"bucket": "candidate-test-artifacts",
|
|
883
|
-
"key": "model-settlement/
|
|
893
|
+
"key": "model-settlement/26102d74d799e7c2138ddfb60591dc2fbe037aecfb49ad6c1f4a69c456a5c4ce",
|
|
884
894
|
"kind": "s3"
|
|
885
895
|
},
|
|
886
|
-
"sha256": "sha256:
|
|
896
|
+
"sha256": "sha256:26102d74d799e7c2138ddfb60591dc2fbe037aecfb49ad6c1f4a69c456a5c4ce"
|
|
887
897
|
},
|
|
888
|
-
"digest": "sha256:
|
|
898
|
+
"digest": "sha256:26102d74d799e7c2138ddfb60591dc2fbe037aecfb49ad6c1f4a69c456a5c4ce",
|
|
889
899
|
"kind": "agent-candidate-model-settlement",
|
|
890
900
|
"material": {
|
|
891
901
|
"calls": [],
|
|
892
902
|
"closed": true,
|
|
893
|
-
"executionPlanDigest": "sha256:
|
|
903
|
+
"executionPlanDigest": "sha256:ab47d989ef65673c99766817a2c193a835edc084828331c4dc6880895e57f78c",
|
|
894
904
|
"grantDigest": "sha256:cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc",
|
|
895
905
|
"kind": "agent-candidate-model-settlement-material",
|
|
896
906
|
"preparationId": "candidate-preparation.nBqsDbKYlsiY4Tto5XaxkR3RCTop9PA9_kX-kOSzR8U",
|
|
@@ -919,15 +929,15 @@ var agent_improvement_proposal_default = {
|
|
|
919
929
|
"byteLength": 478,
|
|
920
930
|
"locator": {
|
|
921
931
|
"bucket": "candidate-test-artifacts",
|
|
922
|
-
"key": "task-outcome/
|
|
932
|
+
"key": "task-outcome/3dd0532c46287c348018bf2a03d9e213bb2ced4adfc9b44021b84fa5313d623e",
|
|
923
933
|
"kind": "s3"
|
|
924
934
|
},
|
|
925
|
-
"sha256": "sha256:
|
|
935
|
+
"sha256": "sha256:3dd0532c46287c348018bf2a03d9e213bb2ced4adfc9b44021b84fa5313d623e"
|
|
926
936
|
},
|
|
927
|
-
"digest": "sha256:
|
|
937
|
+
"digest": "sha256:3dd0532c46287c348018bf2a03d9e213bb2ced4adfc9b44021b84fa5313d623e",
|
|
928
938
|
"kind": "agent-candidate-task-outcome",
|
|
929
939
|
"material": {
|
|
930
|
-
"executionPlanDigest": "sha256:
|
|
940
|
+
"executionPlanDigest": "sha256:ab47d989ef65673c99766817a2c193a835edc084828331c4dc6880895e57f78c",
|
|
931
941
|
"kind": "agent-candidate-task-outcome-material",
|
|
932
942
|
"outcome": {
|
|
933
943
|
"artifact": {
|
|
@@ -961,10 +971,10 @@ var agent_improvement_proposal_default = {
|
|
|
961
971
|
"byteLength": 750,
|
|
962
972
|
"locator": {
|
|
963
973
|
"bucket": "candidate-test-artifacts",
|
|
964
|
-
"key": "trace/
|
|
974
|
+
"key": "trace/4b8b5948ab6ef033359ae0b32fd1b55ff67c5fd6d174bc0943c644bb148e1c29",
|
|
965
975
|
"kind": "s3"
|
|
966
976
|
},
|
|
967
|
-
"sha256": "sha256:
|
|
977
|
+
"sha256": "sha256:4b8b5948ab6ef033359ae0b32fd1b55ff67c5fd6d174bc0943c644bb148e1c29"
|
|
968
978
|
},
|
|
969
979
|
"eventCount": 1,
|
|
970
980
|
"modelCallCount": 0
|
|
@@ -1321,7 +1331,7 @@ var agent_improvement_proposal_default = {
|
|
|
1321
1331
|
}
|
|
1322
1332
|
},
|
|
1323
1333
|
"candidate": {
|
|
1324
|
-
"digest": "sha256:
|
|
1334
|
+
"digest": "sha256:4a3e7a7b18f2601294c4f0a786e5ce924feae0a69e16edd8d785ea528c990e10",
|
|
1325
1335
|
"kind": "agent-candidate-execution-evidence",
|
|
1326
1336
|
"materializationReceipt": {
|
|
1327
1337
|
"benchmark": {
|
|
@@ -1356,16 +1366,16 @@ var agent_improvement_proposal_default = {
|
|
|
1356
1366
|
},
|
|
1357
1367
|
"source": "evaluator-task-container"
|
|
1358
1368
|
},
|
|
1359
|
-
"digest": "sha256:
|
|
1369
|
+
"digest": "sha256:7863c7bbe235a322737eedc4288ca4a3556e2f83a0e1bd13540d0933fc7cc372",
|
|
1360
1370
|
"digestAlgorithm": "rfc8785-sha256",
|
|
1361
1371
|
"executionPlan": {
|
|
1362
1372
|
"artifact": {
|
|
1363
|
-
"byteLength":
|
|
1364
|
-
"content": "
|
|
1373
|
+
"byteLength": 2164,
|
|
1374
|
+
"content": "eyJjb2RlS2luZCI6ImRpc2FibGVkIiwiY29udGFpbmVyIjp7ImltYWdlIjoiZ2hjci5pby9leGFtcGxlL3Rhc2siLCJpbmRleERpZ2VzdCI6InNoYTI1NjphYWFhYWFhYWFhYWFhYWFhYWFhYWFhYWFhYWFhYWFhYWFhYWFhYWFhYWFhYWFhYWFhYWFhYWFhYWFhYWFhYWFhIiwibWFuaWZlc3REaWdlc3QiOiJzaGEyNTY6YmJiYmJiYmJiYmJiYmJiYmJiYmJiYmJiYmJiYmJiYmJiYmJiYmJiYmJiYmJiYmJiYmJiYmJiYmJiYmJiYmJiYiIsInBsYXRmb3JtIjp7ImFyY2hpdGVjdHVyZSI6ImFtZDY0Iiwib3MiOiJsaW51eCJ9LCJzb3VyY2UiOiJldmFsdWF0b3ItdGFzay1jb250YWluZXIifSwiZXhlY3V0aW9uSWQiOiJleHBlcmltZW50LWNhbmRpZGF0ZS0wLTEiLCJoYXJuZXNzIjoiY29kZXgiLCJoYXJuZXNzVmVyc2lvbiI6IjEuMi4zIiwiaW5zdHJ1Y3Rpb25EZWxpdmVyeSI6eyJraW5kIjoic3RkaW4tdXRmOCJ9LCJraW5kIjoiYWdlbnQtY2FuZGlkYXRlLWV4ZWN1dGlvbi1wbGFuLW1hdGVyaWFsIiwibGF1bmNoIjp7ImFyZ3MiOlt7ImtpbmQiOiJwdWJsaWMiLCJ2YWx1ZSI6Ii1jIn0seyJraW5kIjoicHVibGljIiwidmFsdWUiOiJtb2RlbF9pbnN0cnVjdGlvbnNfZmlsZT0uY29kZXgvc3lzdGVtLXByb21wdC5tZCJ9XSwiY3dkIjp7InBhdGgiOiIuIiwid29ya3NwYWNlIjoidGFzayJ9LCJlbnYiOnsiUEFUSCI6eyJraW5kIjoicHVibGljIiwidmFsdWUiOiIvdXNyL2xvY2FsL2JpbjovdXNyL2JpbjovYmluIn19LCJleGVjdXRhYmxlIjoiY29kZXgifSwibGltaXRzIjp7Im1heENvc3RVc2QiOjUsIm1heElucHV0VG9rZW5zIjoxMDAwMDAsIm1heE1vZGVsQ2FsbHMiOjUwLCJtYXhPdXRwdXRUb2tlbnMiOjUwMDAwLCJtYXhTdGVwcyI6MTAwLCJ0aW1lb3V0TXMiOjYwMDAwfSwibWVtb3J5Ijp7Im1vZGUiOiJkaXNhYmxlZCJ9LCJtb2RlbCI6eyJhY2Nlc3MiOnsiZ3JhbnREaWdlc3QiOiJzaGEyNTY6Y2NjY2NjY2NjY2NjY2NjY2NjY2NjY2NjY2NjY2NjY2NjY2NjY2NjY2NjY2NjY2NjY2NjY2NjY2NjY2NjY2NjYyIsImtpbmQiOiJldmFsdWF0b3ItbWVkaWF0ZWQiLCJuZXR3b3JrIjp7ImRvbWFpbnMiOlsicm91dGVyLnRhbmdsZS50b29scyJdLCJtb2RlIjoiZ2F0ZXdheS1vbmx5In19LCJwb2xpY3kiOiJzaW5nbGUiLCJyZXNvbHZlZCI6eyJtb2RlbCI6Im1vZGVsLXNuYXBzaG90IiwicHJvdmlkZXIiOiJwcm92aWRlciIsInJlYXNvbmluZ0VmZm9ydCI6ImhpZ2giLCJyZXF1ZXN0ZWQiOiJwcm92aWRlci9tb2RlbCIsInNuYXBzaG90IjoibW9kZWwtc25hcHNob3QtMjAyNi0wNy0wMSJ9LCJyb3V0ZXMiOlt7ImtpbmQiOiJwcmltYXJ5IiwicmVxdWVzdGVkIjoicHJvdmlkZXIvbW9kZWwifV19LCJuZXR3b3JrIjp7Im1vZGUiOiJkaXNhYmxlZCJ9LCJwcm9maWxlIjp7Im1vdW50UGF0aHMiOlsiLmNvZGV4L3N5c3RlbS1wcm9tcHQubWQiLCJBR0VOVFMubWQiXSwicGxhbkRpZ2VzdCI6InNoYTI1Njo5NDUyY2I2MzJiMTI3YWFlMWRjOTRmYTgzOTZmNzc5MmYxNDI0N2E4ZGQyNGJkOTdmNmFlYzk5OGUxMGUyNjI2IiwidGFyZ2V0V29ya3NwYWNlIjoidGFzayJ9LCJydW5DZWxsIjp7ImFybSI6ImNhbmRpZGF0ZSIsImF0dGVtcHQiOjEsImJ1bmRsZURpZ2VzdCI6InNoYTI1Njo2MGZjYmIxYzcyODE5NGJkNTFkN2QxOWNiNzMyZDFjM2YxODgxZGNlN2UwYTYyNjZiNDFjOGI5OGNmZDY1NjkzIiwiZGlnZXN0Ijoic2hhMjU2OjU2NmNhMDE1MjYyYjA0ZGQ2Y2Y2NTVkOGUxMmY5N2FmNTgxMTZmZmJkYjMzYTMwNDkyMmY5MjI3NmZjNWIxZGIiLCJleHBlcmltZW50RGlnZXN0Ijoic2hhMjU2Ojc5ZDZlNzlkMmNiMGFmNTU0NTczMWY3YzhlZmQ1MmE2OGJjYzc0MzQ5YWE1NTdjMmVjMGZiMWNjYTk1MDE3NDUiLCJraW5kIjoiYWdlbnQtY2FuZGlkYXRlLXJ1bi1jZWxsIiwicmVwZXRpdGlvbiI6MSwic2VlZCI6MTAyLCJzdWl0ZURpZ2VzdCI6InNoYTI1Njo2NTMzNmFmZDg1MjU4Y2NmN2Q5YjI4ZTllMjEyZjhjYzUxZGQwOWEzZTdiMzJjNDY1MThhZTdjYzcwZDQ2NzA4IiwidGFza0RpZ2VzdCI6InNoYTI1Njo2YWE3MTgzMzFiNmRhMzEwZGNhYjZkZjJlMzM5N2NiNzcwODQwZTU1NWE4MmUwNjg5NTc3MTU3Yjg5YjRlMjc3IiwidGFza0luZGV4IjowfSwid29ya3NwYWNlcyI6eyJ0YXNrUm9vdCI6Ii93b3Jrc3BhY2UvdGFzayJ9fQ==",
|
|
1365
1375
|
"encoding": "base64",
|
|
1366
|
-
"sha256": "sha256:
|
|
1376
|
+
"sha256": "sha256:de21c216d858d02cdf57f6007ec52d452e19e6a78f759e934e53016aa7e788a1"
|
|
1367
1377
|
},
|
|
1368
|
-
"digest": "sha256:
|
|
1378
|
+
"digest": "sha256:de21c216d858d02cdf57f6007ec52d452e19e6a78f759e934e53016aa7e788a1",
|
|
1369
1379
|
"kind": "agent-candidate-execution-plan",
|
|
1370
1380
|
"material": {
|
|
1371
1381
|
"codeKind": "disabled",
|
|
@@ -1390,7 +1400,7 @@ var agent_improvement_proposal_default = {
|
|
|
1390
1400
|
"value": "-c"
|
|
1391
1401
|
}, {
|
|
1392
1402
|
"kind": "public",
|
|
1393
|
-
"value": "
|
|
1403
|
+
"value": "model_instructions_file=.codex/system-prompt.md"
|
|
1394
1404
|
}],
|
|
1395
1405
|
"cwd": {
|
|
1396
1406
|
"path": ".",
|
|
@@ -1435,8 +1445,8 @@ var agent_improvement_proposal_default = {
|
|
|
1435
1445
|
},
|
|
1436
1446
|
"network": { "mode": "disabled" },
|
|
1437
1447
|
"profile": {
|
|
1438
|
-
"mountPaths": ["AGENTS.md"],
|
|
1439
|
-
"planDigest": "sha256:
|
|
1448
|
+
"mountPaths": [".codex/system-prompt.md", "AGENTS.md"],
|
|
1449
|
+
"planDigest": "sha256:9452cb632b127aae1dc94fa8396f7792f14247a8dd24bd97f6aec998e10e2626",
|
|
1440
1450
|
"targetWorkspace": "task"
|
|
1441
1451
|
},
|
|
1442
1452
|
"runCell": {
|
|
@@ -1459,8 +1469,12 @@ var agent_improvement_proposal_default = {
|
|
|
1459
1469
|
"harnessVersion": "1.2.3",
|
|
1460
1470
|
"kind": "agent-candidate-materialization",
|
|
1461
1471
|
"profileActivation": {
|
|
1462
|
-
"digest": "sha256:
|
|
1472
|
+
"digest": "sha256:41c3e5ee69665488a2aefdf82b746696422bfbcd528fdedc110bc8e267cb0284",
|
|
1463
1473
|
"files": [{
|
|
1474
|
+
"content": "Return the exact measured answer.",
|
|
1475
|
+
"mode": 420,
|
|
1476
|
+
"path": ".codex/system-prompt.md"
|
|
1477
|
+
}, {
|
|
1464
1478
|
"content": "Inspect the repository, implement the fix, and run tests.\n",
|
|
1465
1479
|
"mode": 420,
|
|
1466
1480
|
"path": "AGENTS.md"
|
|
@@ -1468,16 +1482,20 @@ var agent_improvement_proposal_default = {
|
|
|
1468
1482
|
"kind": "agent-candidate-profile-activation",
|
|
1469
1483
|
"profilePlan": {
|
|
1470
1484
|
"artifact": {
|
|
1471
|
-
"byteLength":
|
|
1472
|
-
"content": "
|
|
1485
|
+
"byteLength": 532,
|
|
1486
|
+
"content": "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",
|
|
1473
1487
|
"encoding": "base64",
|
|
1474
|
-
"sha256": "sha256:
|
|
1488
|
+
"sha256": "sha256:9452cb632b127aae1dc94fa8396f7792f14247a8dd24bd97f6aec998e10e2626"
|
|
1475
1489
|
},
|
|
1476
|
-
"digest": "sha256:
|
|
1490
|
+
"digest": "sha256:9452cb632b127aae1dc94fa8396f7792f14247a8dd24bd97f6aec998e10e2626",
|
|
1477
1491
|
"kind": "agent-profile-workspace-plan",
|
|
1478
1492
|
"material": {
|
|
1479
1493
|
"env": {},
|
|
1480
1494
|
"files": [{
|
|
1495
|
+
"contentSha256": "sha256:7cd9c117d3377708275faa6a564af952f3f0c71352ae55833bf416377962d5b0",
|
|
1496
|
+
"mode": 420,
|
|
1497
|
+
"relPath": ".codex/system-prompt.md"
|
|
1498
|
+
}, {
|
|
1481
1499
|
"contentSha256": "sha256:39adbb9b7f571f0fd586e7795023d25909adbe9586bb21ee134d3d7248bf1599",
|
|
1482
1500
|
"mode": 420,
|
|
1483
1501
|
"relPath": "AGENTS.md"
|
|
@@ -1487,7 +1505,7 @@ var agent_improvement_proposal_default = {
|
|
|
1487
1505
|
"value": "-c"
|
|
1488
1506
|
}, {
|
|
1489
1507
|
"kind": "public",
|
|
1490
|
-
"value": "
|
|
1508
|
+
"value": "model_instructions_file=.codex/system-prompt.md"
|
|
1491
1509
|
}],
|
|
1492
1510
|
"harness": "codex",
|
|
1493
1511
|
"sourceProfileDigest": "sha256:8b1e2f0f7b832958381c29baebe4a961b151af97a0ce9d611a15ad0741aeaa75",
|
|
@@ -1509,12 +1527,12 @@ var agent_improvement_proposal_default = {
|
|
|
1509
1527
|
"byteLength": 1149,
|
|
1510
1528
|
"locator": {
|
|
1511
1529
|
"bucket": "candidate-test-artifacts",
|
|
1512
|
-
"key": "benchmark-result/
|
|
1530
|
+
"key": "benchmark-result/9c0d7cf14dda400f7c809ee8e31065b984c40b0ddd502b4bc8ba8410f78817df",
|
|
1513
1531
|
"kind": "s3"
|
|
1514
1532
|
},
|
|
1515
|
-
"sha256": "sha256:
|
|
1533
|
+
"sha256": "sha256:9c0d7cf14dda400f7c809ee8e31065b984c40b0ddd502b4bc8ba8410f78817df"
|
|
1516
1534
|
},
|
|
1517
|
-
"digest": "sha256:
|
|
1535
|
+
"digest": "sha256:9c0d7cf14dda400f7c809ee8e31065b984c40b0ddd502b4bc8ba8410f78817df",
|
|
1518
1536
|
"kind": "agent-candidate-benchmark-result",
|
|
1519
1537
|
"material": {
|
|
1520
1538
|
"dimensions": [{
|
|
@@ -1530,7 +1548,7 @@ var agent_improvement_proposal_default = {
|
|
|
1530
1548
|
},
|
|
1531
1549
|
"sha256": "sha256:9b9b3a1471309177261cfe65ea9c298e0dd372e4b5d087f8d35d7b732485373d"
|
|
1532
1550
|
},
|
|
1533
|
-
"executionPlanDigest": "sha256:
|
|
1551
|
+
"executionPlanDigest": "sha256:de21c216d858d02cdf57f6007ec52d452e19e6a78f759e934e53016aa7e788a1",
|
|
1534
1552
|
"grader": {
|
|
1535
1553
|
"artifact": {
|
|
1536
1554
|
"byteLength": 25,
|
|
@@ -1564,41 +1582,41 @@ var agent_improvement_proposal_default = {
|
|
|
1564
1582
|
"kind": "agent-candidate-benchmark-result-material",
|
|
1565
1583
|
"passed": true,
|
|
1566
1584
|
"score": 1,
|
|
1567
|
-
"taskOutcomeDigest": "sha256:
|
|
1585
|
+
"taskOutcomeDigest": "sha256:0c09ae883a9cdd298f1dc7c1fe042e286d5c26cca4146788e95107bde8d72f10"
|
|
1568
1586
|
}
|
|
1569
1587
|
},
|
|
1570
1588
|
"bundleDigest": "sha256:60fcbb1c728194bd51d7d19cb732d1c3f1881dce7e0a6266b41c8b98cfd65693",
|
|
1571
|
-
"digest": "sha256:
|
|
1589
|
+
"digest": "sha256:881b0bd00d824a260f4cd8d05b6acf61adf10ac5a5096851667fd9b52f9e7589",
|
|
1572
1590
|
"digestAlgorithm": "rfc8785-sha256",
|
|
1573
|
-
"executionPlanDigest": "sha256:
|
|
1591
|
+
"executionPlanDigest": "sha256:de21c216d858d02cdf57f6007ec52d452e19e6a78f759e934e53016aa7e788a1",
|
|
1574
1592
|
"executorCapture": {
|
|
1575
1593
|
"byteLength": 477,
|
|
1576
1594
|
"locator": {
|
|
1577
1595
|
"bucket": "candidate-test-artifacts",
|
|
1578
|
-
"key": "executor-capture/
|
|
1596
|
+
"key": "executor-capture/35430f2b64d29b9fdcab47bcfa76514fac5d81bbc506560060bc79ecc1dcc6a7",
|
|
1579
1597
|
"kind": "s3"
|
|
1580
1598
|
},
|
|
1581
|
-
"sha256": "sha256:
|
|
1599
|
+
"sha256": "sha256:35430f2b64d29b9fdcab47bcfa76514fac5d81bbc506560060bc79ecc1dcc6a7"
|
|
1582
1600
|
},
|
|
1583
1601
|
"kind": "agent-candidate-run",
|
|
1584
|
-
"materializationReceiptDigest": "sha256:
|
|
1602
|
+
"materializationReceiptDigest": "sha256:7863c7bbe235a322737eedc4288ca4a3556e2f83a0e1bd13540d0933fc7cc372",
|
|
1585
1603
|
"memory": { "mode": "disabled" },
|
|
1586
1604
|
"modelSettlement": {
|
|
1587
1605
|
"artifact": {
|
|
1588
1606
|
"byteLength": 643,
|
|
1589
1607
|
"locator": {
|
|
1590
1608
|
"bucket": "candidate-test-artifacts",
|
|
1591
|
-
"key": "model-settlement/
|
|
1609
|
+
"key": "model-settlement/55c1e80d4f9b3ef9de697d6f7b5fdc55e735aa34ca399a2b9e31e1e533bc10f1",
|
|
1592
1610
|
"kind": "s3"
|
|
1593
1611
|
},
|
|
1594
|
-
"sha256": "sha256:
|
|
1612
|
+
"sha256": "sha256:55c1e80d4f9b3ef9de697d6f7b5fdc55e735aa34ca399a2b9e31e1e533bc10f1"
|
|
1595
1613
|
},
|
|
1596
|
-
"digest": "sha256:
|
|
1614
|
+
"digest": "sha256:55c1e80d4f9b3ef9de697d6f7b5fdc55e735aa34ca399a2b9e31e1e533bc10f1",
|
|
1597
1615
|
"kind": "agent-candidate-model-settlement",
|
|
1598
1616
|
"material": {
|
|
1599
1617
|
"calls": [],
|
|
1600
1618
|
"closed": true,
|
|
1601
|
-
"executionPlanDigest": "sha256:
|
|
1619
|
+
"executionPlanDigest": "sha256:de21c216d858d02cdf57f6007ec52d452e19e6a78f759e934e53016aa7e788a1",
|
|
1602
1620
|
"grantDigest": "sha256:cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc",
|
|
1603
1621
|
"kind": "agent-candidate-model-settlement-material",
|
|
1604
1622
|
"preparationId": "candidate-preparation.nBqsDbKYlsiY4Tto5XaxkR3RCTop9PA9_kX-kOSzR8U",
|
|
@@ -1627,15 +1645,15 @@ var agent_improvement_proposal_default = {
|
|
|
1627
1645
|
"byteLength": 478,
|
|
1628
1646
|
"locator": {
|
|
1629
1647
|
"bucket": "candidate-test-artifacts",
|
|
1630
|
-
"key": "task-outcome/
|
|
1648
|
+
"key": "task-outcome/0c09ae883a9cdd298f1dc7c1fe042e286d5c26cca4146788e95107bde8d72f10",
|
|
1631
1649
|
"kind": "s3"
|
|
1632
1650
|
},
|
|
1633
|
-
"sha256": "sha256:
|
|
1651
|
+
"sha256": "sha256:0c09ae883a9cdd298f1dc7c1fe042e286d5c26cca4146788e95107bde8d72f10"
|
|
1634
1652
|
},
|
|
1635
|
-
"digest": "sha256:
|
|
1653
|
+
"digest": "sha256:0c09ae883a9cdd298f1dc7c1fe042e286d5c26cca4146788e95107bde8d72f10",
|
|
1636
1654
|
"kind": "agent-candidate-task-outcome",
|
|
1637
1655
|
"material": {
|
|
1638
|
-
"executionPlanDigest": "sha256:
|
|
1656
|
+
"executionPlanDigest": "sha256:de21c216d858d02cdf57f6007ec52d452e19e6a78f759e934e53016aa7e788a1",
|
|
1639
1657
|
"kind": "agent-candidate-task-outcome-material",
|
|
1640
1658
|
"outcome": {
|
|
1641
1659
|
"artifact": {
|
|
@@ -1669,10 +1687,10 @@ var agent_improvement_proposal_default = {
|
|
|
1669
1687
|
"byteLength": 750,
|
|
1670
1688
|
"locator": {
|
|
1671
1689
|
"bucket": "candidate-test-artifacts",
|
|
1672
|
-
"key": "trace/
|
|
1690
|
+
"key": "trace/2053f10cccb1b512885ac4bba76ac6606e07e64f7bf409bea3e8e2d397ef33c4",
|
|
1673
1691
|
"kind": "s3"
|
|
1674
1692
|
},
|
|
1675
|
-
"sha256": "sha256:
|
|
1693
|
+
"sha256": "sha256:2053f10cccb1b512885ac4bba76ac6606e07e64f7bf409bea3e8e2d397ef33c4"
|
|
1676
1694
|
},
|
|
1677
1695
|
"eventCount": 1,
|
|
1678
1696
|
"modelCallCount": 0
|
|
@@ -2029,7 +2047,7 @@ var agent_improvement_proposal_default = {
|
|
|
2029
2047
|
}
|
|
2030
2048
|
},
|
|
2031
2049
|
"candidate": {
|
|
2032
|
-
"digest": "sha256:
|
|
2050
|
+
"digest": "sha256:0df77fe605339691ec8f0615445069c363b461ee853473b4d50fa091d2388109",
|
|
2033
2051
|
"kind": "agent-candidate-execution-evidence",
|
|
2034
2052
|
"materializationReceipt": {
|
|
2035
2053
|
"benchmark": {
|
|
@@ -2064,16 +2082,16 @@ var agent_improvement_proposal_default = {
|
|
|
2064
2082
|
},
|
|
2065
2083
|
"source": "evaluator-task-container"
|
|
2066
2084
|
},
|
|
2067
|
-
"digest": "sha256:
|
|
2085
|
+
"digest": "sha256:2ad73653df44003c994fd96f5fe7a1ffeb423b0a3ae89d17300dda426c0daded",
|
|
2068
2086
|
"digestAlgorithm": "rfc8785-sha256",
|
|
2069
2087
|
"executionPlan": {
|
|
2070
2088
|
"artifact": {
|
|
2071
|
-
"byteLength":
|
|
2072
|
-
"content": "
|
|
2089
|
+
"byteLength": 2164,
|
|
2090
|
+
"content": "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",
|
|
2073
2091
|
"encoding": "base64",
|
|
2074
|
-
"sha256": "sha256:
|
|
2092
|
+
"sha256": "sha256:0446656d74500d99e7434eb1391a8e3e654997ff71416e835cfcaeb5d79fe8de"
|
|
2075
2093
|
},
|
|
2076
|
-
"digest": "sha256:
|
|
2094
|
+
"digest": "sha256:0446656d74500d99e7434eb1391a8e3e654997ff71416e835cfcaeb5d79fe8de",
|
|
2077
2095
|
"kind": "agent-candidate-execution-plan",
|
|
2078
2096
|
"material": {
|
|
2079
2097
|
"codeKind": "disabled",
|
|
@@ -2098,7 +2116,7 @@ var agent_improvement_proposal_default = {
|
|
|
2098
2116
|
"value": "-c"
|
|
2099
2117
|
}, {
|
|
2100
2118
|
"kind": "public",
|
|
2101
|
-
"value": "
|
|
2119
|
+
"value": "model_instructions_file=.codex/system-prompt.md"
|
|
2102
2120
|
}],
|
|
2103
2121
|
"cwd": {
|
|
2104
2122
|
"path": ".",
|
|
@@ -2143,8 +2161,8 @@ var agent_improvement_proposal_default = {
|
|
|
2143
2161
|
},
|
|
2144
2162
|
"network": { "mode": "disabled" },
|
|
2145
2163
|
"profile": {
|
|
2146
|
-
"mountPaths": ["AGENTS.md"],
|
|
2147
|
-
"planDigest": "sha256:
|
|
2164
|
+
"mountPaths": [".codex/system-prompt.md", "AGENTS.md"],
|
|
2165
|
+
"planDigest": "sha256:9452cb632b127aae1dc94fa8396f7792f14247a8dd24bd97f6aec998e10e2626",
|
|
2148
2166
|
"targetWorkspace": "task"
|
|
2149
2167
|
},
|
|
2150
2168
|
"runCell": {
|
|
@@ -2167,8 +2185,12 @@ var agent_improvement_proposal_default = {
|
|
|
2167
2185
|
"harnessVersion": "1.2.3",
|
|
2168
2186
|
"kind": "agent-candidate-materialization",
|
|
2169
2187
|
"profileActivation": {
|
|
2170
|
-
"digest": "sha256:
|
|
2188
|
+
"digest": "sha256:41c3e5ee69665488a2aefdf82b746696422bfbcd528fdedc110bc8e267cb0284",
|
|
2171
2189
|
"files": [{
|
|
2190
|
+
"content": "Return the exact measured answer.",
|
|
2191
|
+
"mode": 420,
|
|
2192
|
+
"path": ".codex/system-prompt.md"
|
|
2193
|
+
}, {
|
|
2172
2194
|
"content": "Inspect the repository, implement the fix, and run tests.\n",
|
|
2173
2195
|
"mode": 420,
|
|
2174
2196
|
"path": "AGENTS.md"
|
|
@@ -2176,16 +2198,20 @@ var agent_improvement_proposal_default = {
|
|
|
2176
2198
|
"kind": "agent-candidate-profile-activation",
|
|
2177
2199
|
"profilePlan": {
|
|
2178
2200
|
"artifact": {
|
|
2179
|
-
"byteLength":
|
|
2180
|
-
"content": "
|
|
2201
|
+
"byteLength": 532,
|
|
2202
|
+
"content": "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",
|
|
2181
2203
|
"encoding": "base64",
|
|
2182
|
-
"sha256": "sha256:
|
|
2204
|
+
"sha256": "sha256:9452cb632b127aae1dc94fa8396f7792f14247a8dd24bd97f6aec998e10e2626"
|
|
2183
2205
|
},
|
|
2184
|
-
"digest": "sha256:
|
|
2206
|
+
"digest": "sha256:9452cb632b127aae1dc94fa8396f7792f14247a8dd24bd97f6aec998e10e2626",
|
|
2185
2207
|
"kind": "agent-profile-workspace-plan",
|
|
2186
2208
|
"material": {
|
|
2187
2209
|
"env": {},
|
|
2188
2210
|
"files": [{
|
|
2211
|
+
"contentSha256": "sha256:7cd9c117d3377708275faa6a564af952f3f0c71352ae55833bf416377962d5b0",
|
|
2212
|
+
"mode": 420,
|
|
2213
|
+
"relPath": ".codex/system-prompt.md"
|
|
2214
|
+
}, {
|
|
2189
2215
|
"contentSha256": "sha256:39adbb9b7f571f0fd586e7795023d25909adbe9586bb21ee134d3d7248bf1599",
|
|
2190
2216
|
"mode": 420,
|
|
2191
2217
|
"relPath": "AGENTS.md"
|
|
@@ -2195,7 +2221,7 @@ var agent_improvement_proposal_default = {
|
|
|
2195
2221
|
"value": "-c"
|
|
2196
2222
|
}, {
|
|
2197
2223
|
"kind": "public",
|
|
2198
|
-
"value": "
|
|
2224
|
+
"value": "model_instructions_file=.codex/system-prompt.md"
|
|
2199
2225
|
}],
|
|
2200
2226
|
"harness": "codex",
|
|
2201
2227
|
"sourceProfileDigest": "sha256:8b1e2f0f7b832958381c29baebe4a961b151af97a0ce9d611a15ad0741aeaa75",
|
|
@@ -2217,12 +2243,12 @@ var agent_improvement_proposal_default = {
|
|
|
2217
2243
|
"byteLength": 1149,
|
|
2218
2244
|
"locator": {
|
|
2219
2245
|
"bucket": "candidate-test-artifacts",
|
|
2220
|
-
"key": "benchmark-result/
|
|
2246
|
+
"key": "benchmark-result/1fa99977490c637dcfb7940ad64866dd3c9c0de392b7d6f3b3f57c25ac6b6e3d",
|
|
2221
2247
|
"kind": "s3"
|
|
2222
2248
|
},
|
|
2223
|
-
"sha256": "sha256:
|
|
2249
|
+
"sha256": "sha256:1fa99977490c637dcfb7940ad64866dd3c9c0de392b7d6f3b3f57c25ac6b6e3d"
|
|
2224
2250
|
},
|
|
2225
|
-
"digest": "sha256:
|
|
2251
|
+
"digest": "sha256:1fa99977490c637dcfb7940ad64866dd3c9c0de392b7d6f3b3f57c25ac6b6e3d",
|
|
2226
2252
|
"kind": "agent-candidate-benchmark-result",
|
|
2227
2253
|
"material": {
|
|
2228
2254
|
"dimensions": [{
|
|
@@ -2238,7 +2264,7 @@ var agent_improvement_proposal_default = {
|
|
|
2238
2264
|
},
|
|
2239
2265
|
"sha256": "sha256:9b9b3a1471309177261cfe65ea9c298e0dd372e4b5d087f8d35d7b732485373d"
|
|
2240
2266
|
},
|
|
2241
|
-
"executionPlanDigest": "sha256:
|
|
2267
|
+
"executionPlanDigest": "sha256:0446656d74500d99e7434eb1391a8e3e654997ff71416e835cfcaeb5d79fe8de",
|
|
2242
2268
|
"grader": {
|
|
2243
2269
|
"artifact": {
|
|
2244
2270
|
"byteLength": 25,
|
|
@@ -2272,41 +2298,41 @@ var agent_improvement_proposal_default = {
|
|
|
2272
2298
|
"kind": "agent-candidate-benchmark-result-material",
|
|
2273
2299
|
"passed": true,
|
|
2274
2300
|
"score": 1,
|
|
2275
|
-
"taskOutcomeDigest": "sha256:
|
|
2301
|
+
"taskOutcomeDigest": "sha256:6302cbcc8d554df8e1797747df6dbb65abd32cf7a0450a85959daddb0610e961"
|
|
2276
2302
|
}
|
|
2277
2303
|
},
|
|
2278
2304
|
"bundleDigest": "sha256:60fcbb1c728194bd51d7d19cb732d1c3f1881dce7e0a6266b41c8b98cfd65693",
|
|
2279
|
-
"digest": "sha256:
|
|
2305
|
+
"digest": "sha256:42e03ef75af59ebb2d6d6bbd0624b864d45354e0f56cd8b4e4e68a6af3639cd7",
|
|
2280
2306
|
"digestAlgorithm": "rfc8785-sha256",
|
|
2281
|
-
"executionPlanDigest": "sha256:
|
|
2307
|
+
"executionPlanDigest": "sha256:0446656d74500d99e7434eb1391a8e3e654997ff71416e835cfcaeb5d79fe8de",
|
|
2282
2308
|
"executorCapture": {
|
|
2283
2309
|
"byteLength": 477,
|
|
2284
2310
|
"locator": {
|
|
2285
2311
|
"bucket": "candidate-test-artifacts",
|
|
2286
|
-
"key": "executor-capture/
|
|
2312
|
+
"key": "executor-capture/ed0e2baa2024d09010ab1c9fc4e4ee17e764cbd042e76a1078112e9c307ff90d",
|
|
2287
2313
|
"kind": "s3"
|
|
2288
2314
|
},
|
|
2289
|
-
"sha256": "sha256:
|
|
2315
|
+
"sha256": "sha256:ed0e2baa2024d09010ab1c9fc4e4ee17e764cbd042e76a1078112e9c307ff90d"
|
|
2290
2316
|
},
|
|
2291
2317
|
"kind": "agent-candidate-run",
|
|
2292
|
-
"materializationReceiptDigest": "sha256:
|
|
2318
|
+
"materializationReceiptDigest": "sha256:2ad73653df44003c994fd96f5fe7a1ffeb423b0a3ae89d17300dda426c0daded",
|
|
2293
2319
|
"memory": { "mode": "disabled" },
|
|
2294
2320
|
"modelSettlement": {
|
|
2295
2321
|
"artifact": {
|
|
2296
2322
|
"byteLength": 643,
|
|
2297
2323
|
"locator": {
|
|
2298
2324
|
"bucket": "candidate-test-artifacts",
|
|
2299
|
-
"key": "model-settlement/
|
|
2325
|
+
"key": "model-settlement/1eaaac5858a9318661b2b05c2f5a9d69e3bbfd401fabf9a9bdc52971c32d6df3",
|
|
2300
2326
|
"kind": "s3"
|
|
2301
2327
|
},
|
|
2302
|
-
"sha256": "sha256:
|
|
2328
|
+
"sha256": "sha256:1eaaac5858a9318661b2b05c2f5a9d69e3bbfd401fabf9a9bdc52971c32d6df3"
|
|
2303
2329
|
},
|
|
2304
|
-
"digest": "sha256:
|
|
2330
|
+
"digest": "sha256:1eaaac5858a9318661b2b05c2f5a9d69e3bbfd401fabf9a9bdc52971c32d6df3",
|
|
2305
2331
|
"kind": "agent-candidate-model-settlement",
|
|
2306
2332
|
"material": {
|
|
2307
2333
|
"calls": [],
|
|
2308
2334
|
"closed": true,
|
|
2309
|
-
"executionPlanDigest": "sha256:
|
|
2335
|
+
"executionPlanDigest": "sha256:0446656d74500d99e7434eb1391a8e3e654997ff71416e835cfcaeb5d79fe8de",
|
|
2310
2336
|
"grantDigest": "sha256:cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc",
|
|
2311
2337
|
"kind": "agent-candidate-model-settlement-material",
|
|
2312
2338
|
"preparationId": "candidate-preparation.nBqsDbKYlsiY4Tto5XaxkR3RCTop9PA9_kX-kOSzR8U",
|
|
@@ -2335,15 +2361,15 @@ var agent_improvement_proposal_default = {
|
|
|
2335
2361
|
"byteLength": 478,
|
|
2336
2362
|
"locator": {
|
|
2337
2363
|
"bucket": "candidate-test-artifacts",
|
|
2338
|
-
"key": "task-outcome/
|
|
2364
|
+
"key": "task-outcome/6302cbcc8d554df8e1797747df6dbb65abd32cf7a0450a85959daddb0610e961",
|
|
2339
2365
|
"kind": "s3"
|
|
2340
2366
|
},
|
|
2341
|
-
"sha256": "sha256:
|
|
2367
|
+
"sha256": "sha256:6302cbcc8d554df8e1797747df6dbb65abd32cf7a0450a85959daddb0610e961"
|
|
2342
2368
|
},
|
|
2343
|
-
"digest": "sha256:
|
|
2369
|
+
"digest": "sha256:6302cbcc8d554df8e1797747df6dbb65abd32cf7a0450a85959daddb0610e961",
|
|
2344
2370
|
"kind": "agent-candidate-task-outcome",
|
|
2345
2371
|
"material": {
|
|
2346
|
-
"executionPlanDigest": "sha256:
|
|
2372
|
+
"executionPlanDigest": "sha256:0446656d74500d99e7434eb1391a8e3e654997ff71416e835cfcaeb5d79fe8de",
|
|
2347
2373
|
"kind": "agent-candidate-task-outcome-material",
|
|
2348
2374
|
"outcome": {
|
|
2349
2375
|
"artifact": {
|
|
@@ -2377,10 +2403,10 @@ var agent_improvement_proposal_default = {
|
|
|
2377
2403
|
"byteLength": 750,
|
|
2378
2404
|
"locator": {
|
|
2379
2405
|
"bucket": "candidate-test-artifacts",
|
|
2380
|
-
"key": "trace/
|
|
2406
|
+
"key": "trace/11e6a7b8ae5b5f45db98275619258ba572035fbd6af67bb5c4eff44b36ff0995",
|
|
2381
2407
|
"kind": "s3"
|
|
2382
2408
|
},
|
|
2383
|
-
"sha256": "sha256:
|
|
2409
|
+
"sha256": "sha256:11e6a7b8ae5b5f45db98275619258ba572035fbd6af67bb5c4eff44b36ff0995"
|
|
2384
2410
|
},
|
|
2385
2411
|
"eventCount": 1,
|
|
2386
2412
|
"modelCallCount": 0
|
|
@@ -2737,7 +2763,7 @@ var agent_improvement_proposal_default = {
|
|
|
2737
2763
|
}
|
|
2738
2764
|
},
|
|
2739
2765
|
"candidate": {
|
|
2740
|
-
"digest": "sha256:
|
|
2766
|
+
"digest": "sha256:f08f4f1d6b9daf284f9ca86076e6dcd73cb310d7afa3f807549ee57f7ef41889",
|
|
2741
2767
|
"kind": "agent-candidate-execution-evidence",
|
|
2742
2768
|
"materializationReceipt": {
|
|
2743
2769
|
"benchmark": {
|
|
@@ -2772,16 +2798,16 @@ var agent_improvement_proposal_default = {
|
|
|
2772
2798
|
},
|
|
2773
2799
|
"source": "evaluator-task-container"
|
|
2774
2800
|
},
|
|
2775
|
-
"digest": "sha256:
|
|
2801
|
+
"digest": "sha256:34492818980f61445ea3e76dc055b3ef1e196dba10a1dee311a99a267af0801f",
|
|
2776
2802
|
"digestAlgorithm": "rfc8785-sha256",
|
|
2777
2803
|
"executionPlan": {
|
|
2778
2804
|
"artifact": {
|
|
2779
|
-
"byteLength":
|
|
2780
|
-
"content": "
|
|
2805
|
+
"byteLength": 2164,
|
|
2806
|
+
"content": "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",
|
|
2781
2807
|
"encoding": "base64",
|
|
2782
|
-
"sha256": "sha256:
|
|
2808
|
+
"sha256": "sha256:ef2e2d33a65e201654f55167905633e8360e2ff3375267248eaa710cadbd3486"
|
|
2783
2809
|
},
|
|
2784
|
-
"digest": "sha256:
|
|
2810
|
+
"digest": "sha256:ef2e2d33a65e201654f55167905633e8360e2ff3375267248eaa710cadbd3486",
|
|
2785
2811
|
"kind": "agent-candidate-execution-plan",
|
|
2786
2812
|
"material": {
|
|
2787
2813
|
"codeKind": "disabled",
|
|
@@ -2806,7 +2832,7 @@ var agent_improvement_proposal_default = {
|
|
|
2806
2832
|
"value": "-c"
|
|
2807
2833
|
}, {
|
|
2808
2834
|
"kind": "public",
|
|
2809
|
-
"value": "
|
|
2835
|
+
"value": "model_instructions_file=.codex/system-prompt.md"
|
|
2810
2836
|
}],
|
|
2811
2837
|
"cwd": {
|
|
2812
2838
|
"path": ".",
|
|
@@ -2851,8 +2877,8 @@ var agent_improvement_proposal_default = {
|
|
|
2851
2877
|
},
|
|
2852
2878
|
"network": { "mode": "disabled" },
|
|
2853
2879
|
"profile": {
|
|
2854
|
-
"mountPaths": ["AGENTS.md"],
|
|
2855
|
-
"planDigest": "sha256:
|
|
2880
|
+
"mountPaths": [".codex/system-prompt.md", "AGENTS.md"],
|
|
2881
|
+
"planDigest": "sha256:9452cb632b127aae1dc94fa8396f7792f14247a8dd24bd97f6aec998e10e2626",
|
|
2856
2882
|
"targetWorkspace": "task"
|
|
2857
2883
|
},
|
|
2858
2884
|
"runCell": {
|
|
@@ -2875,8 +2901,12 @@ var agent_improvement_proposal_default = {
|
|
|
2875
2901
|
"harnessVersion": "1.2.3",
|
|
2876
2902
|
"kind": "agent-candidate-materialization",
|
|
2877
2903
|
"profileActivation": {
|
|
2878
|
-
"digest": "sha256:
|
|
2904
|
+
"digest": "sha256:41c3e5ee69665488a2aefdf82b746696422bfbcd528fdedc110bc8e267cb0284",
|
|
2879
2905
|
"files": [{
|
|
2906
|
+
"content": "Return the exact measured answer.",
|
|
2907
|
+
"mode": 420,
|
|
2908
|
+
"path": ".codex/system-prompt.md"
|
|
2909
|
+
}, {
|
|
2880
2910
|
"content": "Inspect the repository, implement the fix, and run tests.\n",
|
|
2881
2911
|
"mode": 420,
|
|
2882
2912
|
"path": "AGENTS.md"
|
|
@@ -2884,16 +2914,20 @@ var agent_improvement_proposal_default = {
|
|
|
2884
2914
|
"kind": "agent-candidate-profile-activation",
|
|
2885
2915
|
"profilePlan": {
|
|
2886
2916
|
"artifact": {
|
|
2887
|
-
"byteLength":
|
|
2888
|
-
"content": "
|
|
2917
|
+
"byteLength": 532,
|
|
2918
|
+
"content": "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",
|
|
2889
2919
|
"encoding": "base64",
|
|
2890
|
-
"sha256": "sha256:
|
|
2920
|
+
"sha256": "sha256:9452cb632b127aae1dc94fa8396f7792f14247a8dd24bd97f6aec998e10e2626"
|
|
2891
2921
|
},
|
|
2892
|
-
"digest": "sha256:
|
|
2922
|
+
"digest": "sha256:9452cb632b127aae1dc94fa8396f7792f14247a8dd24bd97f6aec998e10e2626",
|
|
2893
2923
|
"kind": "agent-profile-workspace-plan",
|
|
2894
2924
|
"material": {
|
|
2895
2925
|
"env": {},
|
|
2896
2926
|
"files": [{
|
|
2927
|
+
"contentSha256": "sha256:7cd9c117d3377708275faa6a564af952f3f0c71352ae55833bf416377962d5b0",
|
|
2928
|
+
"mode": 420,
|
|
2929
|
+
"relPath": ".codex/system-prompt.md"
|
|
2930
|
+
}, {
|
|
2897
2931
|
"contentSha256": "sha256:39adbb9b7f571f0fd586e7795023d25909adbe9586bb21ee134d3d7248bf1599",
|
|
2898
2932
|
"mode": 420,
|
|
2899
2933
|
"relPath": "AGENTS.md"
|
|
@@ -2903,7 +2937,7 @@ var agent_improvement_proposal_default = {
|
|
|
2903
2937
|
"value": "-c"
|
|
2904
2938
|
}, {
|
|
2905
2939
|
"kind": "public",
|
|
2906
|
-
"value": "
|
|
2940
|
+
"value": "model_instructions_file=.codex/system-prompt.md"
|
|
2907
2941
|
}],
|
|
2908
2942
|
"harness": "codex",
|
|
2909
2943
|
"sourceProfileDigest": "sha256:8b1e2f0f7b832958381c29baebe4a961b151af97a0ce9d611a15ad0741aeaa75",
|
|
@@ -2925,12 +2959,12 @@ var agent_improvement_proposal_default = {
|
|
|
2925
2959
|
"byteLength": 1149,
|
|
2926
2960
|
"locator": {
|
|
2927
2961
|
"bucket": "candidate-test-artifacts",
|
|
2928
|
-
"key": "benchmark-result/
|
|
2962
|
+
"key": "benchmark-result/d51189ae2867d126ec817067b29d03926963d77ce0f14d376ef31fe4918664eb",
|
|
2929
2963
|
"kind": "s3"
|
|
2930
2964
|
},
|
|
2931
|
-
"sha256": "sha256:
|
|
2965
|
+
"sha256": "sha256:d51189ae2867d126ec817067b29d03926963d77ce0f14d376ef31fe4918664eb"
|
|
2932
2966
|
},
|
|
2933
|
-
"digest": "sha256:
|
|
2967
|
+
"digest": "sha256:d51189ae2867d126ec817067b29d03926963d77ce0f14d376ef31fe4918664eb",
|
|
2934
2968
|
"kind": "agent-candidate-benchmark-result",
|
|
2935
2969
|
"material": {
|
|
2936
2970
|
"dimensions": [{
|
|
@@ -2946,7 +2980,7 @@ var agent_improvement_proposal_default = {
|
|
|
2946
2980
|
},
|
|
2947
2981
|
"sha256": "sha256:9b9b3a1471309177261cfe65ea9c298e0dd372e4b5d087f8d35d7b732485373d"
|
|
2948
2982
|
},
|
|
2949
|
-
"executionPlanDigest": "sha256:
|
|
2983
|
+
"executionPlanDigest": "sha256:ef2e2d33a65e201654f55167905633e8360e2ff3375267248eaa710cadbd3486",
|
|
2950
2984
|
"grader": {
|
|
2951
2985
|
"artifact": {
|
|
2952
2986
|
"byteLength": 25,
|
|
@@ -2980,41 +3014,41 @@ var agent_improvement_proposal_default = {
|
|
|
2980
3014
|
"kind": "agent-candidate-benchmark-result-material",
|
|
2981
3015
|
"passed": true,
|
|
2982
3016
|
"score": 1,
|
|
2983
|
-
"taskOutcomeDigest": "sha256:
|
|
3017
|
+
"taskOutcomeDigest": "sha256:4e4a15c4ceed3977a625414be391732eb664dd27aab7792a30f27cf58dafc2fe"
|
|
2984
3018
|
}
|
|
2985
3019
|
},
|
|
2986
3020
|
"bundleDigest": "sha256:60fcbb1c728194bd51d7d19cb732d1c3f1881dce7e0a6266b41c8b98cfd65693",
|
|
2987
|
-
"digest": "sha256:
|
|
3021
|
+
"digest": "sha256:f339a77f43d73b83faa392f0758a738b3f56a5a5154bf79323e5cbb78882c54f",
|
|
2988
3022
|
"digestAlgorithm": "rfc8785-sha256",
|
|
2989
|
-
"executionPlanDigest": "sha256:
|
|
3023
|
+
"executionPlanDigest": "sha256:ef2e2d33a65e201654f55167905633e8360e2ff3375267248eaa710cadbd3486",
|
|
2990
3024
|
"executorCapture": {
|
|
2991
3025
|
"byteLength": 477,
|
|
2992
3026
|
"locator": {
|
|
2993
3027
|
"bucket": "candidate-test-artifacts",
|
|
2994
|
-
"key": "executor-capture/
|
|
3028
|
+
"key": "executor-capture/c428a8407b32b7f9b8ca26fe52c515fb6b4f0a324a1948719f91ee90ad2a43e4",
|
|
2995
3029
|
"kind": "s3"
|
|
2996
3030
|
},
|
|
2997
|
-
"sha256": "sha256:
|
|
3031
|
+
"sha256": "sha256:c428a8407b32b7f9b8ca26fe52c515fb6b4f0a324a1948719f91ee90ad2a43e4"
|
|
2998
3032
|
},
|
|
2999
3033
|
"kind": "agent-candidate-run",
|
|
3000
|
-
"materializationReceiptDigest": "sha256:
|
|
3034
|
+
"materializationReceiptDigest": "sha256:34492818980f61445ea3e76dc055b3ef1e196dba10a1dee311a99a267af0801f",
|
|
3001
3035
|
"memory": { "mode": "disabled" },
|
|
3002
3036
|
"modelSettlement": {
|
|
3003
3037
|
"artifact": {
|
|
3004
3038
|
"byteLength": 643,
|
|
3005
3039
|
"locator": {
|
|
3006
3040
|
"bucket": "candidate-test-artifacts",
|
|
3007
|
-
"key": "model-settlement/
|
|
3041
|
+
"key": "model-settlement/badb327758efd85ae21cd2a1b9a8b90c9171b6dea67c903f80b5c1cdea0f4290",
|
|
3008
3042
|
"kind": "s3"
|
|
3009
3043
|
},
|
|
3010
|
-
"sha256": "sha256:
|
|
3044
|
+
"sha256": "sha256:badb327758efd85ae21cd2a1b9a8b90c9171b6dea67c903f80b5c1cdea0f4290"
|
|
3011
3045
|
},
|
|
3012
|
-
"digest": "sha256:
|
|
3046
|
+
"digest": "sha256:badb327758efd85ae21cd2a1b9a8b90c9171b6dea67c903f80b5c1cdea0f4290",
|
|
3013
3047
|
"kind": "agent-candidate-model-settlement",
|
|
3014
3048
|
"material": {
|
|
3015
3049
|
"calls": [],
|
|
3016
3050
|
"closed": true,
|
|
3017
|
-
"executionPlanDigest": "sha256:
|
|
3051
|
+
"executionPlanDigest": "sha256:ef2e2d33a65e201654f55167905633e8360e2ff3375267248eaa710cadbd3486",
|
|
3018
3052
|
"grantDigest": "sha256:cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc",
|
|
3019
3053
|
"kind": "agent-candidate-model-settlement-material",
|
|
3020
3054
|
"preparationId": "candidate-preparation.nBqsDbKYlsiY4Tto5XaxkR3RCTop9PA9_kX-kOSzR8U",
|
|
@@ -3043,15 +3077,15 @@ var agent_improvement_proposal_default = {
|
|
|
3043
3077
|
"byteLength": 478,
|
|
3044
3078
|
"locator": {
|
|
3045
3079
|
"bucket": "candidate-test-artifacts",
|
|
3046
|
-
"key": "task-outcome/
|
|
3080
|
+
"key": "task-outcome/4e4a15c4ceed3977a625414be391732eb664dd27aab7792a30f27cf58dafc2fe",
|
|
3047
3081
|
"kind": "s3"
|
|
3048
3082
|
},
|
|
3049
|
-
"sha256": "sha256:
|
|
3083
|
+
"sha256": "sha256:4e4a15c4ceed3977a625414be391732eb664dd27aab7792a30f27cf58dafc2fe"
|
|
3050
3084
|
},
|
|
3051
|
-
"digest": "sha256:
|
|
3085
|
+
"digest": "sha256:4e4a15c4ceed3977a625414be391732eb664dd27aab7792a30f27cf58dafc2fe",
|
|
3052
3086
|
"kind": "agent-candidate-task-outcome",
|
|
3053
3087
|
"material": {
|
|
3054
|
-
"executionPlanDigest": "sha256:
|
|
3088
|
+
"executionPlanDigest": "sha256:ef2e2d33a65e201654f55167905633e8360e2ff3375267248eaa710cadbd3486",
|
|
3055
3089
|
"kind": "agent-candidate-task-outcome-material",
|
|
3056
3090
|
"outcome": {
|
|
3057
3091
|
"artifact": {
|
|
@@ -3085,10 +3119,10 @@ var agent_improvement_proposal_default = {
|
|
|
3085
3119
|
"byteLength": 750,
|
|
3086
3120
|
"locator": {
|
|
3087
3121
|
"bucket": "candidate-test-artifacts",
|
|
3088
|
-
"key": "trace/
|
|
3122
|
+
"key": "trace/71729abaeb34c5f56445212ab0924b124c16061dd574eeeea67919d79435db1b",
|
|
3089
3123
|
"kind": "s3"
|
|
3090
3124
|
},
|
|
3091
|
-
"sha256": "sha256:
|
|
3125
|
+
"sha256": "sha256:71729abaeb34c5f56445212ab0924b124c16061dd574eeeea67919d79435db1b"
|
|
3092
3126
|
},
|
|
3093
3127
|
"eventCount": 1,
|
|
3094
3128
|
"modelCallCount": 0
|
|
@@ -3445,7 +3479,7 @@ var agent_improvement_proposal_default = {
|
|
|
3445
3479
|
}
|
|
3446
3480
|
},
|
|
3447
3481
|
"candidate": {
|
|
3448
|
-
"digest": "sha256:
|
|
3482
|
+
"digest": "sha256:b55b58b1677b734bca206606a278332745f7be844da243b09bb9a81e3adbaec5",
|
|
3449
3483
|
"kind": "agent-candidate-execution-evidence",
|
|
3450
3484
|
"materializationReceipt": {
|
|
3451
3485
|
"benchmark": {
|
|
@@ -3480,16 +3514,16 @@ var agent_improvement_proposal_default = {
|
|
|
3480
3514
|
},
|
|
3481
3515
|
"source": "evaluator-task-container"
|
|
3482
3516
|
},
|
|
3483
|
-
"digest": "sha256:
|
|
3517
|
+
"digest": "sha256:2bc6a75f77ff4de870e326a766e8d5a7bb5c1f6fb271a6041bb426eb9d846c29",
|
|
3484
3518
|
"digestAlgorithm": "rfc8785-sha256",
|
|
3485
3519
|
"executionPlan": {
|
|
3486
3520
|
"artifact": {
|
|
3487
|
-
"byteLength":
|
|
3488
|
-
"content": "
|
|
3521
|
+
"byteLength": 2164,
|
|
3522
|
+
"content": "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",
|
|
3489
3523
|
"encoding": "base64",
|
|
3490
|
-
"sha256": "sha256:
|
|
3524
|
+
"sha256": "sha256:e247492e04c77a23efb2c86c1f54012bbe29d0e64021b19dbe85df01255fa1f4"
|
|
3491
3525
|
},
|
|
3492
|
-
"digest": "sha256:
|
|
3526
|
+
"digest": "sha256:e247492e04c77a23efb2c86c1f54012bbe29d0e64021b19dbe85df01255fa1f4",
|
|
3493
3527
|
"kind": "agent-candidate-execution-plan",
|
|
3494
3528
|
"material": {
|
|
3495
3529
|
"codeKind": "disabled",
|
|
@@ -3514,7 +3548,7 @@ var agent_improvement_proposal_default = {
|
|
|
3514
3548
|
"value": "-c"
|
|
3515
3549
|
}, {
|
|
3516
3550
|
"kind": "public",
|
|
3517
|
-
"value": "
|
|
3551
|
+
"value": "model_instructions_file=.codex/system-prompt.md"
|
|
3518
3552
|
}],
|
|
3519
3553
|
"cwd": {
|
|
3520
3554
|
"path": ".",
|
|
@@ -3559,8 +3593,8 @@ var agent_improvement_proposal_default = {
|
|
|
3559
3593
|
},
|
|
3560
3594
|
"network": { "mode": "disabled" },
|
|
3561
3595
|
"profile": {
|
|
3562
|
-
"mountPaths": ["AGENTS.md"],
|
|
3563
|
-
"planDigest": "sha256:
|
|
3596
|
+
"mountPaths": [".codex/system-prompt.md", "AGENTS.md"],
|
|
3597
|
+
"planDigest": "sha256:9452cb632b127aae1dc94fa8396f7792f14247a8dd24bd97f6aec998e10e2626",
|
|
3564
3598
|
"targetWorkspace": "task"
|
|
3565
3599
|
},
|
|
3566
3600
|
"runCell": {
|
|
@@ -3583,8 +3617,12 @@ var agent_improvement_proposal_default = {
|
|
|
3583
3617
|
"harnessVersion": "1.2.3",
|
|
3584
3618
|
"kind": "agent-candidate-materialization",
|
|
3585
3619
|
"profileActivation": {
|
|
3586
|
-
"digest": "sha256:
|
|
3620
|
+
"digest": "sha256:41c3e5ee69665488a2aefdf82b746696422bfbcd528fdedc110bc8e267cb0284",
|
|
3587
3621
|
"files": [{
|
|
3622
|
+
"content": "Return the exact measured answer.",
|
|
3623
|
+
"mode": 420,
|
|
3624
|
+
"path": ".codex/system-prompt.md"
|
|
3625
|
+
}, {
|
|
3588
3626
|
"content": "Inspect the repository, implement the fix, and run tests.\n",
|
|
3589
3627
|
"mode": 420,
|
|
3590
3628
|
"path": "AGENTS.md"
|
|
@@ -3592,16 +3630,20 @@ var agent_improvement_proposal_default = {
|
|
|
3592
3630
|
"kind": "agent-candidate-profile-activation",
|
|
3593
3631
|
"profilePlan": {
|
|
3594
3632
|
"artifact": {
|
|
3595
|
-
"byteLength":
|
|
3596
|
-
"content": "
|
|
3633
|
+
"byteLength": 532,
|
|
3634
|
+
"content": "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",
|
|
3597
3635
|
"encoding": "base64",
|
|
3598
|
-
"sha256": "sha256:
|
|
3636
|
+
"sha256": "sha256:9452cb632b127aae1dc94fa8396f7792f14247a8dd24bd97f6aec998e10e2626"
|
|
3599
3637
|
},
|
|
3600
|
-
"digest": "sha256:
|
|
3638
|
+
"digest": "sha256:9452cb632b127aae1dc94fa8396f7792f14247a8dd24bd97f6aec998e10e2626",
|
|
3601
3639
|
"kind": "agent-profile-workspace-plan",
|
|
3602
3640
|
"material": {
|
|
3603
3641
|
"env": {},
|
|
3604
3642
|
"files": [{
|
|
3643
|
+
"contentSha256": "sha256:7cd9c117d3377708275faa6a564af952f3f0c71352ae55833bf416377962d5b0",
|
|
3644
|
+
"mode": 420,
|
|
3645
|
+
"relPath": ".codex/system-prompt.md"
|
|
3646
|
+
}, {
|
|
3605
3647
|
"contentSha256": "sha256:39adbb9b7f571f0fd586e7795023d25909adbe9586bb21ee134d3d7248bf1599",
|
|
3606
3648
|
"mode": 420,
|
|
3607
3649
|
"relPath": "AGENTS.md"
|
|
@@ -3611,7 +3653,7 @@ var agent_improvement_proposal_default = {
|
|
|
3611
3653
|
"value": "-c"
|
|
3612
3654
|
}, {
|
|
3613
3655
|
"kind": "public",
|
|
3614
|
-
"value": "
|
|
3656
|
+
"value": "model_instructions_file=.codex/system-prompt.md"
|
|
3615
3657
|
}],
|
|
3616
3658
|
"harness": "codex",
|
|
3617
3659
|
"sourceProfileDigest": "sha256:8b1e2f0f7b832958381c29baebe4a961b151af97a0ce9d611a15ad0741aeaa75",
|
|
@@ -3633,12 +3675,12 @@ var agent_improvement_proposal_default = {
|
|
|
3633
3675
|
"byteLength": 1149,
|
|
3634
3676
|
"locator": {
|
|
3635
3677
|
"bucket": "candidate-test-artifacts",
|
|
3636
|
-
"key": "benchmark-result/
|
|
3678
|
+
"key": "benchmark-result/d4825d27317a6aee0c214dc4dfcd58f46586f263cc948aa834c45a1845929c29",
|
|
3637
3679
|
"kind": "s3"
|
|
3638
3680
|
},
|
|
3639
|
-
"sha256": "sha256:
|
|
3681
|
+
"sha256": "sha256:d4825d27317a6aee0c214dc4dfcd58f46586f263cc948aa834c45a1845929c29"
|
|
3640
3682
|
},
|
|
3641
|
-
"digest": "sha256:
|
|
3683
|
+
"digest": "sha256:d4825d27317a6aee0c214dc4dfcd58f46586f263cc948aa834c45a1845929c29",
|
|
3642
3684
|
"kind": "agent-candidate-benchmark-result",
|
|
3643
3685
|
"material": {
|
|
3644
3686
|
"dimensions": [{
|
|
@@ -3654,7 +3696,7 @@ var agent_improvement_proposal_default = {
|
|
|
3654
3696
|
},
|
|
3655
3697
|
"sha256": "sha256:9b9b3a1471309177261cfe65ea9c298e0dd372e4b5d087f8d35d7b732485373d"
|
|
3656
3698
|
},
|
|
3657
|
-
"executionPlanDigest": "sha256:
|
|
3699
|
+
"executionPlanDigest": "sha256:e247492e04c77a23efb2c86c1f54012bbe29d0e64021b19dbe85df01255fa1f4",
|
|
3658
3700
|
"grader": {
|
|
3659
3701
|
"artifact": {
|
|
3660
3702
|
"byteLength": 25,
|
|
@@ -3688,41 +3730,41 @@ var agent_improvement_proposal_default = {
|
|
|
3688
3730
|
"kind": "agent-candidate-benchmark-result-material",
|
|
3689
3731
|
"passed": true,
|
|
3690
3732
|
"score": 1,
|
|
3691
|
-
"taskOutcomeDigest": "sha256:
|
|
3733
|
+
"taskOutcomeDigest": "sha256:c6d7c9ba6b62218332948c3f915ae9977f32f1cde32d15eaf544846faaa4c71b"
|
|
3692
3734
|
}
|
|
3693
3735
|
},
|
|
3694
3736
|
"bundleDigest": "sha256:60fcbb1c728194bd51d7d19cb732d1c3f1881dce7e0a6266b41c8b98cfd65693",
|
|
3695
|
-
"digest": "sha256:
|
|
3737
|
+
"digest": "sha256:233b1e655eab6aa77895920b85a8d07d78242cf6d68010e56acabdbfb7761d49",
|
|
3696
3738
|
"digestAlgorithm": "rfc8785-sha256",
|
|
3697
|
-
"executionPlanDigest": "sha256:
|
|
3739
|
+
"executionPlanDigest": "sha256:e247492e04c77a23efb2c86c1f54012bbe29d0e64021b19dbe85df01255fa1f4",
|
|
3698
3740
|
"executorCapture": {
|
|
3699
3741
|
"byteLength": 477,
|
|
3700
3742
|
"locator": {
|
|
3701
3743
|
"bucket": "candidate-test-artifacts",
|
|
3702
|
-
"key": "executor-capture/
|
|
3744
|
+
"key": "executor-capture/7b58c06fb56f5e51d64b5f790092e4178e5381b9e41b2372178d5dcc3bca4ea2",
|
|
3703
3745
|
"kind": "s3"
|
|
3704
3746
|
},
|
|
3705
|
-
"sha256": "sha256:
|
|
3747
|
+
"sha256": "sha256:7b58c06fb56f5e51d64b5f790092e4178e5381b9e41b2372178d5dcc3bca4ea2"
|
|
3706
3748
|
},
|
|
3707
3749
|
"kind": "agent-candidate-run",
|
|
3708
|
-
"materializationReceiptDigest": "sha256:
|
|
3750
|
+
"materializationReceiptDigest": "sha256:2bc6a75f77ff4de870e326a766e8d5a7bb5c1f6fb271a6041bb426eb9d846c29",
|
|
3709
3751
|
"memory": { "mode": "disabled" },
|
|
3710
3752
|
"modelSettlement": {
|
|
3711
3753
|
"artifact": {
|
|
3712
3754
|
"byteLength": 643,
|
|
3713
3755
|
"locator": {
|
|
3714
3756
|
"bucket": "candidate-test-artifacts",
|
|
3715
|
-
"key": "model-settlement/
|
|
3757
|
+
"key": "model-settlement/e44ea0408fdb9f9ee7af5e624b92e9c9d2c617948b66d12d5f3acb552c63e015",
|
|
3716
3758
|
"kind": "s3"
|
|
3717
3759
|
},
|
|
3718
|
-
"sha256": "sha256:
|
|
3760
|
+
"sha256": "sha256:e44ea0408fdb9f9ee7af5e624b92e9c9d2c617948b66d12d5f3acb552c63e015"
|
|
3719
3761
|
},
|
|
3720
|
-
"digest": "sha256:
|
|
3762
|
+
"digest": "sha256:e44ea0408fdb9f9ee7af5e624b92e9c9d2c617948b66d12d5f3acb552c63e015",
|
|
3721
3763
|
"kind": "agent-candidate-model-settlement",
|
|
3722
3764
|
"material": {
|
|
3723
3765
|
"calls": [],
|
|
3724
3766
|
"closed": true,
|
|
3725
|
-
"executionPlanDigest": "sha256:
|
|
3767
|
+
"executionPlanDigest": "sha256:e247492e04c77a23efb2c86c1f54012bbe29d0e64021b19dbe85df01255fa1f4",
|
|
3726
3768
|
"grantDigest": "sha256:cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc",
|
|
3727
3769
|
"kind": "agent-candidate-model-settlement-material",
|
|
3728
3770
|
"preparationId": "candidate-preparation.nBqsDbKYlsiY4Tto5XaxkR3RCTop9PA9_kX-kOSzR8U",
|
|
@@ -3751,15 +3793,15 @@ var agent_improvement_proposal_default = {
|
|
|
3751
3793
|
"byteLength": 478,
|
|
3752
3794
|
"locator": {
|
|
3753
3795
|
"bucket": "candidate-test-artifacts",
|
|
3754
|
-
"key": "task-outcome/
|
|
3796
|
+
"key": "task-outcome/c6d7c9ba6b62218332948c3f915ae9977f32f1cde32d15eaf544846faaa4c71b",
|
|
3755
3797
|
"kind": "s3"
|
|
3756
3798
|
},
|
|
3757
|
-
"sha256": "sha256:
|
|
3799
|
+
"sha256": "sha256:c6d7c9ba6b62218332948c3f915ae9977f32f1cde32d15eaf544846faaa4c71b"
|
|
3758
3800
|
},
|
|
3759
|
-
"digest": "sha256:
|
|
3801
|
+
"digest": "sha256:c6d7c9ba6b62218332948c3f915ae9977f32f1cde32d15eaf544846faaa4c71b",
|
|
3760
3802
|
"kind": "agent-candidate-task-outcome",
|
|
3761
3803
|
"material": {
|
|
3762
|
-
"executionPlanDigest": "sha256:
|
|
3804
|
+
"executionPlanDigest": "sha256:e247492e04c77a23efb2c86c1f54012bbe29d0e64021b19dbe85df01255fa1f4",
|
|
3763
3805
|
"kind": "agent-candidate-task-outcome-material",
|
|
3764
3806
|
"outcome": {
|
|
3765
3807
|
"artifact": {
|
|
@@ -3793,10 +3835,10 @@ var agent_improvement_proposal_default = {
|
|
|
3793
3835
|
"byteLength": 750,
|
|
3794
3836
|
"locator": {
|
|
3795
3837
|
"bucket": "candidate-test-artifacts",
|
|
3796
|
-
"key": "trace/
|
|
3838
|
+
"key": "trace/b813e6baf9c600ffaa6e9cee6b3670f2bb4e9da702d8d863fb30d79ef59fc2aa",
|
|
3797
3839
|
"kind": "s3"
|
|
3798
3840
|
},
|
|
3799
|
-
"sha256": "sha256:
|
|
3841
|
+
"sha256": "sha256:b813e6baf9c600ffaa6e9cee6b3670f2bb4e9da702d8d863fb30d79ef59fc2aa"
|
|
3800
3842
|
},
|
|
3801
3843
|
"eventCount": 1,
|
|
3802
3844
|
"modelCallCount": 0
|
|
@@ -4153,7 +4195,7 @@ var agent_improvement_proposal_default = {
|
|
|
4153
4195
|
}
|
|
4154
4196
|
},
|
|
4155
4197
|
"candidate": {
|
|
4156
|
-
"digest": "sha256:
|
|
4198
|
+
"digest": "sha256:f7846c0b86e6252031b94d7ef394dd1f41757e61a83328c75375d76004c4d45a",
|
|
4157
4199
|
"kind": "agent-candidate-execution-evidence",
|
|
4158
4200
|
"materializationReceipt": {
|
|
4159
4201
|
"benchmark": {
|
|
@@ -4188,16 +4230,16 @@ var agent_improvement_proposal_default = {
|
|
|
4188
4230
|
},
|
|
4189
4231
|
"source": "evaluator-task-container"
|
|
4190
4232
|
},
|
|
4191
|
-
"digest": "sha256:
|
|
4233
|
+
"digest": "sha256:d79626085ba95d13c60ac422fed11e2980593e8e83fef73dc59451336683b69f",
|
|
4192
4234
|
"digestAlgorithm": "rfc8785-sha256",
|
|
4193
4235
|
"executionPlan": {
|
|
4194
4236
|
"artifact": {
|
|
4195
|
-
"byteLength":
|
|
4196
|
-
"content": "
|
|
4237
|
+
"byteLength": 2164,
|
|
4238
|
+
"content": "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",
|
|
4197
4239
|
"encoding": "base64",
|
|
4198
|
-
"sha256": "sha256:
|
|
4240
|
+
"sha256": "sha256:cc4e8f626257eb614658d63483c00c094035cfebffc26aa22f88984eba8de512"
|
|
4199
4241
|
},
|
|
4200
|
-
"digest": "sha256:
|
|
4242
|
+
"digest": "sha256:cc4e8f626257eb614658d63483c00c094035cfebffc26aa22f88984eba8de512",
|
|
4201
4243
|
"kind": "agent-candidate-execution-plan",
|
|
4202
4244
|
"material": {
|
|
4203
4245
|
"codeKind": "disabled",
|
|
@@ -4222,7 +4264,7 @@ var agent_improvement_proposal_default = {
|
|
|
4222
4264
|
"value": "-c"
|
|
4223
4265
|
}, {
|
|
4224
4266
|
"kind": "public",
|
|
4225
|
-
"value": "
|
|
4267
|
+
"value": "model_instructions_file=.codex/system-prompt.md"
|
|
4226
4268
|
}],
|
|
4227
4269
|
"cwd": {
|
|
4228
4270
|
"path": ".",
|
|
@@ -4267,8 +4309,8 @@ var agent_improvement_proposal_default = {
|
|
|
4267
4309
|
},
|
|
4268
4310
|
"network": { "mode": "disabled" },
|
|
4269
4311
|
"profile": {
|
|
4270
|
-
"mountPaths": ["AGENTS.md"],
|
|
4271
|
-
"planDigest": "sha256:
|
|
4312
|
+
"mountPaths": [".codex/system-prompt.md", "AGENTS.md"],
|
|
4313
|
+
"planDigest": "sha256:9452cb632b127aae1dc94fa8396f7792f14247a8dd24bd97f6aec998e10e2626",
|
|
4272
4314
|
"targetWorkspace": "task"
|
|
4273
4315
|
},
|
|
4274
4316
|
"runCell": {
|
|
@@ -4291,8 +4333,12 @@ var agent_improvement_proposal_default = {
|
|
|
4291
4333
|
"harnessVersion": "1.2.3",
|
|
4292
4334
|
"kind": "agent-candidate-materialization",
|
|
4293
4335
|
"profileActivation": {
|
|
4294
|
-
"digest": "sha256:
|
|
4336
|
+
"digest": "sha256:41c3e5ee69665488a2aefdf82b746696422bfbcd528fdedc110bc8e267cb0284",
|
|
4295
4337
|
"files": [{
|
|
4338
|
+
"content": "Return the exact measured answer.",
|
|
4339
|
+
"mode": 420,
|
|
4340
|
+
"path": ".codex/system-prompt.md"
|
|
4341
|
+
}, {
|
|
4296
4342
|
"content": "Inspect the repository, implement the fix, and run tests.\n",
|
|
4297
4343
|
"mode": 420,
|
|
4298
4344
|
"path": "AGENTS.md"
|
|
@@ -4300,16 +4346,20 @@ var agent_improvement_proposal_default = {
|
|
|
4300
4346
|
"kind": "agent-candidate-profile-activation",
|
|
4301
4347
|
"profilePlan": {
|
|
4302
4348
|
"artifact": {
|
|
4303
|
-
"byteLength":
|
|
4304
|
-
"content": "
|
|
4349
|
+
"byteLength": 532,
|
|
4350
|
+
"content": "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",
|
|
4305
4351
|
"encoding": "base64",
|
|
4306
|
-
"sha256": "sha256:
|
|
4352
|
+
"sha256": "sha256:9452cb632b127aae1dc94fa8396f7792f14247a8dd24bd97f6aec998e10e2626"
|
|
4307
4353
|
},
|
|
4308
|
-
"digest": "sha256:
|
|
4354
|
+
"digest": "sha256:9452cb632b127aae1dc94fa8396f7792f14247a8dd24bd97f6aec998e10e2626",
|
|
4309
4355
|
"kind": "agent-profile-workspace-plan",
|
|
4310
4356
|
"material": {
|
|
4311
4357
|
"env": {},
|
|
4312
4358
|
"files": [{
|
|
4359
|
+
"contentSha256": "sha256:7cd9c117d3377708275faa6a564af952f3f0c71352ae55833bf416377962d5b0",
|
|
4360
|
+
"mode": 420,
|
|
4361
|
+
"relPath": ".codex/system-prompt.md"
|
|
4362
|
+
}, {
|
|
4313
4363
|
"contentSha256": "sha256:39adbb9b7f571f0fd586e7795023d25909adbe9586bb21ee134d3d7248bf1599",
|
|
4314
4364
|
"mode": 420,
|
|
4315
4365
|
"relPath": "AGENTS.md"
|
|
@@ -4319,7 +4369,7 @@ var agent_improvement_proposal_default = {
|
|
|
4319
4369
|
"value": "-c"
|
|
4320
4370
|
}, {
|
|
4321
4371
|
"kind": "public",
|
|
4322
|
-
"value": "
|
|
4372
|
+
"value": "model_instructions_file=.codex/system-prompt.md"
|
|
4323
4373
|
}],
|
|
4324
4374
|
"harness": "codex",
|
|
4325
4375
|
"sourceProfileDigest": "sha256:8b1e2f0f7b832958381c29baebe4a961b151af97a0ce9d611a15ad0741aeaa75",
|
|
@@ -4341,12 +4391,12 @@ var agent_improvement_proposal_default = {
|
|
|
4341
4391
|
"byteLength": 1149,
|
|
4342
4392
|
"locator": {
|
|
4343
4393
|
"bucket": "candidate-test-artifacts",
|
|
4344
|
-
"key": "benchmark-result/
|
|
4394
|
+
"key": "benchmark-result/f913c9693e459be416d52cfb5c1ff7faa8986c2874f1bdf6a5abfa94db39d85a",
|
|
4345
4395
|
"kind": "s3"
|
|
4346
4396
|
},
|
|
4347
|
-
"sha256": "sha256:
|
|
4397
|
+
"sha256": "sha256:f913c9693e459be416d52cfb5c1ff7faa8986c2874f1bdf6a5abfa94db39d85a"
|
|
4348
4398
|
},
|
|
4349
|
-
"digest": "sha256:
|
|
4399
|
+
"digest": "sha256:f913c9693e459be416d52cfb5c1ff7faa8986c2874f1bdf6a5abfa94db39d85a",
|
|
4350
4400
|
"kind": "agent-candidate-benchmark-result",
|
|
4351
4401
|
"material": {
|
|
4352
4402
|
"dimensions": [{
|
|
@@ -4362,7 +4412,7 @@ var agent_improvement_proposal_default = {
|
|
|
4362
4412
|
},
|
|
4363
4413
|
"sha256": "sha256:9b9b3a1471309177261cfe65ea9c298e0dd372e4b5d087f8d35d7b732485373d"
|
|
4364
4414
|
},
|
|
4365
|
-
"executionPlanDigest": "sha256:
|
|
4415
|
+
"executionPlanDigest": "sha256:cc4e8f626257eb614658d63483c00c094035cfebffc26aa22f88984eba8de512",
|
|
4366
4416
|
"grader": {
|
|
4367
4417
|
"artifact": {
|
|
4368
4418
|
"byteLength": 25,
|
|
@@ -4396,41 +4446,41 @@ var agent_improvement_proposal_default = {
|
|
|
4396
4446
|
"kind": "agent-candidate-benchmark-result-material",
|
|
4397
4447
|
"passed": true,
|
|
4398
4448
|
"score": 1,
|
|
4399
|
-
"taskOutcomeDigest": "sha256:
|
|
4449
|
+
"taskOutcomeDigest": "sha256:89e6df4a45dd6b99499b8c48f9f783dd6359ab1ebfa38c71f34a6ee6566b3223"
|
|
4400
4450
|
}
|
|
4401
4451
|
},
|
|
4402
4452
|
"bundleDigest": "sha256:60fcbb1c728194bd51d7d19cb732d1c3f1881dce7e0a6266b41c8b98cfd65693",
|
|
4403
|
-
"digest": "sha256:
|
|
4453
|
+
"digest": "sha256:1f989657ce29a8fd82a5fcfac7fc57310b985f491f55cad4ce70f3c21df0905c",
|
|
4404
4454
|
"digestAlgorithm": "rfc8785-sha256",
|
|
4405
|
-
"executionPlanDigest": "sha256:
|
|
4455
|
+
"executionPlanDigest": "sha256:cc4e8f626257eb614658d63483c00c094035cfebffc26aa22f88984eba8de512",
|
|
4406
4456
|
"executorCapture": {
|
|
4407
4457
|
"byteLength": 477,
|
|
4408
4458
|
"locator": {
|
|
4409
4459
|
"bucket": "candidate-test-artifacts",
|
|
4410
|
-
"key": "executor-capture/
|
|
4460
|
+
"key": "executor-capture/321fe808b59d3db1415c9144041dd65cff5f611ad98b4d50c6d815f8c958797d",
|
|
4411
4461
|
"kind": "s3"
|
|
4412
4462
|
},
|
|
4413
|
-
"sha256": "sha256:
|
|
4463
|
+
"sha256": "sha256:321fe808b59d3db1415c9144041dd65cff5f611ad98b4d50c6d815f8c958797d"
|
|
4414
4464
|
},
|
|
4415
4465
|
"kind": "agent-candidate-run",
|
|
4416
|
-
"materializationReceiptDigest": "sha256:
|
|
4466
|
+
"materializationReceiptDigest": "sha256:d79626085ba95d13c60ac422fed11e2980593e8e83fef73dc59451336683b69f",
|
|
4417
4467
|
"memory": { "mode": "disabled" },
|
|
4418
4468
|
"modelSettlement": {
|
|
4419
4469
|
"artifact": {
|
|
4420
4470
|
"byteLength": 643,
|
|
4421
4471
|
"locator": {
|
|
4422
4472
|
"bucket": "candidate-test-artifacts",
|
|
4423
|
-
"key": "model-settlement/
|
|
4473
|
+
"key": "model-settlement/bec47b973ec51b672cd426042c605e696232640101a58d70af4406fe93682d18",
|
|
4424
4474
|
"kind": "s3"
|
|
4425
4475
|
},
|
|
4426
|
-
"sha256": "sha256:
|
|
4476
|
+
"sha256": "sha256:bec47b973ec51b672cd426042c605e696232640101a58d70af4406fe93682d18"
|
|
4427
4477
|
},
|
|
4428
|
-
"digest": "sha256:
|
|
4478
|
+
"digest": "sha256:bec47b973ec51b672cd426042c605e696232640101a58d70af4406fe93682d18",
|
|
4429
4479
|
"kind": "agent-candidate-model-settlement",
|
|
4430
4480
|
"material": {
|
|
4431
4481
|
"calls": [],
|
|
4432
4482
|
"closed": true,
|
|
4433
|
-
"executionPlanDigest": "sha256:
|
|
4483
|
+
"executionPlanDigest": "sha256:cc4e8f626257eb614658d63483c00c094035cfebffc26aa22f88984eba8de512",
|
|
4434
4484
|
"grantDigest": "sha256:cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc",
|
|
4435
4485
|
"kind": "agent-candidate-model-settlement-material",
|
|
4436
4486
|
"preparationId": "candidate-preparation.nBqsDbKYlsiY4Tto5XaxkR3RCTop9PA9_kX-kOSzR8U",
|
|
@@ -4459,15 +4509,15 @@ var agent_improvement_proposal_default = {
|
|
|
4459
4509
|
"byteLength": 478,
|
|
4460
4510
|
"locator": {
|
|
4461
4511
|
"bucket": "candidate-test-artifacts",
|
|
4462
|
-
"key": "task-outcome/
|
|
4512
|
+
"key": "task-outcome/89e6df4a45dd6b99499b8c48f9f783dd6359ab1ebfa38c71f34a6ee6566b3223",
|
|
4463
4513
|
"kind": "s3"
|
|
4464
4514
|
},
|
|
4465
|
-
"sha256": "sha256:
|
|
4515
|
+
"sha256": "sha256:89e6df4a45dd6b99499b8c48f9f783dd6359ab1ebfa38c71f34a6ee6566b3223"
|
|
4466
4516
|
},
|
|
4467
|
-
"digest": "sha256:
|
|
4517
|
+
"digest": "sha256:89e6df4a45dd6b99499b8c48f9f783dd6359ab1ebfa38c71f34a6ee6566b3223",
|
|
4468
4518
|
"kind": "agent-candidate-task-outcome",
|
|
4469
4519
|
"material": {
|
|
4470
|
-
"executionPlanDigest": "sha256:
|
|
4520
|
+
"executionPlanDigest": "sha256:cc4e8f626257eb614658d63483c00c094035cfebffc26aa22f88984eba8de512",
|
|
4471
4521
|
"kind": "agent-candidate-task-outcome-material",
|
|
4472
4522
|
"outcome": {
|
|
4473
4523
|
"artifact": {
|
|
@@ -4501,10 +4551,10 @@ var agent_improvement_proposal_default = {
|
|
|
4501
4551
|
"byteLength": 750,
|
|
4502
4552
|
"locator": {
|
|
4503
4553
|
"bucket": "candidate-test-artifacts",
|
|
4504
|
-
"key": "trace/
|
|
4554
|
+
"key": "trace/4222f404ca11b92771d3fca4949d5ad85e8fa11a7662cf7e3683a8135afb977c",
|
|
4505
4555
|
"kind": "s3"
|
|
4506
4556
|
},
|
|
4507
|
-
"sha256": "sha256:
|
|
4557
|
+
"sha256": "sha256:4222f404ca11b92771d3fca4949d5ad85e8fa11a7662cf7e3683a8135afb977c"
|
|
4508
4558
|
},
|
|
4509
4559
|
"eventCount": 1,
|
|
4510
4560
|
"modelCallCount": 0
|
|
@@ -4515,7 +4565,7 @@ var agent_improvement_proposal_default = {
|
|
|
4515
4565
|
],
|
|
4516
4566
|
"metadata": {
|
|
4517
4567
|
"fixture": "agent-improvement-proposal",
|
|
4518
|
-
"runtimeVersion": "0.
|
|
4568
|
+
"runtimeVersion": "0.131.0"
|
|
4519
4569
|
},
|
|
4520
4570
|
"objectives": [
|
|
4521
4571
|
{
|
|
@@ -4626,8 +4676,8 @@ var agent_improvement_proposal_default = {
|
|
|
4626
4676
|
"baselineContentHash": "sha256:5c21ee53e513fc604cb09754e21c392b24a424da0ef37dbf8f1ee4a8a0b08f09",
|
|
4627
4677
|
"candidateContentHash": "sha256:60fcbb1c728194bd51d7d19cb732d1c3f1881dce7e0a6266b41c8b98cfd65693",
|
|
4628
4678
|
"kind": "agent-eval-loop",
|
|
4629
|
-
"recordDigest": "sha256:
|
|
4630
|
-
"runId": "agent-runtime-0.
|
|
4679
|
+
"recordDigest": "sha256:1be3dab07f5e01c97fb437eec6772825ac7b32a78e875419ecdceb32af1a198d",
|
|
4680
|
+
"runId": "agent-runtime-0.131.0-proposal-fixture",
|
|
4631
4681
|
"schema": "agent-candidate-experiment"
|
|
4632
4682
|
}
|
|
4633
4683
|
},
|
|
@@ -4650,13 +4700,13 @@ var agent_improvement_proposal_default = {
|
|
|
4650
4700
|
}],
|
|
4651
4701
|
kind: "agent-improvement-proposal",
|
|
4652
4702
|
proposedAt: "2026-07-10T01:00:00.000Z",
|
|
4653
|
-
runId: "agent-runtime-0.
|
|
4703
|
+
runId: "agent-runtime-0.131.0-proposal-fixture"
|
|
4654
4704
|
};
|
|
4655
4705
|
//#endregion
|
|
4656
4706
|
//#region src/testing/fixtures/agent-profile-improvement-proposal.json
|
|
4657
4707
|
var agent_profile_improvement_proposal_default = {
|
|
4658
4708
|
changedSurfaces: ["prompt", "skills"],
|
|
4659
|
-
digest: "sha256:
|
|
4709
|
+
digest: "sha256:97ab9bd1ba910859a42954c9b0dc85a377db2dd454516ac4e76bd118c15e69ca",
|
|
4660
4710
|
evaluation: {
|
|
4661
4711
|
"decision": {
|
|
4662
4712
|
"contributingChecks": [
|
|
@@ -6290,7 +6340,7 @@ var agent_profile_improvement_proposal_default = {
|
|
|
6290
6340
|
],
|
|
6291
6341
|
"metadata": {
|
|
6292
6342
|
"fixture": "agent-profile-improvement-proposal",
|
|
6293
|
-
"runtimeVersion": "0.
|
|
6343
|
+
"runtimeVersion": "0.131.0"
|
|
6294
6344
|
},
|
|
6295
6345
|
"objectives": [
|
|
6296
6346
|
{
|
|
@@ -6401,7 +6451,7 @@ var agent_profile_improvement_proposal_default = {
|
|
|
6401
6451
|
"baselineContentHash": "sha256:21c495a37c418c10bde64fbaa188beddeed31f1f051ea60a6a6582a9ee0db704",
|
|
6402
6452
|
"candidateContentHash": "sha256:103f77bc8481601eef1ad5fe6ba84a40dffabc3a44f421f8c8559121edab84e9",
|
|
6403
6453
|
"kind": "agent-eval-loop",
|
|
6404
|
-
"recordDigest": "sha256:
|
|
6454
|
+
"recordDigest": "sha256:30fa9b74d5682ada3d9c40eb90298cda4c3e17c0a45aaf8f821fd422e3bafbf8",
|
|
6405
6455
|
"runId": "profile-improvement-1",
|
|
6406
6456
|
"schema": "agent-profile-improvement-experiment"
|
|
6407
6457
|
}
|
|
@@ -6496,6 +6546,6 @@ function profileStateDigest(profile, recommendedSize) {
|
|
|
6496
6546
|
});
|
|
6497
6547
|
}
|
|
6498
6548
|
//#endregion
|
|
6499
|
-
export { loadAgentImprovementProposalFixture, loadAgentProfileImprovementFixture };
|
|
6549
|
+
export { driverAgent, loadAgentImprovementProposalFixture, loadAgentProfileImprovementFixture, runGraphWithTestBrain, superviseWithTestBrain, supervisorAgentWithTestBrain };
|
|
6500
6550
|
|
|
6501
6551
|
//# sourceMappingURL=testing.js.map
|