{ "schema_version": "alliance-calibration-snapshot.v1", "run_id": "g1-candidate-prompt-1.1.0-2026-08-06", "provider": "claude_cli", "model": "mixed", "prompt_bundle_version": "1.1.0", "benchmark_schema": "vignette.outcome_alliance_benchmark.v1", "generated_at": "2026-08-06T12:32:05.381153Z", "predictions": [ { "case_id": "oas-g0-001", "perspective": "independent_observer", "dimension": "goal", "value": null, "confidence": null, "evidence_turn_indices": [], "status": "error", "error_code": "agent_validationerror", "provider": "claude_cli", "model": "gateway-default", "prompt_bundle_version": "1.1.0", "model_run_id": "84cacb2a-cf27-4317-b2e7-f0b4ed6525d1", "prompt_bundle_hash": "e5ae373bbb90596b0086fcb73985d37f65a920e9b407d145efe89ef76f58034f", "input_evidence_hash": "73c73daff30eb59a0566592162954833c32db3fb8ef971ffeb593003e8b2ec88" }, { "case_id": "oas-g0-001", "perspective": "independent_observer", "dimension": "task", "value": null, "confidence": null, "evidence_turn_indices": [], "status": "error", "error_code": "agent_validationerror", "provider": "claude_cli", "model": "gateway-default", "prompt_bundle_version": "1.1.0", "model_run_id": "84cacb2a-cf27-4317-b2e7-f0b4ed6525d1", "prompt_bundle_hash": "e5ae373bbb90596b0086fcb73985d37f65a920e9b407d145efe89ef76f58034f", "input_evidence_hash": "73c73daff30eb59a0566592162954833c32db3fb8ef971ffeb593003e8b2ec88" }, { "case_id": "oas-g0-001", "perspective": "independent_observer", "dimension": "bond", "value": null, "confidence": null, "evidence_turn_indices": [], "status": "error", "error_code": "agent_validationerror", "provider": "claude_cli", "model": "gateway-default", "prompt_bundle_version": "1.1.0", "model_run_id": "84cacb2a-cf27-4317-b2e7-f0b4ed6525d1", "prompt_bundle_hash": "e5ae373bbb90596b0086fcb73985d37f65a920e9b407d145efe89ef76f58034f", "input_evidence_hash": "73c73daff30eb59a0566592162954833c32db3fb8ef971ffeb593003e8b2ec88" }, { "case_id": "oas-g0-002", "perspective": "independent_observer", "dimension": "goal", "value": 0.35, "confidence": 0.2, "evidence_turn_indices": [ 0, 1, 2 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.1.0", "model_run_id": "e171fb71-ffcc-4dbf-b299-9c6a7c121f85", "prompt_bundle_hash": "e5ae373bbb90596b0086fcb73985d37f65a920e9b407d145efe89ef76f58034f", "input_evidence_hash": "27c4b079f44e906240c6545e9ef02411d276c7fbcb8de23b9ea1ca79106adbbb" }, { "case_id": "oas-g0-002", "perspective": "independent_observer", "dimension": "task", "value": 0.15, "confidence": 0.75, "evidence_turn_indices": [ 0, 1, 2 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.1.0", "model_run_id": "e171fb71-ffcc-4dbf-b299-9c6a7c121f85", "prompt_bundle_hash": "e5ae373bbb90596b0086fcb73985d37f65a920e9b407d145efe89ef76f58034f", "input_evidence_hash": "27c4b079f44e906240c6545e9ef02411d276c7fbcb8de23b9ea1ca79106adbbb" }, { "case_id": "oas-g0-002", "perspective": "independent_observer", "dimension": "bond", "value": 0.25, "confidence": 0.3, "evidence_turn_indices": [ 1, 2 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.1.0", "model_run_id": "e171fb71-ffcc-4dbf-b299-9c6a7c121f85", "prompt_bundle_hash": "e5ae373bbb90596b0086fcb73985d37f65a920e9b407d145efe89ef76f58034f", "input_evidence_hash": "27c4b079f44e906240c6545e9ef02411d276c7fbcb8de23b9ea1ca79106adbbb" }, { "case_id": "oas-g0-003", "perspective": "independent_observer", "dimension": "goal", "value": null, "confidence": null, "evidence_turn_indices": [], "status": "error", "error_code": "agent_validationerror", "provider": "claude_cli", "model": "gateway-default", "prompt_bundle_version": "1.1.0", "model_run_id": "e6102185-889e-42a1-8521-456f8e0b971b", "prompt_bundle_hash": "e5ae373bbb90596b0086fcb73985d37f65a920e9b407d145efe89ef76f58034f", "input_evidence_hash": "eef29a4ff7245ce97c856ec22c6ca74c19ab33c989259f6f0c86d6c60fa04c56" }, { "case_id": "oas-g0-003", "perspective": "independent_observer", "dimension": "task", "value": null, "confidence": null, "evidence_turn_indices": [], "status": "error", "error_code": "agent_validationerror", "provider": "claude_cli", "model": "gateway-default", "prompt_bundle_version": "1.1.0", "model_run_id": "e6102185-889e-42a1-8521-456f8e0b971b", "prompt_bundle_hash": "e5ae373bbb90596b0086fcb73985d37f65a920e9b407d145efe89ef76f58034f", "input_evidence_hash": "eef29a4ff7245ce97c856ec22c6ca74c19ab33c989259f6f0c86d6c60fa04c56" }, { "case_id": "oas-g0-003", "perspective": "independent_observer", "dimension": "bond", "value": null, "confidence": null, "evidence_turn_indices": [], "status": "error", "error_code": "agent_validationerror", "provider": "claude_cli", "model": "gateway-default", "prompt_bundle_version": "1.1.0", "model_run_id": "e6102185-889e-42a1-8521-456f8e0b971b", "prompt_bundle_hash": "e5ae373bbb90596b0086fcb73985d37f65a920e9b407d145efe89ef76f58034f", "input_evidence_hash": "eef29a4ff7245ce97c856ec22c6ca74c19ab33c989259f6f0c86d6c60fa04c56" }, { "case_id": "oas-g0-005", "perspective": "client_agent_report", "dimension": "goal", "value": null, "confidence": null, "evidence_turn_indices": [], "status": "error", "error_code": "agent_validationerror", "provider": "claude_cli", "model": "gateway-default", "prompt_bundle_version": "1.1.0", "model_run_id": "319dc063-5628-499c-bd57-f03ca62d8361", "prompt_bundle_hash": "68b36d7b8e987c40714ce6ad7f2a17010be4a1ebe0822fd26b09fe4742288073", "input_evidence_hash": "df68b447c1d21ea038d185894ea7fcbfacdb9757aee11df5e996dfc7efb600fd" }, { "case_id": "oas-g0-005", "perspective": "client_agent_report", "dimension": "task", "value": null, "confidence": null, "evidence_turn_indices": [], "status": "error", "error_code": "agent_validationerror", "provider": "claude_cli", "model": "gateway-default", "prompt_bundle_version": "1.1.0", "model_run_id": "319dc063-5628-499c-bd57-f03ca62d8361", "prompt_bundle_hash": "68b36d7b8e987c40714ce6ad7f2a17010be4a1ebe0822fd26b09fe4742288073", "input_evidence_hash": "df68b447c1d21ea038d185894ea7fcbfacdb9757aee11df5e996dfc7efb600fd" }, { "case_id": "oas-g0-005", "perspective": "client_agent_report", "dimension": "bond", "value": null, "confidence": null, "evidence_turn_indices": [], "status": "error", "error_code": "agent_validationerror", "provider": "claude_cli", "model": "gateway-default", "prompt_bundle_version": "1.1.0", "model_run_id": "319dc063-5628-499c-bd57-f03ca62d8361", "prompt_bundle_hash": "68b36d7b8e987c40714ce6ad7f2a17010be4a1ebe0822fd26b09fe4742288073", "input_evidence_hash": "df68b447c1d21ea038d185894ea7fcbfacdb9757aee11df5e996dfc7efb600fd" }, { "case_id": "oas-g0-006", "perspective": "client_agent_report", "dimension": "goal", "value": 0.72, "confidence": 0.55, "evidence_turn_indices": [ 0, 2, 3 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.1.0", "model_run_id": "67f7e65d-a914-4f87-b80d-f42ea0b392df", "prompt_bundle_hash": "68b36d7b8e987c40714ce6ad7f2a17010be4a1ebe0822fd26b09fe4742288073", "input_evidence_hash": "425a9d833425dc2547f41667baedfd8f7f5039f220b73a91f48334f7ba77c854" }, { "case_id": "oas-g0-006", "perspective": "client_agent_report", "dimension": "task", "value": 0.78, "confidence": 0.6, "evidence_turn_indices": [ 2, 3 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.1.0", "model_run_id": "67f7e65d-a914-4f87-b80d-f42ea0b392df", "prompt_bundle_hash": "68b36d7b8e987c40714ce6ad7f2a17010be4a1ebe0822fd26b09fe4742288073", "input_evidence_hash": "425a9d833425dc2547f41667baedfd8f7f5039f220b73a91f48334f7ba77c854" }, { "case_id": "oas-g0-006", "perspective": "client_agent_report", "dimension": "bond", "value": 0.52, "confidence": 0.35, "evidence_turn_indices": [ 1, 3 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.1.0", "model_run_id": "67f7e65d-a914-4f87-b80d-f42ea0b392df", "prompt_bundle_hash": "68b36d7b8e987c40714ce6ad7f2a17010be4a1ebe0822fd26b09fe4742288073", "input_evidence_hash": "425a9d833425dc2547f41667baedfd8f7f5039f220b73a91f48334f7ba77c854" }, { "case_id": "oas-g0-007", "perspective": "client_agent_report", "dimension": "goal", "value": 0.1, "confidence": 0.25, "evidence_turn_indices": [ 0, 1, 2 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.1.0", "model_run_id": "6d38b5c8-d763-44ce-8f54-568dace8ff54", "prompt_bundle_hash": "68b36d7b8e987c40714ce6ad7f2a17010be4a1ebe0822fd26b09fe4742288073", "input_evidence_hash": "69ceeff8e25be657653a68fb46ae9b0b5789212806c1ad96c3e41023fe7c3aa6" }, { "case_id": "oas-g0-007", "perspective": "client_agent_report", "dimension": "task", "value": 0.05, "confidence": 0.75, "evidence_turn_indices": [ 0, 1, 2 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.1.0", "model_run_id": "6d38b5c8-d763-44ce-8f54-568dace8ff54", "prompt_bundle_hash": "68b36d7b8e987c40714ce6ad7f2a17010be4a1ebe0822fd26b09fe4742288073", "input_evidence_hash": "69ceeff8e25be657653a68fb46ae9b0b5789212806c1ad96c3e41023fe7c3aa6" }, { "case_id": "oas-g0-007", "perspective": "client_agent_report", "dimension": "bond", "value": 0.05, "confidence": 0.8, "evidence_turn_indices": [ 0, 1, 2 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.1.0", "model_run_id": "6d38b5c8-d763-44ce-8f54-568dace8ff54", "prompt_bundle_hash": "68b36d7b8e987c40714ce6ad7f2a17010be4a1ebe0822fd26b09fe4742288073", "input_evidence_hash": "69ceeff8e25be657653a68fb46ae9b0b5789212806c1ad96c3e41023fe7c3aa6" }, { "case_id": "oas-g0-008", "perspective": "client_agent_report", "dimension": "goal", "value": 0.2, "confidence": 0.55, "evidence_turn_indices": [ 2, 3 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.1.0", "model_run_id": "fbec4430-a4af-4341-9f10-b5e8fe95487b", "prompt_bundle_hash": "68b36d7b8e987c40714ce6ad7f2a17010be4a1ebe0822fd26b09fe4742288073", "input_evidence_hash": "1be15db7456de3aed4450b378396bd1cdd1f01695247654c6133e16073ddfcc9" }, { "case_id": "oas-g0-008", "perspective": "client_agent_report", "dimension": "task", "value": 0.45, "confidence": 0.5, "evidence_turn_indices": [ 0, 1, 3 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.1.0", "model_run_id": "fbec4430-a4af-4341-9f10-b5e8fe95487b", "prompt_bundle_hash": "68b36d7b8e987c40714ce6ad7f2a17010be4a1ebe0822fd26b09fe4742288073", "input_evidence_hash": "1be15db7456de3aed4450b378396bd1cdd1f01695247654c6133e16073ddfcc9" }, { "case_id": "oas-g0-008", "perspective": "client_agent_report", "dimension": "bond", "value": 0.7, "confidence": 0.65, "evidence_turn_indices": [ 0, 1 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.1.0", "model_run_id": "fbec4430-a4af-4341-9f10-b5e8fe95487b", "prompt_bundle_hash": "68b36d7b8e987c40714ce6ad7f2a17010be4a1ebe0822fd26b09fe4742288073", "input_evidence_hash": "1be15db7456de3aed4450b378396bd1cdd1f01695247654c6133e16073ddfcc9" }, { "case_id": "oas-g0-008", "perspective": "independent_observer", "dimension": "goal", "value": 0.2, "confidence": 0.45, "evidence_turn_indices": [ 2, 3 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.1.0", "model_run_id": "071f0970-f9be-4a8b-820b-fb645f2159da", "prompt_bundle_hash": "e5ae373bbb90596b0086fcb73985d37f65a920e9b407d145efe89ef76f58034f", "input_evidence_hash": "1be15db7456de3aed4450b378396bd1cdd1f01695247654c6133e16073ddfcc9" }, { "case_id": "oas-g0-008", "perspective": "independent_observer", "dimension": "task", "value": 0.3, "confidence": 0.4, "evidence_turn_indices": [ 0, 1, 3 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.1.0", "model_run_id": "071f0970-f9be-4a8b-820b-fb645f2159da", "prompt_bundle_hash": "e5ae373bbb90596b0086fcb73985d37f65a920e9b407d145efe89ef76f58034f", "input_evidence_hash": "1be15db7456de3aed4450b378396bd1cdd1f01695247654c6133e16073ddfcc9" }, { "case_id": "oas-g0-008", "perspective": "independent_observer", "dimension": "bond", "value": 0.6, "confidence": 0.5, "evidence_turn_indices": [ 0, 1 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.1.0", "model_run_id": "071f0970-f9be-4a8b-820b-fb645f2159da", "prompt_bundle_hash": "e5ae373bbb90596b0086fcb73985d37f65a920e9b407d145efe89ef76f58034f", "input_evidence_hash": "1be15db7456de3aed4450b378396bd1cdd1f01695247654c6133e16073ddfcc9" } ] }