{ "schema_version": "alliance-calibration-snapshot.v1", "run_id": "g1-candidate-2026-08-06", "provider": "claude_cli", "model": "mixed", "prompt_bundle_version": "1.0.0", "benchmark_schema": "vignette.outcome_alliance_benchmark.v1", "generated_at": "2026-08-06T12:21:16.722685Z", "predictions": [ { "case_id": "oas-g0-001", "perspective": "independent_observer", "dimension": "goal", "value": 0.1, "confidence": 0.7, "evidence_turn_indices": [ 0, 1, 2 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "ed1bc174-584e-4651-b627-562b7752f337", "prompt_bundle_hash": "f09592ae8397cffee096a67773db0bcb4398f419ad8ae1629b7de0497b7c8908", "input_evidence_hash": "73c73daff30eb59a0566592162954833c32db3fb8ef971ffeb593003e8b2ec88" }, { "case_id": "oas-g0-001", "perspective": "independent_observer", "dimension": "task", "value": 0.1, "confidence": 0.65, "evidence_turn_indices": [ 1, 2 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "ed1bc174-584e-4651-b627-562b7752f337", "prompt_bundle_hash": "f09592ae8397cffee096a67773db0bcb4398f419ad8ae1629b7de0497b7c8908", "input_evidence_hash": "73c73daff30eb59a0566592162954833c32db3fb8ef971ffeb593003e8b2ec88" }, { "case_id": "oas-g0-001", "perspective": "independent_observer", "dimension": "bond", "value": 0.25, "confidence": 0.4, "evidence_turn_indices": [ 0, 1, 2 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "ed1bc174-584e-4651-b627-562b7752f337", "prompt_bundle_hash": "f09592ae8397cffee096a67773db0bcb4398f419ad8ae1629b7de0497b7c8908", "input_evidence_hash": "73c73daff30eb59a0566592162954833c32db3fb8ef971ffeb593003e8b2ec88" }, { "case_id": "oas-g0-002", "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": "1.0.0", "model_run_id": "47861f3c-69e9-4f0d-9faa-70428e3b17d8", "prompt_bundle_hash": "f09592ae8397cffee096a67773db0bcb4398f419ad8ae1629b7de0497b7c8908", "input_evidence_hash": "27c4b079f44e906240c6545e9ef02411d276c7fbcb8de23b9ea1ca79106adbbb" }, { "case_id": "oas-g0-002", "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": "1.0.0", "model_run_id": "47861f3c-69e9-4f0d-9faa-70428e3b17d8", "prompt_bundle_hash": "f09592ae8397cffee096a67773db0bcb4398f419ad8ae1629b7de0497b7c8908", "input_evidence_hash": "27c4b079f44e906240c6545e9ef02411d276c7fbcb8de23b9ea1ca79106adbbb" }, { "case_id": "oas-g0-002", "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": "1.0.0", "model_run_id": "47861f3c-69e9-4f0d-9faa-70428e3b17d8", "prompt_bundle_hash": "f09592ae8397cffee096a67773db0bcb4398f419ad8ae1629b7de0497b7c8908", "input_evidence_hash": "27c4b079f44e906240c6545e9ef02411d276c7fbcb8de23b9ea1ca79106adbbb" }, { "case_id": "oas-g0-003", "perspective": "independent_observer", "dimension": "goal", "value": 0.1, "confidence": 0.1, "evidence_turn_indices": [ 0, 1, 2 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "b286da58-99f3-46dd-8052-61524075ae5b", "prompt_bundle_hash": "f09592ae8397cffee096a67773db0bcb4398f419ad8ae1629b7de0497b7c8908", "input_evidence_hash": "eef29a4ff7245ce97c856ec22c6ca74c19ab33c989259f6f0c86d6c60fa04c56" }, { "case_id": "oas-g0-003", "perspective": "independent_observer", "dimension": "task", "value": 0.15, "confidence": 0.1, "evidence_turn_indices": [ 1, 2 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "b286da58-99f3-46dd-8052-61524075ae5b", "prompt_bundle_hash": "f09592ae8397cffee096a67773db0bcb4398f419ad8ae1629b7de0497b7c8908", "input_evidence_hash": "eef29a4ff7245ce97c856ec22c6ca74c19ab33c989259f6f0c86d6c60fa04c56" }, { "case_id": "oas-g0-003", "perspective": "independent_observer", "dimension": "bond", "value": 0.4, "confidence": 0.3, "evidence_turn_indices": [ 0, 1, 2 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "b286da58-99f3-46dd-8052-61524075ae5b", "prompt_bundle_hash": "f09592ae8397cffee096a67773db0bcb4398f419ad8ae1629b7de0497b7c8908", "input_evidence_hash": "eef29a4ff7245ce97c856ec22c6ca74c19ab33c989259f6f0c86d6c60fa04c56" }, { "case_id": "oas-g0-005", "perspective": "client_agent_report", "dimension": "goal", "value": 0.15, "confidence": 0.25, "evidence_turn_indices": [ 0, 1 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "aa57848f-8d71-40ce-b003-32990dbe76ed", "prompt_bundle_hash": "35787f3e2d20a52f402ac5027a44b4a98b8833065c15f5c0949d3172aebeef3d", "input_evidence_hash": "df68b447c1d21ea038d185894ea7fcbfacdb9757aee11df5e996dfc7efb600fd" }, { "case_id": "oas-g0-005", "perspective": "client_agent_report", "dimension": "task", "value": 0.1, "confidence": 0.2, "evidence_turn_indices": [ 0, 2 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "aa57848f-8d71-40ce-b003-32990dbe76ed", "prompt_bundle_hash": "35787f3e2d20a52f402ac5027a44b4a98b8833065c15f5c0949d3172aebeef3d", "input_evidence_hash": "df68b447c1d21ea038d185894ea7fcbfacdb9757aee11df5e996dfc7efb600fd" }, { "case_id": "oas-g0-005", "perspective": "client_agent_report", "dimension": "bond", "value": 0.12, "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": "1.0.0", "model_run_id": "aa57848f-8d71-40ce-b003-32990dbe76ed", "prompt_bundle_hash": "35787f3e2d20a52f402ac5027a44b4a98b8833065c15f5c0949d3172aebeef3d", "input_evidence_hash": "df68b447c1d21ea038d185894ea7fcbfacdb9757aee11df5e996dfc7efb600fd" }, { "case_id": "oas-g0-006", "perspective": "client_agent_report", "dimension": "goal", "value": 0.35, "confidence": 0.3, "evidence_turn_indices": [ 0, 2, 3 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "4e37fde1-9dee-4ceb-bdf9-2470325449e4", "prompt_bundle_hash": "35787f3e2d20a52f402ac5027a44b4a98b8833065c15f5c0949d3172aebeef3d", "input_evidence_hash": "425a9d833425dc2547f41667baedfd8f7f5039f220b73a91f48334f7ba77c854" }, { "case_id": "oas-g0-006", "perspective": "client_agent_report", "dimension": "task", "value": 0.55, "confidence": 0.5, "evidence_turn_indices": [ 2, 3 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "4e37fde1-9dee-4ceb-bdf9-2470325449e4", "prompt_bundle_hash": "35787f3e2d20a52f402ac5027a44b4a98b8833065c15f5c0949d3172aebeef3d", "input_evidence_hash": "425a9d833425dc2547f41667baedfd8f7f5039f220b73a91f48334f7ba77c854" }, { "case_id": "oas-g0-006", "perspective": "client_agent_report", "dimension": "bond", "value": 0.5, "confidence": 0.45, "evidence_turn_indices": [ 0, 1, 3 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "4e37fde1-9dee-4ceb-bdf9-2470325449e4", "prompt_bundle_hash": "35787f3e2d20a52f402ac5027a44b4a98b8833065c15f5c0949d3172aebeef3d", "input_evidence_hash": "425a9d833425dc2547f41667baedfd8f7f5039f220b73a91f48334f7ba77c854" }, { "case_id": "oas-g0-007", "perspective": "client_agent_report", "dimension": "goal", "value": 0.15, "confidence": 0.2, "evidence_turn_indices": [ 1 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "bb14c5b8-55e0-494a-ad5a-62a20711d5c6", "prompt_bundle_hash": "35787f3e2d20a52f402ac5027a44b4a98b8833065c15f5c0949d3172aebeef3d", "input_evidence_hash": "69ceeff8e25be657653a68fb46ae9b0b5789212806c1ad96c3e41023fe7c3aa6" }, { "case_id": "oas-g0-007", "perspective": "client_agent_report", "dimension": "task", "value": 0.1, "confidence": 0.15, "evidence_turn_indices": [ 1 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "bb14c5b8-55e0-494a-ad5a-62a20711d5c6", "prompt_bundle_hash": "35787f3e2d20a52f402ac5027a44b4a98b8833065c15f5c0949d3172aebeef3d", "input_evidence_hash": "69ceeff8e25be657653a68fb46ae9b0b5789212806c1ad96c3e41023fe7c3aa6" }, { "case_id": "oas-g0-007", "perspective": "client_agent_report", "dimension": "bond", "value": 0.1, "confidence": 0.85, "evidence_turn_indices": [ 0, 1, 2 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "bb14c5b8-55e0-494a-ad5a-62a20711d5c6", "prompt_bundle_hash": "35787f3e2d20a52f402ac5027a44b4a98b8833065c15f5c0949d3172aebeef3d", "input_evidence_hash": "69ceeff8e25be657653a68fb46ae9b0b5789212806c1ad96c3e41023fe7c3aa6" }, { "case_id": "oas-g0-008", "perspective": "client_agent_report", "dimension": "goal", "value": 0.15, "confidence": 0.45, "evidence_turn_indices": [ 2, 3 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "12b8464b-943b-4309-a370-3ad0e2f6042e", "prompt_bundle_hash": "35787f3e2d20a52f402ac5027a44b4a98b8833065c15f5c0949d3172aebeef3d", "input_evidence_hash": "1be15db7456de3aed4450b378396bd1cdd1f01695247654c6133e16073ddfcc9" }, { "case_id": "oas-g0-008", "perspective": "client_agent_report", "dimension": "task", "value": 0.1, "confidence": 0.4, "evidence_turn_indices": [ 2, 3 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "12b8464b-943b-4309-a370-3ad0e2f6042e", "prompt_bundle_hash": "35787f3e2d20a52f402ac5027a44b4a98b8833065c15f5c0949d3172aebeef3d", "input_evidence_hash": "1be15db7456de3aed4450b378396bd1cdd1f01695247654c6133e16073ddfcc9" }, { "case_id": "oas-g0-008", "perspective": "client_agent_report", "dimension": "bond", "value": 0.55, "confidence": 0.35, "evidence_turn_indices": [ 0, 1 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "12b8464b-943b-4309-a370-3ad0e2f6042e", "prompt_bundle_hash": "35787f3e2d20a52f402ac5027a44b4a98b8833065c15f5c0949d3172aebeef3d", "input_evidence_hash": "1be15db7456de3aed4450b378396bd1cdd1f01695247654c6133e16073ddfcc9" }, { "case_id": "oas-g0-008", "perspective": "independent_observer", "dimension": "goal", "value": 0.1, "confidence": 0.45, "evidence_turn_indices": [ 2, 3 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "e35b3db9-888c-4e9c-bf4d-f7e24b8aaccc", "prompt_bundle_hash": "f09592ae8397cffee096a67773db0bcb4398f419ad8ae1629b7de0497b7c8908", "input_evidence_hash": "1be15db7456de3aed4450b378396bd1cdd1f01695247654c6133e16073ddfcc9" }, { "case_id": "oas-g0-008", "perspective": "independent_observer", "dimension": "task", "value": 0.1, "confidence": 0.3, "evidence_turn_indices": [ 2, 3 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "e35b3db9-888c-4e9c-bf4d-f7e24b8aaccc", "prompt_bundle_hash": "f09592ae8397cffee096a67773db0bcb4398f419ad8ae1629b7de0497b7c8908", "input_evidence_hash": "1be15db7456de3aed4450b378396bd1cdd1f01695247654c6133e16073ddfcc9" }, { "case_id": "oas-g0-008", "perspective": "independent_observer", "dimension": "bond", "value": 0.55, "confidence": 0.45, "evidence_turn_indices": [ 0, 1 ], "status": "ready", "error_code": null, "provider": "claude_cli", "model": "claude-opus-4-8", "prompt_bundle_version": "1.0.0", "model_run_id": "e35b3db9-888c-4e9c-bf4d-f7e24b8aaccc", "prompt_bundle_hash": "f09592ae8397cffee096a67773db0bcb4398f419ad8ae1629b7de0497b7c8908", "input_evidence_hash": "1be15db7456de3aed4450b378396bd1cdd1f01695247654c6133e16073ddfcc9" } ] }