constraints dict | family stringclasses 8
values | id stringlengths 26 39 | label dict | oracle dict | prompt stringlengths 1.67k 1.76k | schema_version stringclasses 1
value | source stringclasses 1
value | split stringclasses 1
value | target stringlengths 352 485 | task dict | workers listlengths 3 3 |
|---|---|---|---|---|---|---|---|---|---|---|---|
{
"max_calls": 3,
"max_credits": 23.766,
"max_latency_ms": 241531,
"minimum_quality": 0.9506
} | research_synthesis | synthetic-research_synthesis-016901 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 19.519,
"estimated_latency_ms": 170293,
"estimated_pass_rate": 0.7211,
"estimated_quality": 0.9096,
"finalizer": "worker_b",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7211,
"quality": 0.9096,
"quality_target_met": false,
"utility": 70.813256
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":19.519,"estimated_latency_ms":170293,"estimated_pass_rate":0.7211,"estimated_quality":0.9096,"finalizer":"worker_b","independent_verification":true,"max_calls":3,"max_credits":23.766,"max_latency_ms":241531,"policy":"str... | {
"ambiguity": 0.8423,
"category": "research",
"complexity": 0.6075,
"consequence": 0.4493,
"evidence_need": 0.9023,
"family": "research_synthesis",
"input_tokens": 3680,
"request_id": "synthetic-research_synthesis-016901",
"reversibility": 0.8458,
"risk": 0.3863,
"summary": "Compare evidence for ... | [
{
"credits_per_call": 1.127,
"id": "worker_a",
"p95_latency_ms": 18975,
"pass_rate": 0.5123,
"provider_group": "provider_z",
"quality": 0.6471,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 15.119,
"id": "worker_b",
"p95_lat... |
{
"max_calls": 3,
"max_credits": 24.671,
"max_latency_ms": 233330,
"minimum_quality": 0.9385
} | incident_response | synthetic-incident_response-015543 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 17.801,
"estimated_latency_ms": 180259,
"estimated_pass_rate": 0.7536,
"estimated_quality": 0.9074,
"finalizer": "worker_c",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7536,
"quality": 0.9074,
"quality_target_met": false,
"utility": 74.730403
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":17.801,"estimated_latency_ms":180259,"estimated_pass_rate":0.7536,"estimated_quality":0.9074,"finalizer":"worker_c","independent_verification":true,"max_calls":3,"max_credits":24.671,"max_latency_ms":233330,"policy":"str... | {
"ambiguity": 0.5894,
"category": "operations",
"complexity": 0.8331,
"consequence": 0.7332,
"evidence_need": 0.5287,
"family": "incident_response",
"input_tokens": 6840,
"request_id": "synthetic-incident_response-015543",
"reversibility": 0.154,
"risk": 0.5935,
"summary": "Prioritize actions dur... | [
{
"credits_per_call": 3.111,
"id": "worker_a",
"p95_latency_ms": 71252,
"pass_rate": 0.7762,
"provider_group": "provider_y",
"quality": 0.9183,
"strengths": [
"drafting",
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 0.864,
"id": "wor... |
{
"max_calls": 2,
"max_credits": 10.344,
"max_latency_ms": 182663,
"minimum_quality": 0.7966
} | feature_implementation | synthetic-feature_implementation-014611 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 4.54,
"estimated_latency_ms": 68397,
"estimated_pass_rate": 0.589,
"estimated_quality": 0.8268,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 10.3... | {
"candidate_count": 4,
"pass_rate": 0.589,
"quality": 0.8268,
"quality_target_met": true,
"utility": 60.269307
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":4.54,"estimated_latency_ms":68397,"estimated_pass_rate":0.589,"estimated_quality":0.8268,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":10.344,"max_latency_ms":182663,"policy":"balanced","reason_codes":["indep... | {
"ambiguity": 0.2862,
"category": "coding",
"complexity": 0.572,
"consequence": 0.4858,
"evidence_need": 0.601,
"family": "feature_implementation",
"input_tokens": 4240,
"request_id": "synthetic-feature_implementation-014611",
"reversibility": 0.5153,
"risk": 0.3592,
"summary": "Patch a productio... | [
{
"credits_per_call": 11.404,
"id": "worker_a",
"p95_latency_ms": 67963,
"pass_rate": 0.9113,
"provider_group": "provider_x",
"quality": 0.995,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 1.173,
"id": "worker_b",
... |
{
"max_calls": 3,
"max_credits": 18.535,
"max_latency_ms": 171242,
"minimum_quality": 0.7869
} | incident_response | synthetic-incident_response-009039 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 17.329,
"estimated_latency_ms": 152162,
"estimated_pass_rate": 0.7656,
"estimated_quality": 0.9154,
"finalizer": "worker_c",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7656,
"quality": 0.9154,
"quality_target_met": true,
"utility": 74.666081
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":17.329,"estimated_latency_ms":152162,"estimated_pass_rate":0.7656,"estimated_quality":0.9154,"finalizer":"worker_c","independent_verification":true,"max_calls":3,"max_credits":18.535,"max_latency_ms":171242,"policy":"str... | {
"ambiguity": 0.3884,
"category": "operations",
"complexity": 0.7176,
"consequence": 0.6593,
"evidence_need": 0.5032,
"family": "incident_response",
"input_tokens": 6200,
"request_id": "synthetic-incident_response-009039",
"reversibility": 0.5091,
"risk": 0.8694,
"summary": "Prioritize actions du... | [
{
"credits_per_call": 3.287,
"id": "worker_a",
"p95_latency_ms": 65022,
"pass_rate": 0.8275,
"provider_group": "provider_y",
"quality": 0.9064,
"strengths": [
"drafting",
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 0.812,
"id": "wor... |
{
"max_calls": 3,
"max_credits": 23.079,
"max_latency_ms": 240628,
"minimum_quality": 0.9465
} | research_synthesis | synthetic-research_synthesis-019893 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 17.556,
"estimated_latency_ms": 169058,
"estimated_pass_rate": 0.7224,
"estimated_quality": 0.922,
"finalizer": "worker_c",
"independent_verification"... | {
"candidate_count": 15,
"pass_rate": 0.7224,
"quality": 0.922,
"quality_target_met": false,
"utility": 72.093673
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":17.556,"estimated_latency_ms":169058,"estimated_pass_rate":0.7224,"estimated_quality":0.922,"finalizer":"worker_c","independent_verification":true,"max_calls":3,"max_credits":23.079,"max_latency_ms":240628,"policy":"stri... | {
"ambiguity": 0.8758,
"category": "research",
"complexity": 0.5887,
"consequence": 0.2301,
"evidence_need": 0.9336,
"family": "research_synthesis",
"input_tokens": 880,
"request_id": "synthetic-research_synthesis-019893",
"reversibility": 0.8334,
"risk": 0.4028,
"summary": "Produce a decision mem... | [
{
"credits_per_call": 1.03,
"id": "worker_a",
"p95_latency_ms": 32177,
"pass_rate": 0.5125,
"provider_group": "provider_z",
"quality": 0.7157,
"strengths": [
"coding",
"drafting",
"speed",
"synthesis"
]
},
{
"credits_per_call": 3.971,
"id": "worker... |
{
"max_calls": 1,
"max_credits": 1.334,
"max_latency_ms": 40218,
"minimum_quality": 0.7721
} | content_revision | synthetic-content_revision-007434 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 1.152,
"estimated_latency_ms": 21236,
"estimated_pass_rate": 0.332,
"estimated_quality": 0.7219,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 1.334,
"max_latency_... | {
"candidate_count": 1,
"pass_rate": 0.332,
"quality": 0.7219,
"quality_target_met": false,
"utility": 41.864322
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":1.152,"estimated_latency_ms":21236,"estimated_pass_rate":0.332,"estimated_quality":0.7219,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":1.334,"max_latency_ms":40218,"policy":"lean","reason_codes":["best_feasible_bel... | {
"ambiguity": 0.2364,
"category": "writing",
"complexity": 0.3135,
"consequence": 0.0788,
"evidence_need": 0.1114,
"family": "content_revision",
"input_tokens": 4040,
"request_id": "synthetic-content_revision-007434",
"reversibility": 0.9176,
"risk": 0.1177,
"summary": "Rewrite a webhook delivery... | [
{
"credits_per_call": 14.512,
"id": "worker_a",
"p95_latency_ms": 88636,
"pass_rate": 0.995,
"provider_group": "provider_x",
"quality": 0.9606,
"strengths": [
"implementation",
"reasoning",
"synthesis",
"writing"
]
},
{
"credits_per_call": 1.152,
"... |
{
"max_calls": 2,
"max_credits": 9.328,
"max_latency_ms": 173502,
"minimum_quality": 0.7728
} | feature_implementation | synthetic-feature_implementation-012979 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 4.918,
"estimated_latency_ms": 78828,
"estimated_pass_rate": 0.5614,
"estimated_quality": 0.7844,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 9.... | {
"candidate_count": 4,
"pass_rate": 0.5614,
"quality": 0.7844,
"quality_target_met": true,
"utility": 56.034417
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":4.918,"estimated_latency_ms":78828,"estimated_pass_rate":0.5614,"estimated_quality":0.7844,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":9.328,"max_latency_ms":173502,"policy":"balanced","reason_codes":["inde... | {
"ambiguity": 0.3922,
"category": "coding",
"complexity": 0.5924,
"consequence": 0.3682,
"evidence_need": 0.4892,
"family": "feature_implementation",
"input_tokens": 1440,
"request_id": "synthetic-feature_implementation-012979",
"reversibility": 0.7691,
"risk": 0.5068,
"summary": "Patch a product... | [
{
"credits_per_call": 3.945,
"id": "worker_a",
"p95_latency_ms": 47013,
"pass_rate": 0.7728,
"provider_group": "provider_z",
"quality": 0.7964,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 0.973,
"id": "worker_b",
"p95_l... |
{
"max_calls": 2,
"max_credits": 6.132,
"max_latency_ms": 206255,
"minimum_quality": 0.6925
} | research_synthesis | synthetic-research_synthesis-018469 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 4.57,
"estimated_latency_ms": 94526,
"estimated_pass_rate": 0.4849,
"estimated_quality": 0.7741,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 6.1... | {
"candidate_count": 4,
"pass_rate": 0.4849,
"quality": 0.7741,
"quality_target_met": true,
"utility": 52.201541
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":4.57,"estimated_latency_ms":94526,"estimated_pass_rate":0.4849,"estimated_quality":0.7741,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":6.132,"max_latency_ms":206255,"policy":"balanced","reason_codes":["compl... | {
"ambiguity": 0.8514,
"category": "research",
"complexity": 0.6837,
"consequence": 0.3307,
"evidence_need": 0.9141,
"family": "research_synthesis",
"input_tokens": 7240,
"request_id": "synthetic-research_synthesis-018469",
"reversibility": 0.912,
"risk": 0.4807,
"summary": "Review competing appro... | [
{
"credits_per_call": 0.898,
"id": "worker_a",
"p95_latency_ms": 31899,
"pass_rate": 0.4106,
"provider_group": "provider_y",
"quality": 0.7065,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 14.765,
"id": "worker_b",
"p95_lat... |
{
"max_calls": 1,
"max_credits": 3.628,
"max_latency_ms": 40005,
"minimum_quality": 0.4812
} | content_revision | synthetic-content_revision-007882 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 1.179,
"estimated_latency_ms": 21509,
"estimated_pass_rate": 0.3729,
"estimated_quality": 0.7143,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 3.628,
"max_latency... | {
"candidate_count": 1,
"pass_rate": 0.3729,
"quality": 0.7143,
"quality_target_met": true,
"utility": 48.169033
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":1.179,"estimated_latency_ms":21509,"estimated_pass_rate":0.3729,"estimated_quality":0.7143,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":3.628,"max_latency_ms":40005,"policy":"lean","reason_codes":["single_call_suff... | {
"ambiguity": 0.2336,
"category": "writing",
"complexity": 0.3892,
"consequence": 0.2444,
"evidence_need": 0.2937,
"family": "content_revision",
"input_tokens": 7000,
"request_id": "synthetic-content_revision-007882",
"reversibility": 0.927,
"risk": 0.1034,
"summary": "Edit documentation for webh... | [
{
"credits_per_call": 1.179,
"id": "worker_a",
"p95_latency_ms": 21509,
"pass_rate": 0.4431,
"provider_group": "provider_y",
"quality": 0.7278,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 13.06,
"id": "worker_b",
"p95_late... |
{
"max_calls": 2,
"max_credits": 7.501,
"max_latency_ms": 151201,
"minimum_quality": 0.7095
} | product_tradeoff | synthetic-product_tradeoff-016312 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 3.895,
"estimated_latency_ms": 108195,
"estimated_pass_rate": 0.6382,
"estimated_quality": 0.8198,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 7... | {
"candidate_count": 4,
"pass_rate": 0.6382,
"quality": 0.8198,
"quality_target_met": true,
"utility": 58.145265
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":3.895,"estimated_latency_ms":108195,"estimated_pass_rate":0.6382,"estimated_quality":0.8198,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":7.501,"max_latency_ms":151201,"policy":"balanced","reason_codes":["ind... | {
"ambiguity": 0.6881,
"category": "decision",
"complexity": 0.3403,
"consequence": 0.2388,
"evidence_need": 0.6392,
"family": "product_tradeoff",
"input_tokens": 6520,
"request_id": "synthetic-product_tradeoff-016312",
"reversibility": 0.5974,
"risk": 0.4141,
"summary": "Recommend whether to laun... | [
{
"credits_per_call": 2.814,
"id": "worker_a",
"p95_latency_ms": 78737,
"pass_rate": 0.7347,
"provider_group": "provider_x",
"quality": 0.8095,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 1.081,
"id": "worker_b",
"p95_l... |
{
"max_calls": 3,
"max_credits": 6.456,
"max_latency_ms": 186924,
"minimum_quality": 0.8002
} | incident_response | synthetic-incident_response-014679 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 4.822,
"estimated_latency_ms": 87992,
"estimated_pass_rate": 0.6868,
"estimated_quality": 0.8149,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 6.... | {
"candidate_count": 4,
"pass_rate": 0.6868,
"quality": 0.8149,
"quality_target_met": true,
"utility": 63.90426
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":4.822,"estimated_latency_ms":87992,"estimated_pass_rate":0.6868,"estimated_quality":0.8149,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":6.456,"max_latency_ms":186924,"policy":"balanced","reason_codes":["comp... | {
"ambiguity": 0.3914,
"category": "operations",
"complexity": 0.7755,
"consequence": 0.8832,
"evidence_need": 0.5496,
"family": "incident_response",
"input_tokens": 3600,
"request_id": "synthetic-incident_response-014679",
"reversibility": 0.3279,
"risk": 0.761,
"summary": "Plan immediate contain... | [
{
"credits_per_call": 1.048,
"id": "worker_a",
"p95_latency_ms": 18404,
"pass_rate": 0.5739,
"provider_group": "provider_x",
"quality": 0.751,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 3.774,
"id": "worker_b",
"p95_laten... |
{
"max_calls": 3,
"max_credits": 14.234,
"max_latency_ms": 113158,
"minimum_quality": 0.7731
} | incident_response | synthetic-incident_response-017263 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 12.86,
"estimated_latency_ms": 98031,
"estimated_pass_rate": 0.6691,
"estimated_quality": 0.872,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 14.... | {
"candidate_count": 7,
"pass_rate": 0.6691,
"quality": 0.872,
"quality_target_met": true,
"utility": 64.309584
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":12.86,"estimated_latency_ms":98031,"estimated_pass_rate":0.6691,"estimated_quality":0.872,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":14.234,"max_latency_ms":113158,"policy":"balanced","reason_codes":["comp... | {
"ambiguity": 0.7647,
"category": "operations",
"complexity": 0.6351,
"consequence": 0.8386,
"evidence_need": 0.7023,
"family": "incident_response",
"input_tokens": 1640,
"request_id": "synthetic-incident_response-017263",
"reversibility": 0.5296,
"risk": 0.7549,
"summary": "Review an outage invo... | [
{
"credits_per_call": 11.686,
"id": "worker_a",
"p95_latency_ms": 75274,
"pass_rate": 0.8877,
"provider_group": "provider_y",
"quality": 0.995,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 2.952,
"id": "worker_b",
... |
{
"max_calls": 3,
"max_credits": 14.434,
"max_latency_ms": 118499,
"minimum_quality": 0.8322
} | debugging | synthetic-debugging-002132 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 13.824,
"estimated_latency_ms": 79200,
"estimated_pass_rate": 0.6576,
"estimated_quality": 0.8577,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 1... | {
"candidate_count": 7,
"pass_rate": 0.6576,
"quality": 0.8577,
"quality_target_met": true,
"utility": 60.50566
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":13.824,"estimated_latency_ms":79200,"estimated_pass_rate":0.6576,"estimated_quality":0.8577,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":14.434,"max_latency_ms":118499,"policy":"balanced","reason_codes":["co... | {
"ambiguity": 0.7085,
"category": "coding",
"complexity": 0.8815,
"consequence": 0.5402,
"evidence_need": 0.7015,
"family": "debugging",
"input_tokens": 6760,
"request_id": "synthetic-debugging-002132",
"reversibility": 0.6224,
"risk": 0.6086,
"summary": "Explain why OAuth callbacks fails only un... | [
{
"credits_per_call": 12.839,
"id": "worker_a",
"p95_latency_ms": 53078,
"pass_rate": 0.995,
"provider_group": "provider_x",
"quality": 0.9511,
"strengths": [
"coding",
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 0.985,
"i... |
{
"max_calls": 2,
"max_credits": 11.167,
"max_latency_ms": 116860,
"minimum_quality": 0.8846
} | incident_response | synthetic-incident_response-006415 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 4.16,
"estimated_latency_ms": 70121,
"estimated_pass_rate": 0.6669,
"estimated_quality": 0.7925,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 11.... | {
"candidate_count": 4,
"pass_rate": 0.6669,
"quality": 0.7925,
"quality_target_met": false,
"utility": 53.707928
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":4.16,"estimated_latency_ms":70121,"estimated_pass_rate":0.6669,"estimated_quality":0.7925,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":11.167,"max_latency_ms":116860,"policy":"balanced","reason_codes":["best... | {
"ambiguity": 0.4501,
"category": "operations",
"complexity": 0.9368,
"consequence": 0.6527,
"evidence_need": 0.524,
"family": "incident_response",
"input_tokens": 2400,
"request_id": "synthetic-incident_response-006415",
"reversibility": 0.1517,
"risk": 0.8158,
"summary": "Review an outage invol... | [
{
"credits_per_call": 13.622,
"id": "worker_a",
"p95_latency_ms": 68815,
"pass_rate": 0.995,
"provider_group": "provider_x",
"quality": 0.995,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 3.183,
"id": "worker_b",
... |
{
"max_calls": 2,
"max_credits": 10.738,
"max_latency_ms": 105985,
"minimum_quality": 0.7614
} | incident_response | synthetic-incident_response-016879 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 4.133,
"estimated_latency_ms": 90588,
"estimated_pass_rate": 0.6645,
"estimated_quality": 0.8096,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 10... | {
"candidate_count": 4,
"pass_rate": 0.6645,
"quality": 0.8096,
"quality_target_met": true,
"utility": 66.048748
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":4.133,"estimated_latency_ms":90588,"estimated_pass_rate":0.6645,"estimated_quality":0.8096,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":10.738,"max_latency_ms":105985,"policy":"balanced","reason_codes":["com... | {
"ambiguity": 0.7633,
"category": "operations",
"complexity": 0.6139,
"consequence": 0.9446,
"evidence_need": 0.6041,
"family": "incident_response",
"input_tokens": 7960,
"request_id": "synthetic-incident_response-016879",
"reversibility": 0.188,
"risk": 0.9243,
"summary": "Review an outage invol... | [
{
"credits_per_call": 3.296,
"id": "worker_a",
"p95_latency_ms": 68272,
"pass_rate": 0.8316,
"provider_group": "provider_y",
"quality": 0.8563,
"strengths": [
"drafting",
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 0.837,
"id": "wor... |
{
"max_calls": 2,
"max_credits": 11.872,
"max_latency_ms": 71872,
"minimum_quality": 0.7363
} | incident_response | synthetic-incident_response-018367 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 4.84,
"estimated_latency_ms": 65544,
"estimated_pass_rate": 0.641,
"estimated_quality": 0.817,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 11.87... | {
"candidate_count": 5,
"pass_rate": 0.641,
"quality": 0.817,
"quality_target_met": true,
"utility": 61.425172
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":4.84,"estimated_latency_ms":65544,"estimated_pass_rate":0.641,"estimated_quality":0.817,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":11.872,"max_latency_ms":71872,"policy":"balanced","reason_codes":["complex... | {
"ambiguity": 0.6443,
"category": "operations",
"complexity": 0.7704,
"consequence": 0.8091,
"evidence_need": 0.5793,
"family": "incident_response",
"input_tokens": 6560,
"request_id": "synthetic-incident_response-018367",
"reversibility": 0.3963,
"risk": 0.5699,
"summary": "Review an outage invo... | [
{
"credits_per_call": 3.775,
"id": "worker_a",
"p95_latency_ms": 45876,
"pass_rate": 0.8284,
"provider_group": "provider_z",
"quality": 0.9122,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 10.54,
"id": "worker_b",
"p95_l... |
{
"max_calls": 3,
"max_credits": 10.538,
"max_latency_ms": 156395,
"minimum_quality": 0.7752
} | product_tradeoff | synthetic-product_tradeoff-005832 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 5.084,
"estimated_latency_ms": 80179,
"estimated_pass_rate": 0.7554,
"estimated_quality": 0.7846,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 10... | {
"candidate_count": 4,
"pass_rate": 0.7554,
"quality": 0.7846,
"quality_target_met": true,
"utility": 55.926134
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":5.084,"estimated_latency_ms":80179,"estimated_pass_rate":0.7554,"estimated_quality":0.7846,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":10.538,"max_latency_ms":156395,"policy":"balanced","reason_codes":["ind... | {
"ambiguity": 0.7192,
"category": "decision",
"complexity": 0.5446,
"consequence": 0.2685,
"evidence_need": 0.3259,
"family": "product_tradeoff",
"input_tokens": 5080,
"request_id": "synthetic-product_tradeoff-005832",
"reversibility": 0.8027,
"risk": 0.4352,
"summary": "Evaluate build-versus-buy... | [
{
"credits_per_call": 0.913,
"id": "worker_a",
"p95_latency_ms": 27191,
"pass_rate": 0.4645,
"provider_group": "provider_y",
"quality": 0.7284,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 4.171,
"id": "worker_b",
"p95_late... |
{
"max_calls": 3,
"max_credits": 19.37,
"max_latency_ms": 153628,
"minimum_quality": 0.9481
} | debugging | synthetic-debugging-017748 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 16.361,
"estimated_latency_ms": 126256,
"estimated_pass_rate": 0.7434,
"estimated_quality": 0.9304,
"finalizer": "worker_a",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7434,
"quality": 0.9304,
"quality_target_met": false,
"utility": 71.750379
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":16.361,"estimated_latency_ms":126256,"estimated_pass_rate":0.7434,"estimated_quality":0.9304,"finalizer":"worker_a","independent_verification":true,"max_calls":3,"max_credits":19.37,"max_latency_ms":153628,"policy":"stri... | {
"ambiguity": 0.5874,
"category": "coding",
"complexity": 0.6488,
"consequence": 0.3467,
"evidence_need": 0.7483,
"family": "debugging",
"input_tokens": 7400,
"request_id": "synthetic-debugging-017748",
"reversibility": 0.4293,
"risk": 0.5432,
"summary": "Diagnose an intermittent failure in usage... | [
{
"credits_per_call": 11.523,
"id": "worker_a",
"p95_latency_ms": 49725,
"pass_rate": 0.9719,
"provider_group": "provider_x",
"quality": 0.995,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 4.023,
"id": "worker_b",
... |
{
"max_calls": 3,
"max_credits": 19.842,
"max_latency_ms": 250026,
"minimum_quality": 0.8451
} | data_analysis | synthetic-data_analysis-015926 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 16.846,
"estimated_latency_ms": 182148,
"estimated_pass_rate": 0.824,
"estimated_quality": 0.9175,
"finalizer": "worker_c",
"independent_verification"... | {
"candidate_count": 15,
"pass_rate": 0.824,
"quality": 0.9175,
"quality_target_met": true,
"utility": 73.159651
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":16.846,"estimated_latency_ms":182148,"estimated_pass_rate":0.824,"estimated_quality":0.9175,"finalizer":"worker_c","independent_verification":true,"max_calls":3,"max_credits":19.842,"max_latency_ms":250026,"policy":"stri... | {
"ambiguity": 0.6787,
"category": "analysis",
"complexity": 0.6674,
"consequence": 0.6086,
"evidence_need": 0.6262,
"family": "data_analysis",
"input_tokens": 2520,
"request_id": "synthetic-data_analysis-015926",
"reversibility": 0.932,
"risk": 0.4747,
"summary": "Analyze a dataset about usage me... | [
{
"credits_per_call": 3.525,
"id": "worker_a",
"p95_latency_ms": 64218,
"pass_rate": 0.8271,
"provider_group": "provider_z",
"quality": 0.8818,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 0.816,
"id": "worker_b",
"p95_l... |
{
"max_calls": 2,
"max_credits": 17.149,
"max_latency_ms": 163165,
"minimum_quality": 0.7501
} | incident_response | synthetic-incident_response-013567 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 14.889,
"estimated_latency_ms": 135653,
"estimated_pass_rate": 0.8272,
"estimated_quality": 0.8849,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": ... | {
"candidate_count": 9,
"pass_rate": 0.8272,
"quality": 0.8849,
"quality_target_met": true,
"utility": 68.484494
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":14.889,"estimated_latency_ms":135653,"estimated_pass_rate":0.8272,"estimated_quality":0.8849,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":17.149,"max_latency_ms":163165,"policy":"balanced","reason_codes":["c... | {
"ambiguity": 0.5263,
"category": "operations",
"complexity": 0.8463,
"consequence": 0.8103,
"evidence_need": 0.9146,
"family": "incident_response",
"input_tokens": 6840,
"request_id": "synthetic-incident_response-013567",
"reversibility": 0.5393,
"risk": 0.6325,
"summary": "Plan immediate contai... | [
{
"credits_per_call": 1.149,
"id": "worker_a",
"p95_latency_ms": 27523,
"pass_rate": 0.5705,
"provider_group": "provider_x",
"quality": 0.7264,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 11.844,
"id": "worker_b",
"p95_lat... |
{
"max_calls": 2,
"max_credits": 11.804,
"max_latency_ms": 185035,
"minimum_quality": 0.792
} | incident_response | synthetic-incident_response-014047 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 4.547,
"estimated_latency_ms": 88333,
"estimated_pass_rate": 0.5407,
"estimated_quality": 0.7387,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 11... | {
"candidate_count": 4,
"pass_rate": 0.5407,
"quality": 0.7387,
"quality_target_met": false,
"utility": 53.781716
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":4.547,"estimated_latency_ms":88333,"estimated_pass_rate":0.5407,"estimated_quality":0.7387,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":11.804,"max_latency_ms":185035,"policy":"balanced","reason_codes":["bes... | {
"ambiguity": 0.7368,
"category": "operations",
"complexity": 0.7625,
"consequence": 0.6074,
"evidence_need": 0.5411,
"family": "incident_response",
"input_tokens": 6840,
"request_id": "synthetic-incident_response-014047",
"reversibility": 0.226,
"risk": 0.714,
"summary": "Plan immediate containm... | [
{
"credits_per_call": 3.694,
"id": "worker_a",
"p95_latency_ms": 65362,
"pass_rate": 0.7344,
"provider_group": "provider_x",
"quality": 0.8354,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 0.853,
"id": "worker_b",
"p95_l... |
{
"max_calls": 2,
"max_credits": 6.217,
"max_latency_ms": 119817,
"minimum_quality": 0.8622
} | feature_implementation | synthetic-feature_implementation-007763 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 4.043,
"estimated_latency_ms": 90627,
"estimated_pass_rate": 0.7329,
"estimated_quality": 0.7882,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 6.... | {
"candidate_count": 4,
"pass_rate": 0.7329,
"quality": 0.7882,
"quality_target_met": false,
"utility": 51.840797
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":4.043,"estimated_latency_ms":90627,"estimated_pass_rate":0.7329,"estimated_quality":0.7882,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":6.217,"max_latency_ms":119817,"policy":"balanced","reason_codes":["best... | {
"ambiguity": 0.2935,
"category": "coding",
"complexity": 0.746,
"consequence": 0.5926,
"evidence_need": 0.3458,
"family": "feature_implementation",
"input_tokens": 7520,
"request_id": "synthetic-feature_implementation-007763",
"reversibility": 0.7724,
"risk": 0.3663,
"summary": "Design an API fo... | [
{
"credits_per_call": 12.365,
"id": "worker_a",
"p95_latency_ms": 69425,
"pass_rate": 0.8865,
"provider_group": "provider_y",
"quality": 0.9308,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 0.863,
"id": "worker_b",... |
{
"max_calls": 3,
"max_credits": 16.641,
"max_latency_ms": 202534,
"minimum_quality": 0.8268
} | data_analysis | synthetic-data_analysis-006238 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 15.642,
"estimated_latency_ms": 180670,
"estimated_pass_rate": 0.7784,
"estimated_quality": 0.9316,
"finalizer": "worker_a",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7784,
"quality": 0.9316,
"quality_target_met": true,
"utility": 71.275402
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":15.642,"estimated_latency_ms":180670,"estimated_pass_rate":0.7784,"estimated_quality":0.9316,"finalizer":"worker_a","independent_verification":true,"max_calls":3,"max_credits":16.641,"max_latency_ms":202534,"policy":"str... | {
"ambiguity": 0.5189,
"category": "analysis",
"complexity": 0.6899,
"consequence": 0.2334,
"evidence_need": 0.7628,
"family": "data_analysis",
"input_tokens": 3800,
"request_id": "synthetic-data_analysis-006238",
"reversibility": 0.6584,
"risk": 0.4827,
"summary": "Interpret a noisy metric shift ... | [
{
"credits_per_call": 10.578,
"id": "worker_a",
"p95_latency_ms": 74961,
"pass_rate": 0.995,
"provider_group": "provider_z",
"quality": 0.995,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 4.003,
"id": "worker_b",
... |
{
"max_calls": 1,
"max_credits": 1.112,
"max_latency_ms": 36824,
"minimum_quality": 0.7686
} | content_revision | synthetic-content_revision-003970 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 0.925,
"estimated_latency_ms": 19948,
"estimated_pass_rate": 0.4225,
"estimated_quality": 0.7225,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 1.112,
"max_latency... | {
"candidate_count": 1,
"pass_rate": 0.4225,
"quality": 0.7225,
"quality_target_met": false,
"utility": 44.315755
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":0.925,"estimated_latency_ms":19948,"estimated_pass_rate":0.4225,"estimated_quality":0.7225,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":1.112,"max_latency_ms":36824,"policy":"lean","reason_codes":["best_feasible_be... | {
"ambiguity": 0.2174,
"category": "writing",
"complexity": 0.3803,
"consequence": 0.2051,
"evidence_need": 0.2643,
"family": "content_revision",
"input_tokens": 3440,
"request_id": "synthetic-content_revision-003970",
"reversibility": 0.9701,
"risk": 0.1425,
"summary": "Tighten a product explanat... | [
{
"credits_per_call": 3.759,
"id": "worker_a",
"p95_latency_ms": 65042,
"pass_rate": 0.8458,
"provider_group": "provider_z",
"quality": 0.9329,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 13.962,
"id": "worker_b",
"p95_... |
{
"max_calls": 3,
"max_credits": 23.386,
"max_latency_ms": 175178,
"minimum_quality": 0.8366
} | research_synthesis | synthetic-research_synthesis-006349 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 20.353,
"estimated_latency_ms": 140009,
"estimated_pass_rate": 0.6652,
"estimated_quality": 0.9167,
"finalizer": "worker_c",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.6652,
"quality": 0.9167,
"quality_target_met": true,
"utility": 72.195033
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":20.353,"estimated_latency_ms":140009,"estimated_pass_rate":0.6652,"estimated_quality":0.9167,"finalizer":"worker_c","independent_verification":true,"max_calls":3,"max_credits":23.386,"max_latency_ms":175178,"policy":"str... | {
"ambiguity": 0.7214,
"category": "research",
"complexity": 0.7691,
"consequence": 0.3161,
"evidence_need": 0.9047,
"family": "research_synthesis",
"input_tokens": 560,
"request_id": "synthetic-research_synthesis-006349",
"reversibility": 0.9078,
"risk": 0.4712,
"summary": "Review competing appro... | [
{
"credits_per_call": 4.257,
"id": "worker_a",
"p95_latency_ms": 50654,
"pass_rate": 0.8051,
"provider_group": "provider_z",
"quality": 0.8344,
"strengths": [
"reasoning",
"review",
"synthesis",
"writing"
]
},
{
"credits_per_call": 0.855,
"id": "wo... |
{
"max_calls": 1,
"max_credits": 2.317,
"max_latency_ms": 36652,
"minimum_quality": 0.7124
} | content_revision | synthetic-content_revision-004682 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 1.069,
"estimated_latency_ms": 21606,
"estimated_pass_rate": 0.4584,
"estimated_quality": 0.5905,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 2.317,
"max_latency... | {
"candidate_count": 1,
"pass_rate": 0.4584,
"quality": 0.5905,
"quality_target_met": false,
"utility": 36.152086
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":1.069,"estimated_latency_ms":21606,"estimated_pass_rate":0.4584,"estimated_quality":0.5905,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":2.317,"max_latency_ms":36652,"policy":"lean","reason_codes":["best_feasible_be... | {
"ambiguity": 0.3575,
"category": "writing",
"complexity": 0.4271,
"consequence": 0.1424,
"evidence_need": 0.1037,
"family": "content_revision",
"input_tokens": 2960,
"request_id": "synthetic-content_revision-004682",
"reversibility": 0.9971,
"risk": 0.1275,
"summary": "Tighten a product explanat... | [
{
"credits_per_call": 1.069,
"id": "worker_a",
"p95_latency_ms": 21606,
"pass_rate": 0.5442,
"provider_group": "provider_z",
"quality": 0.6209,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 3.064,
"id": "worker_b",
"p95_late... |
{
"max_calls": 3,
"max_credits": 9.784,
"max_latency_ms": 225880,
"minimum_quality": 0.687
} | feature_implementation | synthetic-feature_implementation-006083 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 5.121,
"estimated_latency_ms": 87145,
"estimated_pass_rate": 0.5506,
"estimated_quality": 0.7737,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 9.... | {
"candidate_count": 4,
"pass_rate": 0.5506,
"quality": 0.7737,
"quality_target_met": true,
"utility": 56.710006
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":5.121,"estimated_latency_ms":87145,"estimated_pass_rate":0.5506,"estimated_quality":0.7737,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":9.784,"max_latency_ms":225880,"policy":"balanced","reason_codes":["comp... | {
"ambiguity": 0.4018,
"category": "coding",
"complexity": 0.8278,
"consequence": 0.6003,
"evidence_need": 0.417,
"family": "feature_implementation",
"input_tokens": 4920,
"request_id": "synthetic-feature_implementation-006083",
"reversibility": 0.5119,
"risk": 0.5602,
"summary": "Design an API fo... | [
{
"credits_per_call": 13.625,
"id": "worker_a",
"p95_latency_ms": 69655,
"pass_rate": 0.9376,
"provider_group": "provider_y",
"quality": 0.9724,
"strengths": [
"coding",
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 0.982,
"... |
{
"max_calls": 2,
"max_credits": 1.663,
"max_latency_ms": 20359,
"minimum_quality": 0.7613
} | research_synthesis | synthetic-research_synthesis-006501 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 0.967,
"estimated_latency_ms": 18402,
"estimated_pass_rate": 0.3353,
"estimated_quality": 0.5093,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 1.663,
"max_latency... | {
"candidate_count": 1,
"pass_rate": 0.3353,
"quality": 0.5093,
"quality_target_met": false,
"utility": 20.566706
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":0.967,"estimated_latency_ms":18402,"estimated_pass_rate":0.3353,"estimated_quality":0.5093,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":1.663,"max_latency_ms":20359,"policy":"lean","reason_codes":["best_feasible_be... | {
"ambiguity": 0.7778,
"category": "research",
"complexity": 0.7675,
"consequence": 0.4002,
"evidence_need": 0.662,
"family": "research_synthesis",
"input_tokens": 3640,
"request_id": "synthetic-research_synthesis-006501",
"reversibility": 0.8705,
"risk": 0.4576,
"summary": "Compare evidence for p... | [
{
"credits_per_call": 3.725,
"id": "worker_a",
"p95_latency_ms": 66977,
"pass_rate": 0.8227,
"provider_group": "provider_z",
"quality": 0.8341,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 0.967,
"id": "worker_b",
"p95_l... |
{
"max_calls": 3,
"max_credits": 25.756,
"max_latency_ms": 170701,
"minimum_quality": 0.7862
} | incident_response | synthetic-incident_response-002351 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 18.9,
"estimated_latency_ms": 155016,
"estimated_pass_rate": 0.7337,
"estimated_quality": 0.9164,
"finalizer": "worker_a",
"independent_verification":... | {
"candidate_count": 15,
"pass_rate": 0.7337,
"quality": 0.9164,
"quality_target_met": true,
"utility": 80.966044
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":18.9,"estimated_latency_ms":155016,"estimated_pass_rate":0.7337,"estimated_quality":0.9164,"finalizer":"worker_a","independent_verification":true,"max_calls":3,"max_credits":25.756,"max_latency_ms":170701,"policy":"stric... | {
"ambiguity": 0.659,
"category": "operations",
"complexity": 0.8297,
"consequence": 0.6769,
"evidence_need": 0.7171,
"family": "incident_response",
"input_tokens": 1440,
"request_id": "synthetic-incident_response-002351",
"reversibility": 0.3066,
"risk": 0.907,
"summary": "Prioritize actions duri... | [
{
"credits_per_call": 14.964,
"id": "worker_a",
"p95_latency_ms": 63794,
"pass_rate": 0.9459,
"provider_group": "provider_x",
"quality": 0.9906,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 0.809,
"id": "worker_b",... |
{
"max_calls": 2,
"max_credits": 18.57,
"max_latency_ms": 148845,
"minimum_quality": 0.7474
} | incident_response | synthetic-incident_response-010975 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 17.555,
"estimated_latency_ms": 138255,
"estimated_pass_rate": 0.8718,
"estimated_quality": 0.8948,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": ... | {
"candidate_count": 9,
"pass_rate": 0.8718,
"quality": 0.8948,
"quality_target_met": true,
"utility": 68.792867
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":17.555,"estimated_latency_ms":138255,"estimated_pass_rate":0.8718,"estimated_quality":0.8948,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":18.57,"max_latency_ms":148845,"policy":"balanced","reason_codes":["co... | {
"ambiguity": 0.65,
"category": "operations",
"complexity": 0.7939,
"consequence": 0.7885,
"evidence_need": 0.5701,
"family": "incident_response",
"input_tokens": 7080,
"request_id": "synthetic-incident_response-010975",
"reversibility": 0.4442,
"risk": 0.7287,
"summary": "Review an outage involv... | [
{
"credits_per_call": 4.196,
"id": "worker_a",
"p95_latency_ms": 74182,
"pass_rate": 0.7596,
"provider_group": "provider_z",
"quality": 0.8235,
"strengths": [
"drafting",
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 1.177,
"id": "wor... |
{
"max_calls": 2,
"max_credits": 10.368,
"max_latency_ms": 224831,
"minimum_quality": 0.8119
} | product_tradeoff | synthetic-product_tradeoff-006656 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 5.143,
"estimated_latency_ms": 67689,
"estimated_pass_rate": 0.7456,
"estimated_quality": 0.8167,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 10... | {
"candidate_count": 4,
"pass_rate": 0.7456,
"quality": 0.8167,
"quality_target_met": true,
"utility": 59.121383
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":5.143,"estimated_latency_ms":67689,"estimated_pass_rate":0.7456,"estimated_quality":0.8167,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":10.368,"max_latency_ms":224831,"policy":"balanced","reason_codes":["ind... | {
"ambiguity": 0.633,
"category": "decision",
"complexity": 0.549,
"consequence": 0.2421,
"evidence_need": 0.3351,
"family": "product_tradeoff",
"input_tokens": 1920,
"request_id": "synthetic-product_tradeoff-006656",
"reversibility": 0.6187,
"risk": 0.2086,
"summary": "Choose an implementation st... | [
{
"credits_per_call": 1.178,
"id": "worker_a",
"p95_latency_ms": 22276,
"pass_rate": 0.5583,
"provider_group": "provider_z",
"quality": 0.7402,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 3.965,
"id": "worker_b",
"p95_late... |
{
"max_calls": 3,
"max_credits": 18.742,
"max_latency_ms": 173158,
"minimum_quality": 0.8119
} | research_synthesis | synthetic-research_synthesis-019477 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 15.299,
"estimated_latency_ms": 164919,
"estimated_pass_rate": 0.7272,
"estimated_quality": 0.9164,
"finalizer": "worker_a",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7272,
"quality": 0.9164,
"quality_target_met": true,
"utility": 74.00909
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":15.299,"estimated_latency_ms":164919,"estimated_pass_rate":0.7272,"estimated_quality":0.9164,"finalizer":"worker_a","independent_verification":true,"max_calls":3,"max_credits":18.742,"max_latency_ms":173158,"policy":"str... | {
"ambiguity": 0.6551,
"category": "research",
"complexity": 0.8304,
"consequence": 0.2762,
"evidence_need": 0.9689,
"family": "research_synthesis",
"input_tokens": 4280,
"request_id": "synthetic-research_synthesis-019477",
"reversibility": 0.7782,
"risk": 0.5102,
"summary": "Compare evidence for ... | [
{
"credits_per_call": 10.964,
"id": "worker_a",
"p95_latency_ms": 91069,
"pass_rate": 0.995,
"provider_group": "provider_x",
"quality": 0.9764,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 0.786,
"id": "worker_b",
... |
{
"max_calls": 1,
"max_credits": 2.059,
"max_latency_ms": 36870,
"minimum_quality": 0.5095
} | content_revision | synthetic-content_revision-001490 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 0.807,
"estimated_latency_ms": 27148,
"estimated_pass_rate": 0.32,
"estimated_quality": 0.6406,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 2.059,
"max_latency_m... | {
"candidate_count": 1,
"pass_rate": 0.32,
"quality": 0.6406,
"quality_target_met": true,
"utility": 41.02021
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":0.807,"estimated_latency_ms":27148,"estimated_pass_rate":0.32,"estimated_quality":0.6406,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":2.059,"max_latency_ms":36870,"policy":"lean","reason_codes":["single_call_suffic... | {
"ambiguity": 0.2536,
"category": "writing",
"complexity": 0.4639,
"consequence": 0.0804,
"evidence_need": 0.1092,
"family": "content_revision",
"input_tokens": 7480,
"request_id": "synthetic-content_revision-001490",
"reversibility": 0.8407,
"risk": 0.0717,
"summary": "Edit documentation for pro... | [
{
"credits_per_call": 2.899,
"id": "worker_a",
"p95_latency_ms": 46803,
"pass_rate": 0.9293,
"provider_group": "provider_y",
"quality": 0.9143,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 0.807,
"id": "worker_b",
"p95_l... |
{
"max_calls": 2,
"max_credits": 6.292,
"max_latency_ms": 178580,
"minimum_quality": 0.8967
} | feature_implementation | synthetic-feature_implementation-007643 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 4.464,
"estimated_latency_ms": 104261,
"estimated_pass_rate": 0.576,
"estimated_quality": 0.7708,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 6.... | {
"candidate_count": 4,
"pass_rate": 0.576,
"quality": 0.7708,
"quality_target_met": false,
"utility": 48.82328
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":4.464,"estimated_latency_ms":104261,"estimated_pass_rate":0.576,"estimated_quality":0.7708,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":6.292,"max_latency_ms":178580,"policy":"balanced","reason_codes":["best... | {
"ambiguity": 0.4244,
"category": "coding",
"complexity": 0.6741,
"consequence": 0.4087,
"evidence_need": 0.5318,
"family": "feature_implementation",
"input_tokens": 3760,
"request_id": "synthetic-feature_implementation-007643",
"reversibility": 0.7594,
"risk": 0.5598,
"summary": "Design an API f... | [
{
"credits_per_call": 13.457,
"id": "worker_a",
"p95_latency_ms": 74306,
"pass_rate": 0.9159,
"provider_group": "provider_x",
"quality": 0.932,
"strengths": [
"coding",
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 0.839,
"i... |
{
"max_calls": 3,
"max_credits": 3.27,
"max_latency_ms": 44641,
"minimum_quality": 0.4826
} | debugging | synthetic-debugging-007572 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 0.944,
"estimated_latency_ms": 27087,
"estimated_pass_rate": 0.3713,
"estimated_quality": 0.6704,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 3.27,
"max_latency_... | {
"candidate_count": 1,
"pass_rate": 0.3713,
"quality": 0.6704,
"quality_target_met": true,
"utility": 45.353422
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":0.944,"estimated_latency_ms":27087,"estimated_pass_rate":0.3713,"estimated_quality":0.6704,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":3.27,"max_latency_ms":44641,"policy":"lean","reason_codes":["complex_task"],"r... | {
"ambiguity": 0.7225,
"category": "coding",
"complexity": 0.5887,
"consequence": 0.5084,
"evidence_need": 0.3792,
"family": "debugging",
"input_tokens": 5640,
"request_id": "synthetic-debugging-007572",
"reversibility": 0.4891,
"risk": 0.5479,
"summary": "Find the root cause of a concurrency-sens... | [
{
"credits_per_call": 10.003,
"id": "worker_a",
"p95_latency_ms": 50845,
"pass_rate": 0.9108,
"provider_group": "provider_x",
"quality": 0.927,
"strengths": [
"coding",
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 0.944,
"i... |
{
"max_calls": 2,
"max_credits": 20.117,
"max_latency_ms": 128514,
"minimum_quality": 0.7576
} | product_tradeoff | synthetic-product_tradeoff-011096 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 16.413,
"estimated_latency_ms": 99081,
"estimated_pass_rate": 0.8907,
"estimated_quality": 0.9002,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 2... | {
"candidate_count": 9,
"pass_rate": 0.8907,
"quality": 0.9002,
"quality_target_met": true,
"utility": 66.820588
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":16.413,"estimated_latency_ms":99081,"estimated_pass_rate":0.8907,"estimated_quality":0.9002,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":20.117,"max_latency_ms":128514,"policy":"balanced","reason_codes":["co... | {
"ambiguity": 0.7221,
"category": "decision",
"complexity": 0.6254,
"consequence": 0.2904,
"evidence_need": 0.6437,
"family": "product_tradeoff",
"input_tokens": 3400,
"request_id": "synthetic-product_tradeoff-011096",
"reversibility": 0.6178,
"risk": 0.411,
"summary": "Evaluate build-versus-buy ... | [
{
"credits_per_call": 13.04,
"id": "worker_a",
"p95_latency_ms": 54244,
"pass_rate": 0.995,
"provider_group": "provider_z",
"quality": 0.9388,
"strengths": [
"drafting",
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 3.373,
"... |
{
"max_calls": 3,
"max_credits": 21.173,
"max_latency_ms": 178430,
"minimum_quality": 0.9111
} | research_synthesis | synthetic-research_synthesis-018013 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 17.266,
"estimated_latency_ms": 150931,
"estimated_pass_rate": 0.6951,
"estimated_quality": 0.9157,
"finalizer": "worker_b",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.6951,
"quality": 0.9157,
"quality_target_met": true,
"utility": 73.983658
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":17.266,"estimated_latency_ms":150931,"estimated_pass_rate":0.6951,"estimated_quality":0.9157,"finalizer":"worker_b","independent_verification":true,"max_calls":3,"max_credits":21.173,"max_latency_ms":178430,"policy":"str... | {
"ambiguity": 0.794,
"category": "research",
"complexity": 0.8686,
"consequence": 0.5417,
"evidence_need": 0.7837,
"family": "research_synthesis",
"input_tokens": 4880,
"request_id": "synthetic-research_synthesis-018013",
"reversibility": 0.8846,
"risk": 0.3651,
"summary": "Compare evidence for d... | [
{
"credits_per_call": 1.132,
"id": "worker_a",
"p95_latency_ms": 32556,
"pass_rate": 0.428,
"provider_group": "provider_y",
"quality": 0.6759,
"strengths": [
"coding",
"drafting",
"speed",
"synthesis"
]
},
{
"credits_per_call": 13.313,
"id": "worke... |
{
"max_calls": 2,
"max_credits": 2.512,
"max_latency_ms": 24609,
"minimum_quality": 0.6581
} | data_analysis | synthetic-data_analysis-003926 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 1.169,
"estimated_latency_ms": 21863,
"estimated_pass_rate": 0.3185,
"estimated_quality": 0.6671,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 2.512,
"max_latency... | {
"candidate_count": 1,
"pass_rate": 0.3185,
"quality": 0.6671,
"quality_target_met": true,
"utility": 41.43027
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":1.169,"estimated_latency_ms":21863,"estimated_pass_rate":0.3185,"estimated_quality":0.6671,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":2.512,"max_latency_ms":24609,"policy":"lean","reason_codes":["latency_constrai... | {
"ambiguity": 0.2902,
"category": "analysis",
"complexity": 0.6032,
"consequence": 0.6426,
"evidence_need": 0.4454,
"family": "data_analysis",
"input_tokens": 2920,
"request_id": "synthetic-data_analysis-003926",
"reversibility": 0.8392,
"risk": 0.217,
"summary": "Analyze a dataset about database... | [
{
"credits_per_call": 3.43,
"id": "worker_a",
"p95_latency_ms": 48168,
"pass_rate": 0.8462,
"provider_group": "provider_x",
"quality": 0.9588,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 14.253,
"id": "worker_b",
"p95_l... |
{
"max_calls": 1,
"max_credits": 3.948,
"max_latency_ms": 29048,
"minimum_quality": 0.5628
} | data_analysis | synthetic-data_analysis-007702 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 0.79,
"estimated_latency_ms": 18677,
"estimated_pass_rate": 0.3284,
"estimated_quality": 0.6868,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 3.948,
"max_latency_... | {
"candidate_count": 1,
"pass_rate": 0.3284,
"quality": 0.6868,
"quality_target_met": true,
"utility": 45.840725
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":0.79,"estimated_latency_ms":18677,"estimated_pass_rate":0.3284,"estimated_quality":0.6868,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":3.948,"max_latency_ms":29048,"policy":"lean","reason_codes":["independent_evide... | {
"ambiguity": 0.62,
"category": "analysis",
"complexity": 0.3531,
"consequence": 0.2265,
"evidence_need": 0.7652,
"family": "data_analysis",
"input_tokens": 320,
"request_id": "synthetic-data_analysis-007702",
"reversibility": 0.9242,
"risk": 0.3017,
"summary": "Analyze a dataset about private cr... | [
{
"credits_per_call": 11.525,
"id": "worker_a",
"p95_latency_ms": 84066,
"pass_rate": 0.995,
"provider_group": "provider_x",
"quality": 0.995,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 3.319,
"id": "worker_b",
... |
{
"max_calls": 3,
"max_credits": 19.683,
"max_latency_ms": 202961,
"minimum_quality": 0.8931
} | research_synthesis | synthetic-research_synthesis-011653 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 14.554,
"estimated_latency_ms": 175648,
"estimated_pass_rate": 0.7253,
"estimated_quality": 0.8981,
"finalizer": "worker_c",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7253,
"quality": 0.8981,
"quality_target_met": true,
"utility": 71.252561
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":14.554,"estimated_latency_ms":175648,"estimated_pass_rate":0.7253,"estimated_quality":0.8981,"finalizer":"worker_c","independent_verification":true,"max_calls":3,"max_credits":19.683,"max_latency_ms":202961,"policy":"str... | {
"ambiguity": 0.6556,
"category": "research",
"complexity": 0.8673,
"consequence": 0.474,
"evidence_need": 0.6089,
"family": "research_synthesis",
"input_tokens": 2440,
"request_id": "synthetic-research_synthesis-011653",
"reversibility": 0.8074,
"risk": 0.3176,
"summary": "Produce a decision mem... | [
{
"credits_per_call": 3.413,
"id": "worker_a",
"p95_latency_ms": 75063,
"pass_rate": 0.8117,
"provider_group": "provider_x",
"quality": 0.8338,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 1.064,
"id": "worker_b",
"p95_l... |
{
"max_calls": 1,
"max_credits": 2.077,
"max_latency_ms": 50379,
"minimum_quality": 0.7666
} | data_analysis | synthetic-data_analysis-003622 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 1.085,
"estimated_latency_ms": 29356,
"estimated_pass_rate": 0.3088,
"estimated_quality": 0.4881,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 2.077,
"max_latency... | {
"candidate_count": 1,
"pass_rate": 0.3088,
"quality": 0.4881,
"quality_target_met": false,
"utility": 19.38872
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":1.085,"estimated_latency_ms":29356,"estimated_pass_rate":0.3088,"estimated_quality":0.4881,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":2.077,"max_latency_ms":50379,"policy":"lean","reason_codes":["best_feasible_be... | {
"ambiguity": 0.6803,
"category": "analysis",
"complexity": 0.723,
"consequence": 0.5423,
"evidence_need": 0.4978,
"family": "data_analysis",
"input_tokens": 5040,
"request_id": "synthetic-data_analysis-003622",
"reversibility": 0.9091,
"risk": 0.4232,
"summary": "Compare two experiments on obser... | [
{
"credits_per_call": 2.885,
"id": "worker_a",
"p95_latency_ms": 68341,
"pass_rate": 0.8046,
"provider_group": "provider_y",
"quality": 0.8943,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 11.847,
"id": "worker_b",
"p95_... |
{
"max_calls": 2,
"max_credits": 2.244,
"max_latency_ms": 48728,
"minimum_quality": 0.7468
} | content_revision | synthetic-content_revision-008410 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 0.851,
"estimated_latency_ms": 24881,
"estimated_pass_rate": 0.4057,
"estimated_quality": 0.6749,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 2.244,
"max_latency... | {
"candidate_count": 1,
"pass_rate": 0.4057,
"quality": 0.6749,
"quality_target_met": false,
"utility": 43.202721
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":0.851,"estimated_latency_ms":24881,"estimated_pass_rate":0.4057,"estimated_quality":0.6749,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":2.244,"max_latency_ms":48728,"policy":"lean","reason_codes":["best_feasible_be... | {
"ambiguity": 0.3623,
"category": "writing",
"complexity": 0.2118,
"consequence": 0.2565,
"evidence_need": 0.0722,
"family": "content_revision",
"input_tokens": 5400,
"request_id": "synthetic-content_revision-008410",
"reversibility": 0.8469,
"risk": 0.1823,
"summary": "Tighten a product explanat... | [
{
"credits_per_call": 3.131,
"id": "worker_a",
"p95_latency_ms": 73835,
"pass_rate": 0.8093,
"provider_group": "provider_y",
"quality": 0.8711,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 11.026,
"id": "worker_b",
"p95_... |
{
"max_calls": 1,
"max_credits": 1.547,
"max_latency_ms": 37746,
"minimum_quality": 0.6503
} | research_synthesis | synthetic-research_synthesis-005157 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 1.089,
"estimated_latency_ms": 18924,
"estimated_pass_rate": 0.3499,
"estimated_quality": 0.51,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 1.547,
"max_latency_m... | {
"candidate_count": 1,
"pass_rate": 0.3499,
"quality": 0.51,
"quality_target_met": false,
"utility": 26.591418
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":1.089,"estimated_latency_ms":18924,"estimated_pass_rate":0.3499,"estimated_quality":0.51,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":1.547,"max_latency_ms":37746,"policy":"lean","reason_codes":["best_feasible_belo... | {
"ambiguity": 0.8751,
"category": "research",
"complexity": 0.7591,
"consequence": 0.5676,
"evidence_need": 0.6258,
"family": "research_synthesis",
"input_tokens": 2840,
"request_id": "synthetic-research_synthesis-005157",
"reversibility": 0.884,
"risk": 0.3431,
"summary": "Review competing appro... | [
{
"credits_per_call": 3.188,
"id": "worker_a",
"p95_latency_ms": 57176,
"pass_rate": 0.755,
"provider_group": "provider_x",
"quality": 0.9127,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 1.089,
"id": "worker_b",
"p95_la... |
{
"max_calls": 2,
"max_credits": 14.366,
"max_latency_ms": 254340,
"minimum_quality": 0.6988
} | product_tradeoff | synthetic-product_tradeoff-001008 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 13.158,
"estimated_latency_ms": 145598,
"estimated_pass_rate": 0.8723,
"estimated_quality": 0.9266,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": ... | {
"candidate_count": 9,
"pass_rate": 0.8723,
"quality": 0.9266,
"quality_target_met": true,
"utility": 68.307902
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":13.158,"estimated_latency_ms":145598,"estimated_pass_rate":0.8723,"estimated_quality":0.9266,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":14.366,"max_latency_ms":254340,"policy":"balanced","reason_codes":["c... | {
"ambiguity": 0.6796,
"category": "decision",
"complexity": 0.6253,
"consequence": 0.4291,
"evidence_need": 0.484,
"family": "product_tradeoff",
"input_tokens": 3600,
"request_id": "synthetic-product_tradeoff-001008",
"reversibility": 0.8106,
"risk": 0.333,
"summary": "Evaluate build-versus-buy o... | [
{
"credits_per_call": 9.951,
"id": "worker_a",
"p95_latency_ms": 89801,
"pass_rate": 0.969,
"provider_group": "provider_z",
"quality": 0.995,
"strengths": [
"drafting",
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 3.207,
"i... |
{
"max_calls": 2,
"max_credits": 22.133,
"max_latency_ms": 152133,
"minimum_quality": 0.7903
} | incident_response | synthetic-incident_response-008359 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 17.661,
"estimated_latency_ms": 109194,
"estimated_pass_rate": 0.8632,
"estimated_quality": 0.9156,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": ... | {
"candidate_count": 9,
"pass_rate": 0.8632,
"quality": 0.9156,
"quality_target_met": true,
"utility": 70.513457
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":17.661,"estimated_latency_ms":109194,"estimated_pass_rate":0.8632,"estimated_quality":0.9156,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":22.133,"max_latency_ms":152133,"policy":"balanced","reason_codes":["c... | {
"ambiguity": 0.5382,
"category": "operations",
"complexity": 0.84,
"consequence": 0.6107,
"evidence_need": 0.7626,
"family": "incident_response",
"input_tokens": 2200,
"request_id": "synthetic-incident_response-008359",
"reversibility": 0.2836,
"risk": 0.6711,
"summary": "Prioritize actions duri... | [
{
"credits_per_call": 1.177,
"id": "worker_a",
"p95_latency_ms": 23993,
"pass_rate": 0.5191,
"provider_group": "provider_y",
"quality": 0.7774,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 3.201,
"id": "worker_b",
"p95_late... |
{
"max_calls": 3,
"max_credits": 9.52,
"max_latency_ms": 103782,
"minimum_quality": 0.7167
} | feature_implementation | synthetic-feature_implementation-005187 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 4.121,
"estimated_latency_ms": 81844,
"estimated_pass_rate": 0.6158,
"estimated_quality": 0.7717,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 9.... | {
"candidate_count": 4,
"pass_rate": 0.6158,
"quality": 0.7717,
"quality_target_met": true,
"utility": 56.4529
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":4.121,"estimated_latency_ms":81844,"estimated_pass_rate":0.6158,"estimated_quality":0.7717,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":9.52,"max_latency_ms":103782,"policy":"balanced","reason_codes":["compl... | {
"ambiguity": 0.2679,
"category": "coding",
"complexity": 0.7622,
"consequence": 0.6207,
"evidence_need": 0.4092,
"family": "feature_implementation",
"input_tokens": 7720,
"request_id": "synthetic-feature_implementation-005187",
"reversibility": 0.5761,
"risk": 0.4614,
"summary": "Implement datab... | [
{
"credits_per_call": 1.056,
"id": "worker_a",
"p95_latency_ms": 19855,
"pass_rate": 0.5559,
"provider_group": "provider_z",
"quality": 0.745,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 12.629,
"id": "worker_b",
"p95_late... |
{
"max_calls": 1,
"max_credits": 3.955,
"max_latency_ms": 44896,
"minimum_quality": 0.529
} | research_synthesis | synthetic-research_synthesis-006565 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 0.801,
"estimated_latency_ms": 31800,
"estimated_pass_rate": 0.3421,
"estimated_quality": 0.5226,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 3.955,
"max_latency... | {
"candidate_count": 1,
"pass_rate": 0.3421,
"quality": 0.5226,
"quality_target_met": false,
"utility": 35.395635
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":0.801,"estimated_latency_ms":31800,"estimated_pass_rate":0.3421,"estimated_quality":0.5226,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":3.955,"max_latency_ms":44896,"policy":"lean","reason_codes":["best_feasible_be... | {
"ambiguity": 0.8281,
"category": "research",
"complexity": 0.4812,
"consequence": 0.4834,
"evidence_need": 0.8193,
"family": "research_synthesis",
"input_tokens": 5800,
"request_id": "synthetic-research_synthesis-006565",
"reversibility": 0.8829,
"risk": 0.3886,
"summary": "Review competing appr... | [
{
"credits_per_call": 0.801,
"id": "worker_a",
"p95_latency_ms": 31800,
"pass_rate": 0.4907,
"provider_group": "provider_z",
"quality": 0.6132,
"strengths": [
"coding",
"drafting",
"speed",
"synthesis"
]
},
{
"credits_per_call": 2.925,
"id": "worke... |
{
"max_calls": 2,
"max_credits": 17.783,
"max_latency_ms": 170306,
"minimum_quality": 0.7897
} | incident_response | synthetic-incident_response-013759 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 15.871,
"estimated_latency_ms": 128118,
"estimated_pass_rate": 0.9127,
"estimated_quality": 0.9195,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": ... | {
"candidate_count": 9,
"pass_rate": 0.9127,
"quality": 0.9195,
"quality_target_met": true,
"utility": 70.023296
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":15.871,"estimated_latency_ms":128118,"estimated_pass_rate":0.9127,"estimated_quality":0.9195,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":17.783,"max_latency_ms":170306,"policy":"balanced","reason_codes":["c... | {
"ambiguity": 0.395,
"category": "operations",
"complexity": 0.5074,
"consequence": 0.5653,
"evidence_need": 0.6005,
"family": "incident_response",
"input_tokens": 7160,
"request_id": "synthetic-incident_response-013759",
"reversibility": 0.232,
"risk": 0.567,
"summary": "Prioritize actions durin... | [
{
"credits_per_call": 12.02,
"id": "worker_a",
"p95_latency_ms": 63014,
"pass_rate": 0.9901,
"provider_group": "provider_z",
"quality": 0.9303,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 1.008,
"id": "worker_b",
... |
{
"max_calls": 2,
"max_credits": 11.361,
"max_latency_ms": 139316,
"minimum_quality": 0.8553
} | product_tradeoff | synthetic-product_tradeoff-003784 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 4.527,
"estimated_latency_ms": 87407,
"estimated_pass_rate": 0.5246,
"estimated_quality": 0.7746,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 11... | {
"candidate_count": 4,
"pass_rate": 0.5246,
"quality": 0.7746,
"quality_target_met": false,
"utility": 50.901196
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":4.527,"estimated_latency_ms":87407,"estimated_pass_rate":0.5246,"estimated_quality":0.7746,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":11.361,"max_latency_ms":139316,"policy":"balanced","reason_codes":["bes... | {
"ambiguity": 0.6113,
"category": "decision",
"complexity": 0.6645,
"consequence": 0.6118,
"evidence_need": 0.6098,
"family": "product_tradeoff",
"input_tokens": 3680,
"request_id": "synthetic-product_tradeoff-003784",
"reversibility": 0.7944,
"risk": 0.2168,
"summary": "Choose an implementation ... | [
{
"credits_per_call": 3.606,
"id": "worker_a",
"p95_latency_ms": 60454,
"pass_rate": 0.759,
"provider_group": "provider_x",
"quality": 0.8211,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 0.921,
"id": "worker_b",
"p95_la... |
{
"max_calls": 2,
"max_credits": 9.888,
"max_latency_ms": 228126,
"minimum_quality": 0.7979
} | feature_implementation | synthetic-feature_implementation-010403 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 4.232,
"estimated_latency_ms": 76123,
"estimated_pass_rate": 0.5271,
"estimated_quality": 0.7467,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 9.... | {
"candidate_count": 4,
"pass_rate": 0.5271,
"quality": 0.7467,
"quality_target_met": false,
"utility": 51.304349
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":4.232,"estimated_latency_ms":76123,"estimated_pass_rate":0.5271,"estimated_quality":0.7467,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":9.888,"max_latency_ms":228126,"policy":"balanced","reason_codes":["best... | {
"ambiguity": 0.2261,
"category": "coding",
"complexity": 0.8441,
"consequence": 0.3071,
"evidence_need": 0.5043,
"family": "feature_implementation",
"input_tokens": 6600,
"request_id": "synthetic-feature_implementation-010403",
"reversibility": 0.5781,
"risk": 0.361,
"summary": "Patch a producti... | [
{
"credits_per_call": 14.81,
"id": "worker_a",
"p95_latency_ms": 86014,
"pass_rate": 0.9452,
"provider_group": "provider_y",
"quality": 0.938,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 3.312,
"id": "worker_b",
... |
{
"max_calls": 3,
"max_credits": 15.578,
"max_latency_ms": 163653,
"minimum_quality": 0.8371
} | incident_response | synthetic-incident_response-013039 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 15.001,
"estimated_latency_ms": 147190,
"estimated_pass_rate": 0.7638,
"estimated_quality": 0.9081,
"finalizer": "worker_b",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7638,
"quality": 0.9081,
"quality_target_met": true,
"utility": 77.322395
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":15.001,"estimated_latency_ms":147190,"estimated_pass_rate":0.7638,"estimated_quality":0.9081,"finalizer":"worker_b","independent_verification":true,"max_calls":3,"max_credits":15.578,"max_latency_ms":163653,"policy":"str... | {
"ambiguity": 0.7502,
"category": "operations",
"complexity": 0.8262,
"consequence": 0.5697,
"evidence_need": 0.8101,
"family": "incident_response",
"input_tokens": 3400,
"request_id": "synthetic-incident_response-013039",
"reversibility": 0.395,
"risk": 0.6982,
"summary": "Prioritize actions dur... | [
{
"credits_per_call": 1.103,
"id": "worker_a",
"p95_latency_ms": 31182,
"pass_rate": 0.5597,
"provider_group": "provider_y",
"quality": 0.7006,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 11.113,
"id": "worker_b",
"p95_lat... |
{
"max_calls": 2,
"max_credits": 11.781,
"max_latency_ms": 226180,
"minimum_quality": 0.8432
} | feature_implementation | synthetic-feature_implementation-008699 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 4.092,
"estimated_latency_ms": 81377,
"estimated_pass_rate": 0.7307,
"estimated_quality": 0.7945,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 11... | {
"candidate_count": 4,
"pass_rate": 0.7307,
"quality": 0.7945,
"quality_target_met": false,
"utility": 56.616798
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":4.092,"estimated_latency_ms":81377,"estimated_pass_rate":0.7307,"estimated_quality":0.7945,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":11.781,"max_latency_ms":226180,"policy":"balanced","reason_codes":["bes... | {
"ambiguity": 0.2046,
"category": "coding",
"complexity": 0.7649,
"consequence": 0.4131,
"evidence_need": 0.5224,
"family": "feature_implementation",
"input_tokens": 4880,
"request_id": "synthetic-feature_implementation-008699",
"reversibility": 0.4941,
"risk": 0.2854,
"summary": "Implement backg... | [
{
"credits_per_call": 12.715,
"id": "worker_a",
"p95_latency_ms": 79129,
"pass_rate": 0.8948,
"provider_group": "provider_x",
"quality": 0.9631,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 1.053,
"id": "worker_b",... |
{
"max_calls": 1,
"max_credits": 2.059,
"max_latency_ms": 24861,
"minimum_quality": 0.7528
} | debugging | synthetic-debugging-007612 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 0.842,
"estimated_latency_ms": 20728,
"estimated_pass_rate": 0.3514,
"estimated_quality": 0.6435,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 2.059,
"max_latency... | {
"candidate_count": 1,
"pass_rate": 0.3514,
"quality": 0.6435,
"quality_target_met": false,
"utility": 36.774145
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":0.842,"estimated_latency_ms":20728,"estimated_pass_rate":0.3514,"estimated_quality":0.6435,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":2.059,"max_latency_ms":24861,"policy":"lean","reason_codes":["best_feasible_be... | {
"ambiguity": 0.6994,
"category": "coding",
"complexity": 0.8081,
"consequence": 0.3612,
"evidence_need": 0.5945,
"family": "debugging",
"input_tokens": 7600,
"request_id": "synthetic-debugging-007612",
"reversibility": 0.5694,
"risk": 0.5592,
"summary": "Find the root cause of a concurrency-sens... | [
{
"credits_per_call": 3.825,
"id": "worker_a",
"p95_latency_ms": 69177,
"pass_rate": 0.8406,
"provider_group": "provider_x",
"quality": 0.8962,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 0.842,
"id": "worker_b",
"p95_l... |
{
"max_calls": 2,
"max_credits": 9.921,
"max_latency_ms": 117287,
"minimum_quality": 0.7826
} | incident_response | synthetic-incident_response-000679 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 4.828,
"estimated_latency_ms": 105757,
"estimated_pass_rate": 0.5642,
"estimated_quality": 0.7821,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 9... | {
"candidate_count": 4,
"pass_rate": 0.5642,
"quality": 0.7821,
"quality_target_met": false,
"utility": 58.094685
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":4.828,"estimated_latency_ms":105757,"estimated_pass_rate":0.5642,"estimated_quality":0.7821,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":9.921,"max_latency_ms":117287,"policy":"balanced","reason_codes":["bes... | {
"ambiguity": 0.7911,
"category": "operations",
"complexity": 0.8286,
"consequence": 0.8502,
"evidence_need": 0.6558,
"family": "incident_response",
"input_tokens": 1240,
"request_id": "synthetic-incident_response-000679",
"reversibility": 0.1691,
"risk": 0.651,
"summary": "Review an outage invol... | [
{
"credits_per_call": 1.075,
"id": "worker_a",
"p95_latency_ms": 24410,
"pass_rate": 0.4762,
"provider_group": "provider_y",
"quality": 0.753,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 13.028,
"id": "worker_b",
"p95_late... |
{
"max_calls": 2,
"max_credits": 3.847,
"max_latency_ms": 42129,
"minimum_quality": 0.699
} | content_revision | synthetic-content_revision-000858 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 1.151,
"estimated_latency_ms": 25805,
"estimated_pass_rate": 0.3484,
"estimated_quality": 0.73,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 3.847,
"max_latency_m... | {
"candidate_count": 1,
"pass_rate": 0.3484,
"quality": 0.73,
"quality_target_met": true,
"utility": 48.422485
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":1.151,"estimated_latency_ms":25805,"estimated_pass_rate":0.3484,"estimated_quality":0.73,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":3.847,"max_latency_ms":42129,"policy":"lean","reason_codes":["single_call_suffic... | {
"ambiguity": 0.334,
"category": "writing",
"complexity": 0.2115,
"consequence": 0.1154,
"evidence_need": 0.2964,
"family": "content_revision",
"input_tokens": 5880,
"request_id": "synthetic-content_revision-000858",
"reversibility": 0.898,
"risk": 0.1566,
"summary": "Tighten a product explanatio... | [
{
"credits_per_call": 13.813,
"id": "worker_a",
"p95_latency_ms": 51143,
"pass_rate": 0.995,
"provider_group": "provider_z",
"quality": 0.9634,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 3.186,
"id": "worker_b",
... |
{
"max_calls": 3,
"max_credits": 19.838,
"max_latency_ms": 172758,
"minimum_quality": 0.8711
} | research_synthesis | synthetic-research_synthesis-012509 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 15.307,
"estimated_latency_ms": 152348,
"estimated_pass_rate": 0.6614,
"estimated_quality": 0.9023,
"finalizer": "worker_b",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.6614,
"quality": 0.9023,
"quality_target_met": true,
"utility": 69.373607
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":15.307,"estimated_latency_ms":152348,"estimated_pass_rate":0.6614,"estimated_quality":0.9023,"finalizer":"worker_b","independent_verification":true,"max_calls":3,"max_credits":19.838,"max_latency_ms":172758,"policy":"str... | {
"ambiguity": 0.5242,
"category": "research",
"complexity": 0.8653,
"consequence": 0.4543,
"evidence_need": 0.9888,
"family": "research_synthesis",
"input_tokens": 6720,
"request_id": "synthetic-research_synthesis-012509",
"reversibility": 0.9478,
"risk": 0.2034,
"summary": "Review competing appr... | [
{
"credits_per_call": 1.117,
"id": "worker_a",
"p95_latency_ms": 31707,
"pass_rate": 0.4147,
"provider_group": "provider_y",
"quality": 0.6109,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 10.694,
"id": "worker_b",
"p95_lat... |
{
"max_calls": 3,
"max_credits": 19.566,
"max_latency_ms": 176626,
"minimum_quality": 0.8051
} | incident_response | synthetic-incident_response-012063 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 15.726,
"estimated_latency_ms": 160056,
"estimated_pass_rate": 0.7163,
"estimated_quality": 0.8872,
"finalizer": "worker_b",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7163,
"quality": 0.8872,
"quality_target_met": true,
"utility": 78.361322
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":15.726,"estimated_latency_ms":160056,"estimated_pass_rate":0.7163,"estimated_quality":0.8872,"finalizer":"worker_b","independent_verification":true,"max_calls":3,"max_credits":19.566,"max_latency_ms":176626,"policy":"str... | {
"ambiguity": 0.7188,
"category": "operations",
"complexity": 0.8653,
"consequence": 0.9077,
"evidence_need": 0.8233,
"family": "incident_response",
"input_tokens": 7480,
"request_id": "synthetic-incident_response-012063",
"reversibility": 0.2412,
"risk": 0.6617,
"summary": "Plan immediate contai... | [
{
"credits_per_call": 2.997,
"id": "worker_a",
"p95_latency_ms": 47347,
"pass_rate": 0.7627,
"provider_group": "provider_x",
"quality": 0.9049,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 11.754,
"id": "worker_b",
"p95_... |
{
"max_calls": 3,
"max_credits": 24.275,
"max_latency_ms": 154894,
"minimum_quality": 0.8054
} | research_synthesis | synthetic-research_synthesis-009469 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 18.845,
"estimated_latency_ms": 134024,
"estimated_pass_rate": 0.7328,
"estimated_quality": 0.9149,
"finalizer": "worker_b",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7328,
"quality": 0.9149,
"quality_target_met": true,
"utility": 70.737548
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":18.845,"estimated_latency_ms":134024,"estimated_pass_rate":0.7328,"estimated_quality":0.9149,"finalizer":"worker_b","independent_verification":true,"max_calls":3,"max_credits":24.275,"max_latency_ms":154894,"policy":"str... | {
"ambiguity": 0.8464,
"category": "research",
"complexity": 0.5383,
"consequence": 0.2979,
"evidence_need": 0.7601,
"family": "research_synthesis",
"input_tokens": 1280,
"request_id": "synthetic-research_synthesis-009469",
"reversibility": 0.7711,
"risk": 0.4091,
"summary": "Review competing appr... | [
{
"credits_per_call": 3.653,
"id": "worker_a",
"p95_latency_ms": 55883,
"pass_rate": 0.8602,
"provider_group": "provider_x",
"quality": 0.8728,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 14.37,
"id": "worker_b",
"p95_l... |
{
"max_calls": 3,
"max_credits": 7.476,
"max_latency_ms": 218870,
"minimum_quality": 0.8227
} | incident_response | synthetic-incident_response-006807 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 4.704,
"estimated_latency_ms": 111801,
"estimated_pass_rate": 0.696,
"estimated_quality": 0.8123,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 7.... | {
"candidate_count": 4,
"pass_rate": 0.696,
"quality": 0.8123,
"quality_target_met": false,
"utility": 64.315594
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":4.704,"estimated_latency_ms":111801,"estimated_pass_rate":0.696,"estimated_quality":0.8123,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":7.476,"max_latency_ms":218870,"policy":"balanced","reason_codes":["best... | {
"ambiguity": 0.59,
"category": "operations",
"complexity": 0.8065,
"consequence": 0.7228,
"evidence_need": 0.7248,
"family": "incident_response",
"input_tokens": 4200,
"request_id": "synthetic-incident_response-006807",
"reversibility": 0.3197,
"risk": 0.8665,
"summary": "Plan immediate containm... | [
{
"credits_per_call": 10.065,
"id": "worker_a",
"p95_latency_ms": 68063,
"pass_rate": 0.9623,
"provider_group": "provider_y",
"quality": 0.9352,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 3.92,
"id": "worker_b",
... |
{
"max_calls": 2,
"max_credits": 13.885,
"max_latency_ms": 217640,
"minimum_quality": 0.7278
} | incident_response | synthetic-incident_response-004911 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 12.955,
"estimated_latency_ms": 111856,
"estimated_pass_rate": 0.7408,
"estimated_quality": 0.8857,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": ... | {
"candidate_count": 7,
"pass_rate": 0.7408,
"quality": 0.8857,
"quality_target_met": true,
"utility": 71.346581
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":12.955,"estimated_latency_ms":111856,"estimated_pass_rate":0.7408,"estimated_quality":0.8857,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":13.885,"max_latency_ms":217640,"policy":"balanced","reason_codes":["c... | {
"ambiguity": 0.4418,
"category": "operations",
"complexity": 0.8733,
"consequence": 0.9285,
"evidence_need": 0.9405,
"family": "incident_response",
"input_tokens": 2560,
"request_id": "synthetic-incident_response-004911",
"reversibility": 0.5492,
"risk": 0.8981,
"summary": "Plan immediate contai... | [
{
"credits_per_call": 11.764,
"id": "worker_a",
"p95_latency_ms": 89105,
"pass_rate": 0.9012,
"provider_group": "provider_x",
"quality": 0.995,
"strengths": [
"drafting",
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 1.191,
... |
{
"max_calls": 3,
"max_credits": 20.81,
"max_latency_ms": 185976,
"minimum_quality": 0.8821
} | research_synthesis | synthetic-research_synthesis-011533 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 18.908,
"estimated_latency_ms": 162473,
"estimated_pass_rate": 0.6791,
"estimated_quality": 0.9014,
"finalizer": "worker_a",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.6791,
"quality": 0.9014,
"quality_target_met": true,
"utility": 70.734256
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":18.908,"estimated_latency_ms":162473,"estimated_pass_rate":0.6791,"estimated_quality":0.9014,"finalizer":"worker_a","independent_verification":true,"max_calls":3,"max_credits":20.81,"max_latency_ms":185976,"policy":"stri... | {
"ambiguity": 0.7073,
"category": "research",
"complexity": 0.8359,
"consequence": 0.4734,
"evidence_need": 0.6995,
"family": "research_synthesis",
"input_tokens": 5320,
"request_id": "synthetic-research_synthesis-011533",
"reversibility": 0.7059,
"risk": 0.4157,
"summary": "Review competing appr... | [
{
"credits_per_call": 15.033,
"id": "worker_a",
"p95_latency_ms": 67406,
"pass_rate": 0.995,
"provider_group": "provider_z",
"quality": 0.9782,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 0.85,
"id": "worker_b",
... |
{
"max_calls": 3,
"max_credits": 21.338,
"max_latency_ms": 165266,
"minimum_quality": 0.824
} | debugging | synthetic-debugging-016012 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 17.401,
"estimated_latency_ms": 142504,
"estimated_pass_rate": 0.7191,
"estimated_quality": 0.8892,
"finalizer": "worker_c",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7191,
"quality": 0.8892,
"quality_target_met": true,
"utility": 70.396722
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":17.401,"estimated_latency_ms":142504,"estimated_pass_rate":0.7191,"estimated_quality":0.8892,"finalizer":"worker_c","independent_verification":true,"max_calls":3,"max_credits":21.338,"max_latency_ms":165266,"policy":"str... | {
"ambiguity": 0.6289,
"category": "coding",
"complexity": 0.6955,
"consequence": 0.4847,
"evidence_need": 0.6989,
"family": "debugging",
"input_tokens": 1600,
"request_id": "synthetic-debugging-016012",
"reversibility": 0.6871,
"risk": 0.5085,
"summary": "Find the root cause of a concurrency-sens... | [
{
"credits_per_call": 0.854,
"id": "worker_a",
"p95_latency_ms": 22468,
"pass_rate": 0.4856,
"provider_group": "provider_z",
"quality": 0.6764,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 3.906,
"id": "worker_b",
"p95_late... |
{
"max_calls": 3,
"max_credits": 20.617,
"max_latency_ms": 148628,
"minimum_quality": 0.9432
} | research_synthesis | synthetic-research_synthesis-013965 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 18.429,
"estimated_latency_ms": 141188,
"estimated_pass_rate": 0.7069,
"estimated_quality": 0.9227,
"finalizer": "worker_c",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7069,
"quality": 0.9227,
"quality_target_met": false,
"utility": 74.446346
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":18.429,"estimated_latency_ms":141188,"estimated_pass_rate":0.7069,"estimated_quality":0.9227,"finalizer":"worker_c","independent_verification":true,"max_calls":3,"max_credits":20.617,"max_latency_ms":148628,"policy":"str... | {
"ambiguity": 0.8998,
"category": "research",
"complexity": 0.792,
"consequence": 0.4887,
"evidence_need": 0.8497,
"family": "research_synthesis",
"input_tokens": 4080,
"request_id": "synthetic-research_synthesis-013965",
"reversibility": 0.7901,
"risk": 0.5055,
"summary": "Compare evidence for w... | [
{
"credits_per_call": 2.824,
"id": "worker_a",
"p95_latency_ms": 47551,
"pass_rate": 0.837,
"provider_group": "provider_x",
"quality": 0.8809,
"strengths": [
"reasoning",
"review",
"synthesis",
"writing"
]
},
{
"credits_per_call": 0.967,
"id": "wor... |
{
"max_calls": 3,
"max_credits": 20.761,
"max_latency_ms": 228537,
"minimum_quality": 0.841
} | debugging | synthetic-debugging-015660 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 16.233,
"estimated_latency_ms": 130293,
"estimated_pass_rate": 0.909,
"estimated_quality": 0.9381,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 2... | {
"candidate_count": 15,
"pass_rate": 0.909,
"quality": 0.9381,
"quality_target_met": true,
"utility": 72.700405
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":16.233,"estimated_latency_ms":130293,"estimated_pass_rate":0.909,"estimated_quality":0.9381,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":20.761,"max_latency_ms":228537,"policy":"balanced","reason_codes":["in... | {
"ambiguity": 0.5127,
"category": "coding",
"complexity": 0.6183,
"consequence": 0.6966,
"evidence_need": 0.5831,
"family": "debugging",
"input_tokens": 5640,
"request_id": "synthetic-debugging-015660",
"reversibility": 0.6928,
"risk": 0.4668,
"summary": "Diagnose an intermittent failure in usage... | [
{
"credits_per_call": 13.169,
"id": "worker_a",
"p95_latency_ms": 78480,
"pass_rate": 0.995,
"provider_group": "provider_z",
"quality": 0.9811,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 1.206,
"id": "worker_b",
... |
{
"max_calls": 2,
"max_credits": 18.265,
"max_latency_ms": 208460,
"minimum_quality": 0.7153
} | product_tradeoff | synthetic-product_tradeoff-010272 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 15.771,
"estimated_latency_ms": 154135,
"estimated_pass_rate": 0.8523,
"estimated_quality": 0.8872,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": ... | {
"candidate_count": 9,
"pass_rate": 0.8523,
"quality": 0.8872,
"quality_target_met": true,
"utility": 63.980865
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":15.771,"estimated_latency_ms":154135,"estimated_pass_rate":0.8523,"estimated_quality":0.8872,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":18.265,"max_latency_ms":208460,"policy":"balanced","reason_codes":["c... | {
"ambiguity": 0.7032,
"category": "decision",
"complexity": 0.7444,
"consequence": 0.2254,
"evidence_need": 0.5111,
"family": "product_tradeoff",
"input_tokens": 560,
"request_id": "synthetic-product_tradeoff-010272",
"reversibility": 0.6586,
"risk": 0.1152,
"summary": "Recommend whether to launc... | [
{
"credits_per_call": 11.956,
"id": "worker_a",
"p95_latency_ms": 91859,
"pass_rate": 0.9497,
"provider_group": "provider_z",
"quality": 0.9436,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 1.212,
"id": "worker_b",... |
{
"max_calls": 3,
"max_credits": 19.552,
"max_latency_ms": 157998,
"minimum_quality": 0.8968
} | feature_implementation | synthetic-feature_implementation-008139 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 15.698,
"estimated_latency_ms": 153021,
"estimated_pass_rate": 0.7959,
"estimated_quality": 0.8978,
"finalizer": "worker_b",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7959,
"quality": 0.8978,
"quality_target_met": true,
"utility": 64.018651
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":15.698,"estimated_latency_ms":153021,"estimated_pass_rate":0.7959,"estimated_quality":0.8978,"finalizer":"worker_b","independent_verification":true,"max_calls":3,"max_credits":19.552,"max_latency_ms":157998,"policy":"str... | {
"ambiguity": 0.5212,
"category": "coding",
"complexity": 0.7546,
"consequence": 0.5084,
"evidence_need": 0.2852,
"family": "feature_implementation",
"input_tokens": 600,
"request_id": "synthetic-feature_implementation-008139",
"reversibility": 0.66,
"risk": 0.4363,
"summary": "Implement backgrou... | [
{
"credits_per_call": 3.034,
"id": "worker_a",
"p95_latency_ms": 53935,
"pass_rate": 0.7665,
"provider_group": "provider_z",
"quality": 0.8704,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 11.47,
"id": "worker_b",
"p95_l... |
{
"max_calls": 3,
"max_credits": 20.873,
"max_latency_ms": 223058,
"minimum_quality": 0.8416
} | debugging | synthetic-debugging-014252 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 16.997,
"estimated_latency_ms": 176629,
"estimated_pass_rate": 0.7571,
"estimated_quality": 0.912,
"finalizer": "worker_a",
"independent_verification"... | {
"candidate_count": 15,
"pass_rate": 0.7571,
"quality": 0.912,
"quality_target_met": true,
"utility": 73.185967
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":16.997,"estimated_latency_ms":176629,"estimated_pass_rate":0.7571,"estimated_quality":0.912,"finalizer":"worker_a","independent_verification":true,"max_calls":3,"max_credits":20.873,"max_latency_ms":223058,"policy":"stri... | {
"ambiguity": 0.6081,
"category": "coding",
"complexity": 0.8364,
"consequence": 0.3343,
"evidence_need": 0.5003,
"family": "debugging",
"input_tokens": 1480,
"request_id": "synthetic-debugging-014252",
"reversibility": 0.5463,
"risk": 0.5646,
"summary": "Explain why background jobs fails only un... | [
{
"credits_per_call": 12.071,
"id": "worker_a",
"p95_latency_ms": 86557,
"pass_rate": 0.995,
"provider_group": "provider_x",
"quality": 0.995,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 3.726,
"id": "worker_b",
... |
{
"max_calls": 2,
"max_credits": 13.801,
"max_latency_ms": 110571,
"minimum_quality": 0.7492
} | incident_response | synthetic-incident_response-013279 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 4.493,
"estimated_latency_ms": 86171,
"estimated_pass_rate": 0.6606,
"estimated_quality": 0.7725,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 13... | {
"candidate_count": 4,
"pass_rate": 0.6606,
"quality": 0.7725,
"quality_target_met": true,
"utility": 57.443353
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":4.493,"estimated_latency_ms":86171,"estimated_pass_rate":0.6606,"estimated_quality":0.7725,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":13.801,"max_latency_ms":110571,"policy":"balanced","reason_codes":["com... | {
"ambiguity": 0.6228,
"category": "operations",
"complexity": 0.9404,
"consequence": 0.8666,
"evidence_need": 0.6294,
"family": "incident_response",
"input_tokens": 6880,
"request_id": "synthetic-incident_response-013279",
"reversibility": 0.4042,
"risk": 0.5825,
"summary": "Plan immediate contai... | [
{
"credits_per_call": 3.467,
"id": "worker_a",
"p95_latency_ms": 57415,
"pass_rate": 0.7833,
"provider_group": "provider_y",
"quality": 0.8922,
"strengths": [
"drafting",
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 1.026,
"id": "wor... |
{
"max_calls": 1,
"max_credits": 3.442,
"max_latency_ms": 24318,
"minimum_quality": 0.6351
} | data_analysis | synthetic-data_analysis-002350 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 0.983,
"estimated_latency_ms": 20482,
"estimated_pass_rate": 0.323,
"estimated_quality": 0.5763,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 3.442,
"max_latency_... | {
"candidate_count": 1,
"pass_rate": 0.323,
"quality": 0.5763,
"quality_target_met": false,
"utility": 34.900716
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":0.983,"estimated_latency_ms":20482,"estimated_pass_rate":0.323,"estimated_quality":0.5763,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":3.442,"max_latency_ms":24318,"policy":"lean","reason_codes":["best_feasible_bel... | {
"ambiguity": 0.6904,
"category": "analysis",
"complexity": 0.4931,
"consequence": 0.6473,
"evidence_need": 0.6543,
"family": "data_analysis",
"input_tokens": 5120,
"request_id": "synthetic-data_analysis-002350",
"reversibility": 0.7291,
"risk": 0.4429,
"summary": "Compare two experiments on rate... | [
{
"credits_per_call": 3.902,
"id": "worker_a",
"p95_latency_ms": 74281,
"pass_rate": 0.9117,
"provider_group": "provider_y",
"quality": 0.9139,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 11.195,
"id": "worker_b",
"p95_... |
{
"max_calls": 3,
"max_credits": 20.531,
"max_latency_ms": 159831,
"minimum_quality": 0.8012
} | data_analysis | synthetic-data_analysis-013206 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 19.451,
"estimated_latency_ms": 137579,
"estimated_pass_rate": 0.7405,
"estimated_quality": 0.9156,
"finalizer": "worker_b",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7405,
"quality": 0.9156,
"quality_target_met": true,
"utility": 68.95908
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":19.451,"estimated_latency_ms":137579,"estimated_pass_rate":0.7405,"estimated_quality":0.9156,"finalizer":"worker_b","independent_verification":true,"max_calls":3,"max_credits":20.531,"max_latency_ms":159831,"policy":"str... | {
"ambiguity": 0.5822,
"category": "analysis",
"complexity": 0.7685,
"consequence": 0.5012,
"evidence_need": 0.6217,
"family": "data_analysis",
"input_tokens": 2440,
"request_id": "synthetic-data_analysis-013206",
"reversibility": 0.8748,
"risk": 0.2529,
"summary": "Compare two experiments on prom... | [
{
"credits_per_call": 1.112,
"id": "worker_a",
"p95_latency_ms": 20912,
"pass_rate": 0.5261,
"provider_group": "provider_x",
"quality": 0.6867,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 14.656,
"id": "worker_b",
"p95_lat... |
{
"max_calls": 1,
"max_credits": 2.836,
"max_latency_ms": 20098,
"minimum_quality": 0.5928
} | product_tradeoff | synthetic-product_tradeoff-004576 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 1.159,
"estimated_latency_ms": 18754,
"estimated_pass_rate": 0.3519,
"estimated_quality": 0.7113,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 2.836,
"max_latency... | {
"candidate_count": 1,
"pass_rate": 0.3519,
"quality": 0.7113,
"quality_target_met": true,
"utility": 45.134506
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":1.159,"estimated_latency_ms":18754,"estimated_pass_rate":0.3519,"estimated_quality":0.7113,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":2.836,"max_latency_ms":20098,"policy":"lean","reason_codes":["latency_constrai... | {
"ambiguity": 0.3847,
"category": "decision",
"complexity": 0.5771,
"consequence": 0.3357,
"evidence_need": 0.3994,
"family": "product_tradeoff",
"input_tokens": 2080,
"request_id": "synthetic-product_tradeoff-004576",
"reversibility": 0.6303,
"risk": 0.3594,
"summary": "Evaluate build-versus-buy... | [
{
"credits_per_call": 3.691,
"id": "worker_a",
"p95_latency_ms": 47538,
"pass_rate": 0.7208,
"provider_group": "provider_y",
"quality": 0.8483,
"strengths": [
"drafting",
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 13.279,
"id": "wo... |
{
"max_calls": 2,
"max_credits": 12.299,
"max_latency_ms": 195638,
"minimum_quality": 0.788
} | incident_response | synthetic-incident_response-015151 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 3.794,
"estimated_latency_ms": 95297,
"estimated_pass_rate": 0.6793,
"estimated_quality": 0.8474,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 12... | {
"candidate_count": 5,
"pass_rate": 0.6793,
"quality": 0.8474,
"quality_target_met": true,
"utility": 68.096285
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":3.794,"estimated_latency_ms":95297,"estimated_pass_rate":0.6793,"estimated_quality":0.8474,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":12.299,"max_latency_ms":195638,"policy":"balanced","reason_codes":["com... | {
"ambiguity": 0.5823,
"category": "operations",
"complexity": 0.762,
"consequence": 0.6654,
"evidence_need": 0.9057,
"family": "incident_response",
"input_tokens": 7240,
"request_id": "synthetic-incident_response-015151",
"reversibility": 0.5154,
"risk": 0.7566,
"summary": "Review an outage invol... | [
{
"credits_per_call": 12.039,
"id": "worker_a",
"p95_latency_ms": 85195,
"pass_rate": 0.995,
"provider_group": "provider_z",
"quality": 0.9306,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 0.861,
"id": "worker_b",
... |
{
"max_calls": 3,
"max_credits": 23.043,
"max_latency_ms": 164886,
"minimum_quality": 0.7903
} | incident_response | synthetic-incident_response-017631 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 17.262,
"estimated_latency_ms": 124830,
"estimated_pass_rate": 0.7491,
"estimated_quality": 0.8796,
"finalizer": "worker_c",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7491,
"quality": 0.8796,
"quality_target_met": true,
"utility": 78.484464
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":17.262,"estimated_latency_ms":124830,"estimated_pass_rate":0.7491,"estimated_quality":0.8796,"finalizer":"worker_c","independent_verification":true,"max_calls":3,"max_credits":23.043,"max_latency_ms":164886,"policy":"str... | {
"ambiguity": 0.4741,
"category": "operations",
"complexity": 0.9143,
"consequence": 0.8487,
"evidence_need": 0.8246,
"family": "incident_response",
"input_tokens": 4120,
"request_id": "synthetic-incident_response-017631",
"reversibility": 0.3495,
"risk": 0.6913,
"summary": "Prioritize actions du... | [
{
"credits_per_call": 3.606,
"id": "worker_a",
"p95_latency_ms": 51645,
"pass_rate": 0.7282,
"provider_group": "provider_y",
"quality": 0.8334,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 1.22,
"id": "worker_b",
"p95_la... |
{
"max_calls": 3,
"max_credits": 16.173,
"max_latency_ms": 146788,
"minimum_quality": 0.8891
} | debugging | synthetic-debugging-016940 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 14.03,
"estimated_latency_ms": 138337,
"estimated_pass_rate": 0.7616,
"estimated_quality": 0.9168,
"finalizer": "worker_b",
"independent_verification"... | {
"candidate_count": 15,
"pass_rate": 0.7616,
"quality": 0.9168,
"quality_target_met": true,
"utility": 73.690446
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":14.03,"estimated_latency_ms":138337,"estimated_pass_rate":0.7616,"estimated_quality":0.9168,"finalizer":"worker_b","independent_verification":true,"max_calls":3,"max_credits":16.173,"max_latency_ms":146788,"policy":"stri... | {
"ambiguity": 0.3524,
"category": "coding",
"complexity": 0.8067,
"consequence": 0.5658,
"evidence_need": 0.6377,
"family": "debugging",
"input_tokens": 4800,
"request_id": "synthetic-debugging-016940",
"reversibility": 0.607,
"risk": 0.6807,
"summary": "Find the root cause of a concurrency-sensi... | [
{
"credits_per_call": 2.782,
"id": "worker_a",
"p95_latency_ms": 57883,
"pass_rate": 0.7328,
"provider_group": "provider_z",
"quality": 0.8835,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 10.31,
"id": "worker_b",
"p95_l... |
{
"max_calls": 2,
"max_credits": 22.705,
"max_latency_ms": 266919,
"minimum_quality": 0.8655
} | research_synthesis | synthetic-research_synthesis-002381 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 18.107,
"estimated_latency_ms": 148071,
"estimated_pass_rate": 0.8974,
"estimated_quality": 0.941,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 2... | {
"candidate_count": 9,
"pass_rate": 0.8974,
"quality": 0.941,
"quality_target_met": true,
"utility": 72.268711
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":18.107,"estimated_latency_ms":148071,"estimated_pass_rate":0.8974,"estimated_quality":0.941,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":22.705,"max_latency_ms":266919,"policy":"balanced","reason_codes":["in... | {
"ambiguity": 0.6179,
"category": "research",
"complexity": 0.527,
"consequence": 0.5215,
"evidence_need": 0.7247,
"family": "research_synthesis",
"input_tokens": 1160,
"request_id": "synthetic-research_synthesis-002381",
"reversibility": 0.9362,
"risk": 0.5226,
"summary": "Review competing appro... | [
{
"credits_per_call": 1.17,
"id": "worker_a",
"p95_latency_ms": 24939,
"pass_rate": 0.4344,
"provider_group": "provider_x",
"quality": 0.6942,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 14.448,
"id": "worker_b",
"p95_late... |
{
"max_calls": 3,
"max_credits": 20.587,
"max_latency_ms": 201858,
"minimum_quality": 0.7814
} | research_synthesis | synthetic-research_synthesis-020581 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 18.015,
"estimated_latency_ms": 165567,
"estimated_pass_rate": 0.7115,
"estimated_quality": 0.896,
"finalizer": "worker_c",
"independent_verification"... | {
"candidate_count": 15,
"pass_rate": 0.7115,
"quality": 0.896,
"quality_target_met": true,
"utility": 69.030084
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":18.015,"estimated_latency_ms":165567,"estimated_pass_rate":0.7115,"estimated_quality":0.896,"finalizer":"worker_c","independent_verification":true,"max_calls":3,"max_credits":20.587,"max_latency_ms":201858,"policy":"stri... | {
"ambiguity": 0.7298,
"category": "research",
"complexity": 0.6893,
"consequence": 0.5394,
"evidence_need": 0.7516,
"family": "research_synthesis",
"input_tokens": 960,
"request_id": "synthetic-research_synthesis-020581",
"reversibility": 0.8649,
"risk": 0.2811,
"summary": "Produce a decision mem... | [
{
"credits_per_call": 3.396,
"id": "worker_a",
"p95_latency_ms": 77740,
"pass_rate": 0.7345,
"provider_group": "provider_y",
"quality": 0.7952,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 1.196,
"id": "worker_b",
"p95_l... |
{
"max_calls": 2,
"max_credits": 5.69,
"max_latency_ms": 144150,
"minimum_quality": 0.8456
} | feature_implementation | synthetic-feature_implementation-010707 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 3.948,
"estimated_latency_ms": 79711,
"estimated_pass_rate": 0.5967,
"estimated_quality": 0.8013,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 5.... | {
"candidate_count": 4,
"pass_rate": 0.5967,
"quality": 0.8013,
"quality_target_met": false,
"utility": 54.170275
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":3.948,"estimated_latency_ms":79711,"estimated_pass_rate":0.5967,"estimated_quality":0.8013,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":5.69,"max_latency_ms":144150,"policy":"balanced","reason_codes":["best_... | {
"ambiguity": 0.463,
"category": "coding",
"complexity": 0.7943,
"consequence": 0.5363,
"evidence_need": 0.6472,
"family": "feature_implementation",
"input_tokens": 6120,
"request_id": "synthetic-feature_implementation-010707",
"reversibility": 0.736,
"risk": 0.2718,
"summary": "Implement deploym... | [
{
"credits_per_call": 1.152,
"id": "worker_a",
"p95_latency_ms": 21626,
"pass_rate": 0.5782,
"provider_group": "provider_x",
"quality": 0.7893,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 12.327,
"id": "worker_b",
"p95_lat... |
{
"max_calls": 2,
"max_credits": 9.002,
"max_latency_ms": 166037,
"minimum_quality": 0.8695
} | incident_response | synthetic-incident_response-014247 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 5.346,
"estimated_latency_ms": 101499,
"estimated_pass_rate": 0.6358,
"estimated_quality": 0.7868,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 9... | {
"candidate_count": 4,
"pass_rate": 0.6358,
"quality": 0.7868,
"quality_target_met": false,
"utility": 57.746403
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":5.346,"estimated_latency_ms":101499,"estimated_pass_rate":0.6358,"estimated_quality":0.7868,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":9.002,"max_latency_ms":166037,"policy":"balanced","reason_codes":["bes... | {
"ambiguity": 0.7626,
"category": "operations",
"complexity": 0.6604,
"consequence": 0.8036,
"evidence_need": 0.5638,
"family": "incident_response",
"input_tokens": 2480,
"request_id": "synthetic-incident_response-014247",
"reversibility": 0.1662,
"risk": 0.8029,
"summary": "Prioritize actions du... | [
{
"credits_per_call": 10.586,
"id": "worker_a",
"p95_latency_ms": 76894,
"pass_rate": 0.995,
"provider_group": "provider_z",
"quality": 0.9336,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 4.14,
"id": "worker_b",
... |
{
"max_calls": 1,
"max_credits": 1.953,
"max_latency_ms": 28949,
"minimum_quality": 0.5611
} | product_tradeoff | synthetic-product_tradeoff-007880 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 1.023,
"estimated_latency_ms": 24349,
"estimated_pass_rate": 0.4221,
"estimated_quality": 0.6896,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 1.953,
"max_latency... | {
"candidate_count": 1,
"pass_rate": 0.4221,
"quality": 0.6896,
"quality_target_met": true,
"utility": 45.011884
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":1.023,"estimated_latency_ms":24349,"estimated_pass_rate":0.4221,"estimated_quality":0.6896,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":1.953,"max_latency_ms":28949,"policy":"lean","reason_codes":["latency_constrai... | {
"ambiguity": 0.3551,
"category": "decision",
"complexity": 0.5701,
"consequence": 0.6205,
"evidence_need": 0.3271,
"family": "product_tradeoff",
"input_tokens": 6920,
"request_id": "synthetic-product_tradeoff-007880",
"reversibility": 0.5711,
"risk": 0.2492,
"summary": "Recommend whether to laun... | [
{
"credits_per_call": 3.762,
"id": "worker_a",
"p95_latency_ms": 49859,
"pass_rate": 0.76,
"provider_group": "provider_x",
"quality": 0.8931,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 12.382,
"id": "worker_b",
"p95_la... |
{
"max_calls": 3,
"max_credits": 27.574,
"max_latency_ms": 170703,
"minimum_quality": 0.8719
} | product_tradeoff | synthetic-product_tradeoff-001736 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 20.485,
"estimated_latency_ms": 153718,
"estimated_pass_rate": 0.7309,
"estimated_quality": 0.918,
"finalizer": "worker_a",
"independent_verification"... | {
"candidate_count": 15,
"pass_rate": 0.7309,
"quality": 0.918,
"quality_target_met": true,
"utility": 68.833324
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":20.485,"estimated_latency_ms":153718,"estimated_pass_rate":0.7309,"estimated_quality":0.918,"finalizer":"worker_a","independent_verification":true,"max_calls":3,"max_credits":27.574,"max_latency_ms":170703,"policy":"stri... | {
"ambiguity": 0.6836,
"category": "decision",
"complexity": 0.6477,
"consequence": 0.314,
"evidence_need": 0.3256,
"family": "product_tradeoff",
"input_tokens": 5800,
"request_id": "synthetic-product_tradeoff-001736",
"reversibility": 0.6563,
"risk": 0.4052,
"summary": "Evaluate build-versus-buy ... | [
{
"credits_per_call": 15.251,
"id": "worker_a",
"p95_latency_ms": 75276,
"pass_rate": 0.9456,
"provider_group": "provider_y",
"quality": 0.9877,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 1.195,
"id": "worker_b",... |
{
"max_calls": 3,
"max_credits": 20.921,
"max_latency_ms": 217071,
"minimum_quality": 0.8887
} | product_tradeoff | synthetic-product_tradeoff-018656 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 18.009,
"estimated_latency_ms": 152313,
"estimated_pass_rate": 0.8931,
"estimated_quality": 0.9368,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": ... | {
"candidate_count": 15,
"pass_rate": 0.8931,
"quality": 0.9368,
"quality_target_met": true,
"utility": 68.714824
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":18.009,"estimated_latency_ms":152313,"estimated_pass_rate":0.8931,"estimated_quality":0.9368,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":20.921,"max_latency_ms":217071,"policy":"balanced","reason_codes":["c... | {
"ambiguity": 0.7692,
"category": "decision",
"complexity": 0.5776,
"consequence": 0.2936,
"evidence_need": 0.6431,
"family": "product_tradeoff",
"input_tokens": 7240,
"request_id": "synthetic-product_tradeoff-018656",
"reversibility": 0.8056,
"risk": 0.2621,
"summary": "Recommend whether to laun... | [
{
"credits_per_call": 1.136,
"id": "worker_a",
"p95_latency_ms": 30312,
"pass_rate": 0.5462,
"provider_group": "provider_z",
"quality": 0.7345,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 14.849,
"id": "worker_b",
"p95_lat... |
{
"max_calls": 3,
"max_credits": 25.939,
"max_latency_ms": 214529,
"minimum_quality": 0.8157
} | debugging | synthetic-debugging-014516 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 19.349,
"estimated_latency_ms": 162505,
"estimated_pass_rate": 0.7209,
"estimated_quality": 0.9203,
"finalizer": "worker_c",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7209,
"quality": 0.9203,
"quality_target_met": true,
"utility": 71.873836
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":19.349,"estimated_latency_ms":162505,"estimated_pass_rate":0.7209,"estimated_quality":0.9203,"finalizer":"worker_c","independent_verification":true,"max_calls":3,"max_credits":25.939,"max_latency_ms":214529,"policy":"str... | {
"ambiguity": 0.5902,
"category": "coding",
"complexity": 0.6748,
"consequence": 0.4109,
"evidence_need": 0.6252,
"family": "debugging",
"input_tokens": 1520,
"request_id": "synthetic-debugging-014516",
"reversibility": 0.4402,
"risk": 0.4978,
"summary": "Find the root cause of a concurrency-sens... | [
{
"credits_per_call": 0.936,
"id": "worker_a",
"p95_latency_ms": 20960,
"pass_rate": 0.5019,
"provider_group": "provider_z",
"quality": 0.7645,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 3.087,
"id": "worker_b",
"p95_late... |
{
"max_calls": 2,
"max_credits": 16.015,
"max_latency_ms": 240291,
"minimum_quality": 0.7934
} | research_synthesis | synthetic-research_synthesis-010453 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 15.752,
"estimated_latency_ms": 141743,
"estimated_pass_rate": 0.8755,
"estimated_quality": 0.91,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 16... | {
"candidate_count": 9,
"pass_rate": 0.8755,
"quality": 0.91,
"quality_target_met": true,
"utility": 66.943368
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":15.752,"estimated_latency_ms":141743,"estimated_pass_rate":0.8755,"estimated_quality":0.91,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":16.015,"max_latency_ms":240291,"policy":"balanced","reason_codes":["com... | {
"ambiguity": 0.7443,
"category": "research",
"complexity": 0.8893,
"consequence": 0.4833,
"evidence_need": 0.6069,
"family": "research_synthesis",
"input_tokens": 4160,
"request_id": "synthetic-research_synthesis-010453",
"reversibility": 0.953,
"risk": 0.2988,
"summary": "Produce a decision mem... | [
{
"credits_per_call": 0.88,
"id": "worker_a",
"p95_latency_ms": 32655,
"pass_rate": 0.533,
"provider_group": "provider_y",
"quality": 0.7251,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 11.838,
"id": "worker_b",
"p95_laten... |
{
"max_calls": 3,
"max_credits": 8.42,
"max_latency_ms": 132172,
"minimum_quality": 0.7735
} | debugging | synthetic-debugging-014604 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 4.404,
"estimated_latency_ms": 92527,
"estimated_pass_rate": 0.5581,
"estimated_quality": 0.7833,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 8.... | {
"candidate_count": 4,
"pass_rate": 0.5581,
"quality": 0.7833,
"quality_target_met": true,
"utility": 55.374657
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":4.404,"estimated_latency_ms":92527,"estimated_pass_rate":0.5581,"estimated_quality":0.7833,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":8.42,"max_latency_ms":132172,"policy":"balanced","reason_codes":["indep... | {
"ambiguity": 0.6141,
"category": "coding",
"complexity": 0.6778,
"consequence": 0.3225,
"evidence_need": 0.6567,
"family": "debugging",
"input_tokens": 1160,
"request_id": "synthetic-debugging-014604",
"reversibility": 0.7063,
"risk": 0.6984,
"summary": "Explain why OAuth callbacks fails only un... | [
{
"credits_per_call": 3.263,
"id": "worker_a",
"p95_latency_ms": 59522,
"pass_rate": 0.735,
"provider_group": "provider_z",
"quality": 0.8632,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 10.06,
"id": "worker_b",
"p95_la... |
{
"max_calls": 2,
"max_credits": 9.074,
"max_latency_ms": 258566,
"minimum_quality": 0.8375
} | incident_response | synthetic-incident_response-006799 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 4,
"estimated_latency_ms": 113958,
"estimated_pass_rate": 0.6381,
"estimated_quality": 0.8007,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 9.074... | {
"candidate_count": 4,
"pass_rate": 0.6381,
"quality": 0.8007,
"quality_target_met": false,
"utility": 58.864762
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":4.0,"estimated_latency_ms":113958,"estimated_pass_rate":0.6381,"estimated_quality":0.8007,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":9.074,"max_latency_ms":258566,"policy":"balanced","reason_codes":["best_... | {
"ambiguity": 0.7311,
"category": "operations",
"complexity": 0.8153,
"consequence": 0.7635,
"evidence_need": 0.6239,
"family": "incident_response",
"input_tokens": 7080,
"request_id": "synthetic-incident_response-006799",
"reversibility": 0.1736,
"risk": 0.7146,
"summary": "Plan immediate contai... | [
{
"credits_per_call": 3.114,
"id": "worker_a",
"p95_latency_ms": 82192,
"pass_rate": 0.7228,
"provider_group": "provider_z",
"quality": 0.9002,
"strengths": [
"drafting",
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 13.729,
"id": "wo... |
{
"max_calls": 3,
"max_credits": 17.376,
"max_latency_ms": 199643,
"minimum_quality": 0.9266
} | debugging | synthetic-debugging-006372 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 15.351,
"estimated_latency_ms": 155359,
"estimated_pass_rate": 0.7643,
"estimated_quality": 0.9313,
"finalizer": "worker_b",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7643,
"quality": 0.9313,
"quality_target_met": true,
"utility": 68.669152
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":15.351,"estimated_latency_ms":155359,"estimated_pass_rate":0.7643,"estimated_quality":0.9313,"finalizer":"worker_b","independent_verification":true,"max_calls":3,"max_credits":17.376,"max_latency_ms":199643,"policy":"str... | {
"ambiguity": 0.4837,
"category": "coding",
"complexity": 0.7682,
"consequence": 0.4544,
"evidence_need": 0.5794,
"family": "debugging",
"input_tokens": 4840,
"request_id": "synthetic-debugging-006372",
"reversibility": 0.5201,
"risk": 0.5268,
"summary": "Find the root cause of a concurrency-sens... | [
{
"credits_per_call": 3.939,
"id": "worker_a",
"p95_latency_ms": 72013,
"pass_rate": 0.7279,
"provider_group": "provider_x",
"quality": 0.9132,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 10.578,
"id": "worker_b",
"p95_... |
{
"max_calls": 3,
"max_credits": 21.392,
"max_latency_ms": 240891,
"minimum_quality": 0.8736
} | incident_response | synthetic-incident_response-004327 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 16.897,
"estimated_latency_ms": 196231,
"estimated_pass_rate": 0.7701,
"estimated_quality": 0.8975,
"finalizer": "worker_b",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7701,
"quality": 0.8975,
"quality_target_met": true,
"utility": 75.011972
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":16.897,"estimated_latency_ms":196231,"estimated_pass_rate":0.7701,"estimated_quality":0.8975,"finalizer":"worker_b","independent_verification":true,"max_calls":3,"max_credits":21.392,"max_latency_ms":240891,"policy":"str... | {
"ambiguity": 0.429,
"category": "operations",
"complexity": 0.5651,
"consequence": 0.8023,
"evidence_need": 0.7092,
"family": "incident_response",
"input_tokens": 4320,
"request_id": "synthetic-incident_response-004327",
"reversibility": 0.2127,
"risk": 0.7767,
"summary": "Prioritize actions dur... | [
{
"credits_per_call": 4.147,
"id": "worker_a",
"p95_latency_ms": 82122,
"pass_rate": 0.791,
"provider_group": "provider_x",
"quality": 0.8737,
"strengths": [
"drafting",
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 11.784,
"id": "wor... |
{
"max_calls": 3,
"max_credits": 22.551,
"max_latency_ms": 203879,
"minimum_quality": 0.9257
} | debugging | synthetic-debugging-010204 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 16.671,
"estimated_latency_ms": 140751,
"estimated_pass_rate": 0.7525,
"estimated_quality": 0.8894,
"finalizer": "worker_c",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7525,
"quality": 0.8894,
"quality_target_met": false,
"utility": 73.03646
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":16.671,"estimated_latency_ms":140751,"estimated_pass_rate":0.7525,"estimated_quality":0.8894,"finalizer":"worker_c","independent_verification":true,"max_calls":3,"max_credits":22.551,"max_latency_ms":203879,"policy":"str... | {
"ambiguity": 0.7087,
"category": "coding",
"complexity": 0.7295,
"consequence": 0.4704,
"evidence_need": 0.6378,
"family": "debugging",
"input_tokens": 6880,
"request_id": "synthetic-debugging-010204",
"reversibility": 0.4398,
"risk": 0.6483,
"summary": "Explain why access control fails only und... | [
{
"credits_per_call": 1.061,
"id": "worker_a",
"p95_latency_ms": 28575,
"pass_rate": 0.5789,
"provider_group": "provider_y",
"quality": 0.7186,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 4.214,
"id": "worker_b",
"p95_late... |
{
"max_calls": 3,
"max_credits": 21.124,
"max_latency_ms": 204648,
"minimum_quality": 0.9074
} | incident_response | synthetic-incident_response-006575 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 16.248,
"estimated_latency_ms": 155613,
"estimated_pass_rate": 0.818,
"estimated_quality": 0.9249,
"finalizer": "worker_b",
"independent_verification"... | {
"candidate_count": 15,
"pass_rate": 0.818,
"quality": 0.9249,
"quality_target_met": true,
"utility": 84.286231
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":16.248,"estimated_latency_ms":155613,"estimated_pass_rate":0.818,"estimated_quality":0.9249,"finalizer":"worker_b","independent_verification":true,"max_calls":3,"max_credits":21.124,"max_latency_ms":204648,"policy":"stri... | {
"ambiguity": 0.3568,
"category": "operations",
"complexity": 0.7918,
"consequence": 0.8267,
"evidence_need": 0.8798,
"family": "incident_response",
"input_tokens": 1400,
"request_id": "synthetic-incident_response-006575",
"reversibility": 0.4656,
"risk": 0.9294,
"summary": "Prioritize actions du... | [
{
"credits_per_call": 1.197,
"id": "worker_a",
"p95_latency_ms": 30430,
"pass_rate": 0.6161,
"provider_group": "provider_z",
"quality": 0.7817,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 11.109,
"id": "worker_b",
"p95_lat... |
{
"max_calls": 3,
"max_credits": 2.848,
"max_latency_ms": 32031,
"minimum_quality": 0.7131
} | feature_implementation | synthetic-feature_implementation-007891 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 0.933,
"estimated_latency_ms": 17731,
"estimated_pass_rate": 0.4007,
"estimated_quality": 0.59,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 2.848,
"max_latency_m... | {
"candidate_count": 1,
"pass_rate": 0.4007,
"quality": 0.59,
"quality_target_met": false,
"utility": 35.841703
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":0.933,"estimated_latency_ms":17731,"estimated_pass_rate":0.4007,"estimated_quality":0.59,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":2.848,"max_latency_ms":32031,"policy":"lean","reason_codes":["best_feasible_belo... | {
"ambiguity": 0.2673,
"category": "coding",
"complexity": 0.6667,
"consequence": 0.5391,
"evidence_need": 0.5956,
"family": "feature_implementation",
"input_tokens": 7920,
"request_id": "synthetic-feature_implementation-007891",
"reversibility": 0.4842,
"risk": 0.4009,
"summary": "Implement usage... | [
{
"credits_per_call": 0.933,
"id": "worker_a",
"p95_latency_ms": 17731,
"pass_rate": 0.529,
"provider_group": "provider_x",
"quality": 0.6774,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 12.469,
"id": "worker_b",
"p95_late... |
{
"max_calls": 1,
"max_credits": 1.97,
"max_latency_ms": 38235,
"minimum_quality": 0.491
} | product_tradeoff | synthetic-product_tradeoff-002720 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 1.097,
"estimated_latency_ms": 24779,
"estimated_pass_rate": 0.4218,
"estimated_quality": 0.6853,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 1.97,
"max_latency_... | {
"candidate_count": 1,
"pass_rate": 0.4218,
"quality": 0.6853,
"quality_target_met": true,
"utility": 45.37571
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":1.097,"estimated_latency_ms":24779,"estimated_pass_rate":0.4218,"estimated_quality":0.6853,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":1.97,"max_latency_ms":38235,"policy":"lean","reason_codes":["single_call_suffi... | {
"ambiguity": 0.3927,
"category": "decision",
"complexity": 0.3864,
"consequence": 0.2335,
"evidence_need": 0.3459,
"family": "product_tradeoff",
"input_tokens": 2120,
"request_id": "synthetic-product_tradeoff-002720",
"reversibility": 0.5868,
"risk": 0.2154,
"summary": "Recommend whether to laun... | [
{
"credits_per_call": 14.162,
"id": "worker_a",
"p95_latency_ms": 56596,
"pass_rate": 0.995,
"provider_group": "provider_y",
"quality": 0.9279,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 3.631,
"id": "worker_b",
... |
{
"max_calls": 2,
"max_credits": 18.437,
"max_latency_ms": 209755,
"minimum_quality": 0.8114
} | debugging | synthetic-debugging-007948 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 13.768,
"estimated_latency_ms": 121551,
"estimated_pass_rate": 0.9115,
"estimated_quality": 0.9252,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": ... | {
"candidate_count": 9,
"pass_rate": 0.9115,
"quality": 0.9252,
"quality_target_met": true,
"utility": 70.826852
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":13.768,"estimated_latency_ms":121551,"estimated_pass_rate":0.9115,"estimated_quality":0.9252,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":18.437,"max_latency_ms":209755,"policy":"balanced","reason_codes":["i... | {
"ambiguity": 0.4323,
"category": "coding",
"complexity": 0.6012,
"consequence": 0.5675,
"evidence_need": 0.5857,
"family": "debugging",
"input_tokens": 2000,
"request_id": "synthetic-debugging-007948",
"reversibility": 0.58,
"risk": 0.3312,
"summary": "Diagnose an intermittent failure in tenant-... | [
{
"credits_per_call": 2.738,
"id": "worker_a",
"p95_latency_ms": 68983,
"pass_rate": 0.8514,
"provider_group": "provider_x",
"quality": 0.8481,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 1.089,
"id": "worker_b",
"p95_l... |
{
"max_calls": 3,
"max_credits": 20.32,
"max_latency_ms": 191862,
"minimum_quality": 0.8172
} | debugging | synthetic-debugging-020500 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 17.648,
"estimated_latency_ms": 155675,
"estimated_pass_rate": 0.7408,
"estimated_quality": 0.9024,
"finalizer": "worker_c",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7408,
"quality": 0.9024,
"quality_target_met": true,
"utility": 68.532
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":17.648,"estimated_latency_ms":155675,"estimated_pass_rate":0.7408,"estimated_quality":0.9024,"finalizer":"worker_c","independent_verification":true,"max_calls":3,"max_credits":20.32,"max_latency_ms":191862,"policy":"stri... | {
"ambiguity": 0.393,
"category": "coding",
"complexity": 0.7252,
"consequence": 0.4132,
"evidence_need": 0.7387,
"family": "debugging",
"input_tokens": 4920,
"request_id": "synthetic-debugging-020500",
"reversibility": 0.5714,
"risk": 0.3664,
"summary": "Diagnose an intermittent failure in data r... | [
{
"credits_per_call": 0.8,
"id": "worker_a",
"p95_latency_ms": 23308,
"pass_rate": 0.5824,
"provider_group": "provider_z",
"quality": 0.8016,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 3.815,
"id": "worker_b",
"p95_latenc... |
{
"max_calls": 3,
"max_credits": 24.619,
"max_latency_ms": 213896,
"minimum_quality": 0.7871
} | research_synthesis | synthetic-research_synthesis-005325 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 18.874,
"estimated_latency_ms": 153587,
"estimated_pass_rate": 0.7129,
"estimated_quality": 0.9074,
"finalizer": "worker_b",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.7129,
"quality": 0.9074,
"quality_target_met": true,
"utility": 72.244937
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":18.874,"estimated_latency_ms":153587,"estimated_pass_rate":0.7129,"estimated_quality":0.9074,"finalizer":"worker_b","independent_verification":true,"max_calls":3,"max_credits":24.619,"max_latency_ms":213896,"policy":"str... | {
"ambiguity": 0.8185,
"category": "research",
"complexity": 0.5743,
"consequence": 0.355,
"evidence_need": 0.7683,
"family": "research_synthesis",
"input_tokens": 6960,
"request_id": "synthetic-research_synthesis-005325",
"reversibility": 0.775,
"risk": 0.5297,
"summary": "Produce a decision memo... | [
{
"credits_per_call": 1.048,
"id": "worker_a",
"p95_latency_ms": 21957,
"pass_rate": 0.464,
"provider_group": "provider_z",
"quality": 0.6636,
"strengths": [
"coding",
"drafting",
"speed"
]
},
{
"credits_per_call": 14.306,
"id": "worker_b",
"p95_late... |
{
"max_calls": 3,
"max_credits": 23.027,
"max_latency_ms": 188869,
"minimum_quality": 0.9558
} | research_synthesis | synthetic-research_synthesis-012797 | {
"access": {
"finalizer": [
"writer",
"reviewer"
],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 17.605,
"estimated_latency_ms": 166537,
"estimated_pass_rate": 0.6796,
"estimated_quality": 0.9075,
"finalizer": "worker_b",
"independent_verification... | {
"candidate_count": 15,
"pass_rate": 0.6796,
"quality": 0.9075,
"quality_target_met": false,
"utility": 70.588426
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":["writer","reviewer"],"reviewer":["writer"],"writer":[]},"estimated_credits":17.605,"estimated_latency_ms":166537,"estimated_pass_rate":0.6796,"estimated_quality":0.9075,"finalizer":"worker_b","independent_verification":true,"max_calls":3,"max_credits":23.027,"max_latency_ms":188869,"policy":"str... | {
"ambiguity": 0.8745,
"category": "research",
"complexity": 0.5488,
"consequence": 0.5137,
"evidence_need": 0.9246,
"family": "research_synthesis",
"input_tokens": 1760,
"request_id": "synthetic-research_synthesis-012797",
"reversibility": 0.8125,
"risk": 0.5216,
"summary": "Produce a decision me... | [
{
"credits_per_call": 3.082,
"id": "worker_a",
"p95_latency_ms": 63186,
"pass_rate": 0.7323,
"provider_group": "provider_x",
"quality": 0.8665,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 13.526,
"id": "worker_b",
"p95_... |
{
"max_calls": 1,
"max_credits": 1.802,
"max_latency_ms": 47986,
"minimum_quality": 0.75
} | incident_response | synthetic-incident_response-003511 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 1.208,
"estimated_latency_ms": 30457,
"estimated_pass_rate": 0.3654,
"estimated_quality": 0.5832,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 1.802,
"max_latency... | {
"candidate_count": 1,
"pass_rate": 0.3654,
"quality": 0.5832,
"quality_target_met": false,
"utility": 12.827861
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":1.208,"estimated_latency_ms":30457,"estimated_pass_rate":0.3654,"estimated_quality":0.5832,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":1.802,"max_latency_ms":47986,"policy":"lean","reason_codes":["best_feasible_be... | {
"ambiguity": 0.5314,
"category": "operations",
"complexity": 0.6897,
"consequence": 0.9556,
"evidence_need": 0.7821,
"family": "incident_response",
"input_tokens": 4160,
"request_id": "synthetic-incident_response-003511",
"reversibility": 0.4046,
"risk": 0.7119,
"summary": "Review an outage invo... | [
{
"credits_per_call": 10.14,
"id": "worker_a",
"p95_latency_ms": 72485,
"pass_rate": 0.995,
"provider_group": "provider_z",
"quality": 0.9163,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 1.208,
"id": "worker_b",
... |
{
"max_calls": 3,
"max_credits": 13.309,
"max_latency_ms": 185225,
"minimum_quality": 0.8952
} | incident_response | synthetic-incident_response-000903 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 11.266,
"estimated_latency_ms": 91137,
"estimated_pass_rate": 0.7088,
"estimated_quality": 0.8789,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": 1... | {
"candidate_count": 7,
"pass_rate": 0.7088,
"quality": 0.8789,
"quality_target_met": false,
"utility": 64.656333
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":11.266,"estimated_latency_ms":91137,"estimated_pass_rate":0.7088,"estimated_quality":0.8789,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":13.309,"max_latency_ms":185225,"policy":"balanced","reason_codes":["be... | {
"ambiguity": 0.6489,
"category": "operations",
"complexity": 0.648,
"consequence": 0.6218,
"evidence_need": 0.7274,
"family": "incident_response",
"input_tokens": 7040,
"request_id": "synthetic-incident_response-000903",
"reversibility": 0.3062,
"risk": 0.6395,
"summary": "Plan immediate contain... | [
{
"credits_per_call": 10.419,
"id": "worker_a",
"p95_latency_ms": 60463,
"pass_rate": 0.9457,
"provider_group": "provider_y",
"quality": 0.9537,
"strengths": [
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 0.847,
"id": "worker_b",... |
{
"max_calls": 1,
"max_credits": 3.07,
"max_latency_ms": 47286,
"minimum_quality": 0.7017
} | content_revision | synthetic-content_revision-007658 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 0.899,
"estimated_latency_ms": 28657,
"estimated_pass_rate": 0.3989,
"estimated_quality": 0.6864,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 3.07,
"max_latency_... | {
"candidate_count": 1,
"pass_rate": 0.3989,
"quality": 0.6864,
"quality_target_met": false,
"utility": 46.311903
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":0.899,"estimated_latency_ms":28657,"estimated_pass_rate":0.3989,"estimated_quality":0.6864,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":3.07,"max_latency_ms":47286,"policy":"lean","reason_codes":["best_feasible_bel... | {
"ambiguity": 0.2302,
"category": "writing",
"complexity": 0.3666,
"consequence": 0.2502,
"evidence_need": 0.2268,
"family": "content_revision",
"input_tokens": 7320,
"request_id": "synthetic-content_revision-007658",
"reversibility": 0.8025,
"risk": 0.0531,
"summary": "Rewrite a background jobs ... | [
{
"credits_per_call": 2.864,
"id": "worker_a",
"p95_latency_ms": 77802,
"pass_rate": 0.9151,
"provider_group": "provider_y",
"quality": 0.9758,
"strengths": [
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 0.899,
"id": "worker_b",
"p95_l... |
{
"max_calls": 2,
"max_credits": 23.101,
"max_latency_ms": 262667,
"minimum_quality": 0.8484
} | product_tradeoff | synthetic-product_tradeoff-009768 | {
"access": {
"finalizer": [],
"reviewer": [
"writer"
],
"writer": []
},
"estimated_credits": 19.207,
"estimated_latency_ms": 150754,
"estimated_pass_rate": 0.8906,
"estimated_quality": 0.9135,
"finalizer": null,
"independent_verification": true,
"max_calls": 2,
"max_credits": ... | {
"candidate_count": 9,
"pass_rate": 0.8906,
"quality": 0.9135,
"quality_target_met": true,
"utility": 68.115905
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":["writer"],"writer":[]},"estimated_credits":19.207,"estimated_latency_ms":150754,"estimated_pass_rate":0.8906,"estimated_quality":0.9135,"finalizer":null,"independent_verification":true,"max_calls":2,"max_credits":23.101,"max_latency_ms":262667,"policy":"balanced","reason_codes":["c... | {
"ambiguity": 0.5324,
"category": "decision",
"complexity": 0.3992,
"consequence": 0.2409,
"evidence_need": 0.4703,
"family": "product_tradeoff",
"input_tokens": 1480,
"request_id": "synthetic-product_tradeoff-009768",
"reversibility": 0.7374,
"risk": 0.4078,
"summary": "Evaluate build-versus-buy... | [
{
"credits_per_call": 4.245,
"id": "worker_a",
"p95_latency_ms": 62943,
"pass_rate": 0.7915,
"provider_group": "provider_y",
"quality": 0.8813,
"strengths": [
"drafting",
"reasoning",
"review",
"writing"
]
},
{
"credits_per_call": 14.962,
"id": "wo... |
{
"max_calls": 1,
"max_credits": 3.244,
"max_latency_ms": 31406,
"minimum_quality": 0.4995
} | incident_response | synthetic-incident_response-006711 | {
"access": {
"finalizer": [],
"reviewer": [],
"writer": []
},
"estimated_credits": 0.979,
"estimated_latency_ms": 18362,
"estimated_pass_rate": 0.3651,
"estimated_quality": 0.6682,
"finalizer": null,
"independent_verification": false,
"max_calls": 1,
"max_credits": 3.244,
"max_latency... | {
"candidate_count": 1,
"pass_rate": 0.3651,
"quality": 0.6682,
"quality_target_met": true,
"utility": 45.082191
} | You are the Dot Loom conductor. Select the highest-utility execution plan that obeys all hard call, credit, and latency limits. Use cheap workers when sufficient. Require an independent provider when consequences and verification needs justify it. Return exactly one compact JSON object and no prose.
INPUT={"constraints... | loom-conductor-v2 | synthetic-empirical-v1 | train | {"access":{"finalizer":[],"reviewer":[],"writer":[]},"estimated_credits":0.979,"estimated_latency_ms":18362,"estimated_pass_rate":0.3651,"estimated_quality":0.6682,"finalizer":null,"independent_verification":false,"max_calls":1,"max_credits":3.244,"max_latency_ms":31406,"policy":"lean","reason_codes":["complex_task","i... | {
"ambiguity": 0.7374,
"category": "operations",
"complexity": 0.6723,
"consequence": 0.5803,
"evidence_need": 0.7241,
"family": "incident_response",
"input_tokens": 6880,
"request_id": "synthetic-incident_response-006711",
"reversibility": 0.403,
"risk": 0.5999,
"summary": "Prioritize actions dur... | [
{
"credits_per_call": 12.356,
"id": "worker_a",
"p95_latency_ms": 81862,
"pass_rate": 0.9406,
"provider_group": "provider_z",
"quality": 0.9735,
"strengths": [
"drafting",
"implementation",
"reasoning",
"synthesis"
]
},
{
"credits_per_call": 3.629,
... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.