{ "version": 1, "date": "2026-09-07", "scope": "Offline productive ACP orchestration; both versions are modernization checkpoints", "fixture": { "fixture_sha256": "f5ca65e474286a561519f536304f3b73c55680fbbb6923cb21edd9bac63127b4", "project_builder_sha256": "a71c1d57ce3a406d04bec3cd1e94074d4188a070191d78a4c5c1cf2db5f2b2c6", "benchmark_sha256": "0f9c8703d78993b5cb38902f5c53064a52366e7f851e93e27daeb75b85149c09", "conversations": 50, "history": 100 }, "execution_order": [ "baseline-2", "current-2", "current-3", "baseline-3", "baseline-4", "current-4" ], "inference": false, "implementations": { "baseline": { "source_commit": "3b3db470af8350f69e98201106e2f7238f77a932", "wheel_sha256": "99c20f6b1278f99aad6f9d8d5ba4d7e459385c2c3208b6d01617eb8a45965582", "version": "2.6.0", "package_sha256": "620d2334060a318cffb11a693ea503b721244e3ea2a3b74a93793e0edee6970b", "python": "3.13.7", "dependencies": { "agent-client-protocol": "0.11.1", "pydantic": "2.13.5", "pydantic-core": "2.46.5", "pyyaml": "6.0.3", "click": "8.5.0", "packaging": "26.3" }, "schema_sha256": "fbfb3fd8230fe6f62e4b3a093f3e3c1fd1ffbe35498b6c4239e97a3ff16ed4d2", "samples": [ { "sample": 2, "raw_report_sha256": "e38aadb307a87b3bc72a6f93ccac8a519615aa3cbc7e4551112aeae909c943d7", "elapsed_seconds": 17.09709992283024, "daemon_runtime_reads": 1145, "stage_seconds": [ 5.877149854786694, 4.922188670840114, 6.297535531921312 ], "run_count": 3, "applied_results": 3, "successful_commands": 3, "independent_test_cases": 6, "restart_unchanged": true, "roles": [ { "role": "DEVELOPER", "context_bytes": 7977, "run_offsets": { "cleanup_complete": 2.859226, "context_ready": 0.240693, "first_prompt_activity": 2.723456, "first_prompt_started": 0.430708, "first_protocol_activity": 2.708466, "process_recorded": 0.295322, "protocol_ready": 0.351972, "session_ready": 0.397137, "workspace_ready": 0.201455 }, "queue_scope": null, "queue_wait_seconds": null, "claim_to_transition_seconds": null, "preparation_validation_seconds": null, "preparation_validation_attempts": null, "application_elapsed_seconds": null, "received_to_resolution_seconds": null }, { "role": "TESTER", "context_bytes": 9437, "run_offsets": { "cleanup_complete": 2.662947, "context_ready": 0.15341, "first_prompt_activity": 2.621046, "first_prompt_started": 0.313233, "first_protocol_activity": 2.60525, "process_recorded": 0.184477, "protocol_ready": 0.239747, "session_ready": 0.287619, "workspace_ready": 0.118396 }, "queue_scope": null, "queue_wait_seconds": null, "claim_to_transition_seconds": null, "preparation_validation_seconds": null, "preparation_validation_attempts": null, "application_elapsed_seconds": null, "received_to_resolution_seconds": null }, { "role": "ARCHITECT-REVIEWER", "context_bytes": 10539, "run_offsets": { "cleanup_complete": 2.93617, "context_ready": 0.17661, "first_prompt_activity": 2.884851, "first_prompt_started": 0.357379, "first_protocol_activity": 2.864758, "process_recorded": 0.211665, "protocol_ready": 0.259278, "session_ready": 0.322943, "workspace_ready": 0.11759 }, "queue_scope": null, "queue_wait_seconds": null, "claim_to_transition_seconds": null, "preparation_validation_seconds": null, "preparation_validation_attempts": null, "application_elapsed_seconds": null, "received_to_resolution_seconds": null } ] }, { "sample": 3, "raw_report_sha256": "256752065718fe10a7988ab0a1682a6d8c57084218be5a42a4cfd2a74ba8635a", "elapsed_seconds": 17.39515896793455, "daemon_runtime_reads": 1203, "stage_seconds": [ 5.028630197048187, 6.285037057008594, 6.081349877174944 ], "run_count": 3, "applied_results": 3, "successful_commands": 3, "independent_test_cases": 6, "restart_unchanged": true, "roles": [ { "role": "DEVELOPER", "context_bytes": 7977, "run_offsets": { "cleanup_complete": 2.653891, "context_ready": 0.143174, "first_prompt_activity": 2.628181, "first_prompt_started": 0.29868, "first_protocol_activity": 2.611767, "process_recorded": 0.173294, "protocol_ready": 0.225816, "session_ready": 0.269221, "workspace_ready": 0.112007 }, "queue_scope": null, "queue_wait_seconds": null, "claim_to_transition_seconds": null, "preparation_validation_seconds": null, "preparation_validation_attempts": null, "application_elapsed_seconds": null, "received_to_resolution_seconds": null }, { "role": "TESTER", "context_bytes": 9437, "run_offsets": { "cleanup_complete": 3.368319, "context_ready": 0.159052, "first_prompt_activity": 3.16908, "first_prompt_started": 0.326796, "first_protocol_activity": 3.152801, "process_recorded": 0.199673, "protocol_ready": 0.254871, "session_ready": 0.296124, "workspace_ready": 0.11577 }, "queue_scope": null, "queue_wait_seconds": null, "claim_to_transition_seconds": null, "preparation_validation_seconds": null, "preparation_validation_attempts": null, "application_elapsed_seconds": null, "received_to_resolution_seconds": null }, { "role": "ARCHITECT-REVIEWER", "context_bytes": 10539, "run_offsets": { "cleanup_complete": 2.937281, "context_ready": 0.171298, "first_prompt_activity": 2.878213, "first_prompt_started": 0.349757, "first_protocol_activity": 2.831252, "process_recorded": 0.212819, "protocol_ready": 0.266841, "session_ready": 0.313096, "workspace_ready": 0.116973 }, "queue_scope": null, "queue_wait_seconds": null, "claim_to_transition_seconds": null, "preparation_validation_seconds": null, "preparation_validation_attempts": null, "application_elapsed_seconds": null, "received_to_resolution_seconds": null } ] }, { "sample": 4, "raw_report_sha256": "51ed640df220a24b2fc4ede20ea43e354e289d1eb38a84e6efa6e1c9f305b4b3", "elapsed_seconds": 16.930358740035444, "daemon_runtime_reads": 1145, "stage_seconds": [ 5.994538901140913, 5.79529067594558, 5.140361288096756 ], "run_count": 3, "applied_results": 3, "successful_commands": 3, "independent_test_cases": 6, "restart_unchanged": true, "roles": [ { "role": "DEVELOPER", "context_bytes": 7977, "run_offsets": { "cleanup_complete": 2.832187, "context_ready": 0.145805, "first_prompt_activity": 2.638354, "first_prompt_started": 0.325765, "first_protocol_activity": 2.623368, "process_recorded": 0.189997, "protocol_ready": 0.250229, "session_ready": 0.289712, "workspace_ready": 0.111096 }, "queue_scope": null, "queue_wait_seconds": null, "claim_to_transition_seconds": null, "preparation_validation_seconds": null, "preparation_validation_attempts": null, "application_elapsed_seconds": null, "received_to_resolution_seconds": null }, { "role": "TESTER", "context_bytes": 9437, "run_offsets": { "cleanup_complete": 2.856704, "context_ready": 0.158358, "first_prompt_activity": 2.833156, "first_prompt_started": 0.340331, "first_protocol_activity": 2.803803, "process_recorded": 0.207618, "protocol_ready": 0.262444, "session_ready": 0.312182, "workspace_ready": 0.120269 }, "queue_scope": null, "queue_wait_seconds": null, "claim_to_transition_seconds": null, "preparation_validation_seconds": null, "preparation_validation_attempts": null, "application_elapsed_seconds": null, "received_to_resolution_seconds": null }, { "role": "ARCHITECT-REVIEWER", "context_bytes": 10539, "run_offsets": { "cleanup_complete": 2.721087, "context_ready": 0.14991, "first_prompt_activity": 2.703034, "first_prompt_started": 0.33039, "first_protocol_activity": 2.690939, "process_recorded": 0.191002, "protocol_ready": 0.24791, "session_ready": 0.286185, "workspace_ready": 0.119691 }, "queue_scope": null, "queue_wait_seconds": null, "claim_to_transition_seconds": null, "preparation_validation_seconds": null, "preparation_validation_attempts": null, "application_elapsed_seconds": null, "received_to_resolution_seconds": null } ] } ], "median_seconds": 17.09709992283024, "median_daemon_runtime_reads": 1145 }, "current": { "source_commit": "4961016641140c0d331d8537de5422b85ebec282", "wheel_sha256": "5fb7bd1a13acaac3fc0eb19f0a1842210c0e468507ed92087ed666c2ca818fd3", "version": "2.6.0", "package_sha256": "17075ae4e78a9cebe591b24d77156a901c84949df4b00d9de450ac6ce5a7bf9c", "python": "3.13.7", "dependencies": { "agent-client-protocol": "0.11.1", "pydantic": "2.13.5", "pydantic-core": "2.46.5", "pyyaml": "6.0.3", "click": "8.5.0", "packaging": "26.3" }, "schema_sha256": "fbfb3fd8230fe6f62e4b3a093f3e3c1fd1ffbe35498b6c4239e97a3ff16ed4d2", "samples": [ { "sample": 2, "raw_report_sha256": "c89d9755212cde435f799f0ff0220f0cd315feda86a265ad1bdbdf610e6f5f45", "elapsed_seconds": 16.567118116188794, "daemon_runtime_reads": 355, "stage_seconds": [ 5.602873431053013, 5.372635341947898, 5.591440385906026 ], "run_count": 3, "applied_results": 3, "successful_commands": 3, "independent_test_cases": 6, "restart_unchanged": true, "roles": [ { "role": "DEVELOPER", "context_bytes": 7972, "run_offsets": { "cleanup_complete": 2.73223, "context_ready": 0.152894, "first_prompt_activity": 2.634241, "first_prompt_started": 0.356788, "first_protocol_activity": 2.623786, "process_recorded": 0.184551, "protocol_ready": 0.231932, "session_ready": 0.303902, "workspace_ready": 0.124417 }, "queue_scope": "observed_revision_wait", "queue_wait_seconds": 0.026003599166870117, "claim_to_transition_seconds": 3.9639134407043457, "preparation_validation_seconds": 0.2249331851489842, "preparation_validation_attempts": 1, "application_elapsed_seconds": 0.1686544418334961, "received_to_resolution_seconds": 1.3751318454742432 }, { "role": "TESTER", "context_bytes": 9432, "run_offsets": { "cleanup_complete": 2.684469, "context_ready": 0.146486, "first_prompt_activity": 2.520719, "first_prompt_started": 0.401355, "first_protocol_activity": 2.498122, "process_recorded": 0.188417, "protocol_ready": 0.24025, "session_ready": 0.335243, "workspace_ready": 0.112327 }, "queue_scope": "accepted_transition_wait", "queue_wait_seconds": 1.6025450229644775, "claim_to_transition_seconds": 3.81892466545105, "preparation_validation_seconds": 0.1721218810416758, "preparation_validation_attempts": 1, "application_elapsed_seconds": 0.16267681121826172, "received_to_resolution_seconds": 1.352360486984253 }, { "role": "ARCHITECT-REVIEWER", "context_bytes": 10533, "run_offsets": { "cleanup_complete": 2.769975, "context_ready": 0.158611, "first_prompt_activity": 2.581121, "first_prompt_started": 0.406326, "first_protocol_activity": 2.553403, "process_recorded": 0.198762, "protocol_ready": 0.256437, "session_ready": 0.351187, "workspace_ready": 0.117731 }, "queue_scope": "accepted_transition_wait", "queue_wait_seconds": 1.5724830627441406, "claim_to_transition_seconds": 3.9620423316955566, "preparation_validation_seconds": 0.1317164390347898, "preparation_validation_attempts": 1, "application_elapsed_seconds": 0.25601863861083984, "received_to_resolution_seconds": 1.5144360065460205 } ] }, { "sample": 3, "raw_report_sha256": "6a04e0dcf6b6dfa5b62bee1d5cbea840408f97c5efb7e46343bed51aa7f451e4", "elapsed_seconds": 16.691363190067932, "daemon_runtime_reads": 355, "stage_seconds": [ 5.601948914118111, 5.428259026957676, 5.66097871796228 ], "run_count": 3, "applied_results": 3, "successful_commands": 3, "independent_test_cases": 6, "restart_unchanged": true, "roles": [ { "role": "DEVELOPER", "context_bytes": 7972, "run_offsets": { "cleanup_complete": 2.778037, "context_ready": 0.145743, "first_prompt_activity": 2.693973, "first_prompt_started": 0.37664, "first_protocol_activity": 2.672271, "process_recorded": 0.187853, "protocol_ready": 0.251126, "session_ready": 0.319943, "workspace_ready": 0.11957 }, "queue_scope": "observed_revision_wait", "queue_wait_seconds": 0.024758100509643555, "claim_to_transition_seconds": 4.010199546813965, "preparation_validation_seconds": 0.23787082522176206, "preparation_validation_attempts": 1, "application_elapsed_seconds": 0.15726327896118164, "received_to_resolution_seconds": 1.373368740081787 }, { "role": "TESTER", "context_bytes": 9432, "run_offsets": { "cleanup_complete": 2.911186, "context_ready": 0.153227, "first_prompt_activity": 2.729626, "first_prompt_started": 0.388338, "first_protocol_activity": 2.691989, "process_recorded": 0.197132, "protocol_ready": 0.246531, "session_ready": 0.337234, "workspace_ready": 0.112272 }, "queue_scope": "accepted_transition_wait", "queue_wait_seconds": 1.617194652557373, "claim_to_transition_seconds": 3.8325514793395996, "preparation_validation_seconds": 0.17137938598170877, "preparation_validation_attempts": 1, "application_elapsed_seconds": 0.15537333488464355, "received_to_resolution_seconds": 1.1752161979675293 }, { "role": "ARCHITECT-REVIEWER", "context_bytes": 10533, "run_offsets": { "cleanup_complete": 2.919695, "context_ready": 0.155131, "first_prompt_activity": 2.745194, "first_prompt_started": 0.41089, "first_protocol_activity": 2.728571, "process_recorded": 0.197052, "protocol_ready": 0.262253, "session_ready": 0.352105, "workspace_ready": 0.122715 }, "queue_scope": "accepted_transition_wait", "queue_wait_seconds": 1.5886554718017578, "claim_to_transition_seconds": 4.008255958557129, "preparation_validation_seconds": 0.1507945330813527, "preparation_validation_attempts": 1, "application_elapsed_seconds": 0.2516670227050781, "received_to_resolution_seconds": 1.3149151802062988 } ] }, { "sample": 4, "raw_report_sha256": "1f18ad4b0b6e25d3b9f08e86c09b208972bef15970f95c7e08b34395997fbe09", "elapsed_seconds": 16.45087797101587, "daemon_runtime_reads": 355, "stage_seconds": [ 5.1625477720517665, 5.507202971959487, 5.780966232996434 ], "run_count": 3, "applied_results": 3, "successful_commands": 3, "independent_test_cases": 6, "restart_unchanged": true, "roles": [ { "role": "DEVELOPER", "context_bytes": 7972, "run_offsets": { "cleanup_complete": 2.900262, "context_ready": 0.153036, "first_prompt_activity": 2.660402, "first_prompt_started": 0.37394, "first_protocol_activity": 2.643759, "process_recorded": 0.184741, "protocol_ready": 0.242146, "session_ready": 0.315025, "workspace_ready": 0.112766 }, "queue_scope": "observed_revision_wait", "queue_wait_seconds": 0.02086329460144043, "claim_to_transition_seconds": 3.5599312782287598, "preparation_validation_seconds": 0.23581884405575693, "preparation_validation_attempts": 1, "application_elapsed_seconds": 0.1606307029724121, "received_to_resolution_seconds": 0.9553203582763672 }, { "role": "TESTER", "context_bytes": 9432, "run_offsets": { "cleanup_complete": 2.891119, "context_ready": 0.15193, "first_prompt_activity": 2.713767, "first_prompt_started": 0.392746, "first_protocol_activity": 2.676384, "process_recorded": 0.178632, "protocol_ready": 0.239725, "session_ready": 0.332142, "workspace_ready": 0.113255 }, "queue_scope": "accepted_transition_wait", "queue_wait_seconds": 1.6266589164733887, "claim_to_transition_seconds": 3.8205506801605225, "preparation_validation_seconds": 0.17605566699057817, "preparation_validation_attempts": 1, "application_elapsed_seconds": 0.17109417915344238, "received_to_resolution_seconds": 1.1819748878479004 }, { "role": "ARCHITECT-REVIEWER", "context_bytes": 10533, "run_offsets": { "cleanup_complete": 2.815369, "context_ready": 0.148555, "first_prompt_activity": 2.713446, "first_prompt_started": 0.398304, "first_protocol_activity": 2.699481, "process_recorded": 0.181435, "protocol_ready": 0.241984, "session_ready": 0.334022, "workspace_ready": 0.115168 }, "queue_scope": "accepted_transition_wait", "queue_wait_seconds": 1.6745591163635254, "claim_to_transition_seconds": 4.125742197036743, "preparation_validation_seconds": 0.13687382289208472, "preparation_validation_attempts": 1, "application_elapsed_seconds": 0.25287342071533203, "received_to_resolution_seconds": 1.4561073780059814 } ] } ], "median_seconds": 16.567118116188794, "median_daemon_runtime_reads": 355 } }, "limitations": [ "Deterministic fixture agents, supplied plan, local checks and merge; no inference or model-quality measurement.", "Baseline 3b3db47 is already modernized; this is not an original pre-modernization pipeline measurement.", "Three fresh projects per version on one host; small latency differences are descriptive, not a statistical guarantee.", "Read counts cover RunStore reads in the daemon process, excluding subprocesses and conversation journals.", "Setup and final independent validation/restart assertions are outside the timed three-stage pipeline.", "Intermediate changes include protocol evidence, retention, usage budgets and timing; do not attribute every difference to one optimization.", "Missing baseline queue/validation/progress measurements remain null rather than reconstructed." ], "comparison": { "observed_median_wall_time_reduction_percent": 3.0998345276893713, "median_daemon_runtime_read_reduction_percent": 68.99563318777292 } }