From 63c970812b8b5c6664400c9360e22c427844820c Mon Sep 17 00:00:00 2001 From: gensokyo-dev <342580529+gensokyo-dev@users.noreply.github.com> Date: Sun, 18 Oct 2026 19:19:20 +0800 Subject: [PATCH] perf(hopper): reuse decoded records for literal string matches (#5017) Keep complete decoded string records in Hopper's existing search inventory and directly project literal matches, removing the redundant full record projection and superseded value-only representation. Record real Linux CLI/MCP correctness and Evidence checks, shorter warm queries, negative whole-task costs, native coverage gaps and resource stops. Fixes #5010. Co-authored-by: Kaoru0822 <208281322+kaoru0822-kitauji@users.noreply.github.com> --- bridge/hopper_bridge.py | 30 +- experiments/analysis-efficiency/README.md | 4 + .../hopper-string-records.json | 1563 +++++++++++++++++ .../hopper-string-records.md | 236 +++ 4 files changed, 1813 insertions(+), 20 deletions(-) create mode 100644 experiments/analysis-efficiency/hopper-string-records.json create mode 100644 experiments/analysis-efficiency/hopper-string-records.md diff --git a/bridge/hopper_bridge.py b/bridge/hopper_bridge.py index 07a8adff51..02be09c844 100644 --- a/bridge/hopper_bridge.py +++ b/bridge/hopper_bridge.py @@ -431,10 +431,6 @@ def _procedure_map(document): return result -def _strings(document): - return {address: record["value"] for address, record in _string_inventory(document).items()} - - def _string_record(segment, address, display): """Read one native typed object, retaining its shortened display separately.""" length = segment.getObjectLength(address) @@ -521,7 +517,7 @@ def _search_inventory(document, kind): if kind == "procedure": values = _procedure_map(document) elif kind == "string": - values = _strings(document) + values = _string_inventory(document) elif kind == "name": values = {_hex(address): value for address, value in _name_map(document).items()} else: @@ -548,23 +544,17 @@ def _search_results(document, kind, params): else: raise CapabilityUnavailableError("Regex matching requires REA's supervised worker; the native bridge only executes literal searches") - selected = [] - for item in _search_inventory(document, kind): - if not matches(item[1]): - continue - selected.append(item) - if not selected: - return [] if kind == "string": - records = _string_inventory(document) - return [{"address": address, **records[address]} for address, _ in selected] - return [ - { - "address": address, - "value": value, - } - for address, value in selected + return [ + {"address": address, **record} + for address, record in _search_inventory(document, kind) + if matches(record["value"]) ] + return [ + {"address": address, "value": value} + for address, value in _search_inventory(document, kind) + if matches(value) + ] def _unavailable(reason): diff --git a/experiments/analysis-efficiency/README.md b/experiments/analysis-efficiency/README.md index 6e99dd6400..e014521fdc 100644 --- a/experiments/analysis-efficiency/README.md +++ b/experiments/analysis-efficiency/README.md @@ -1,5 +1,9 @@ # Analysis efficiency measurements +For Hopper literal string queries, see the [decoded-record checkpoint](hopper-string-records.md). +It measures record reuse through real Linux CLI/MCP workflows, preserving +native recognition gaps, guard-stopped attempts and negative whole-task costs. + For JEB on Linux, see the [CE native audit](jeb-linux-native.md). It qualifies the direct headless launch route and measures selected function reads through real CLI and production stdio MCP, preserving timeout attempts and the limits diff --git a/experiments/analysis-efficiency/hopper-string-records.json b/experiments/analysis-efficiency/hopper-string-records.json new file mode 100644 index 0000000000..913d7e929e --- /dev/null +++ b/experiments/analysis-efficiency/hopper-string-records.json @@ -0,0 +1,1563 @@ +{ + "question": "Remove duplicate full-inventory projection from literal search without changing native string facts, metadata, Evidence or public CLI/MCP contracts.", + "issue": "https://github.com/morluto/rea/issues/5010", + "date_utc": "2026-10-18", + "baseline": "c12d9eda87aa5a022ee7e18ea96d8d03bfd12e96", + "rea_version": "6.7.0", + "node": "24.18.0", + "host": "Linux x86-64 Debian 12; Linux 6.1.0-53-cloud-amd64", + "native_provider": { + "reported_release_version": "6.6.0", + "official_url": "https://www.hopperapp.com:443/downloader/public/Hopper-6.6.0-Linux-demo.deb", + "archive_bytes": 35748722, + "archive_sha256": "59705e50f7ebf9104abe1e72ac3949bed1a8e43ec67a8c9510a1c5056c200d27", + "publisher_primary_fingerprint": "7C2D73BAFAF5B9A710A69DA1A26DF815099B1BCD", + "archive_and_launcher_authenticated": true, + "launcher_sha256": "b2271008fb82bf14cfae5c3644cdde2b11fc6804bd7c99d67032ec1b1942d53c" + }, + "runtime": { + "rootfs_url": "https://cdimage.ubuntu.com/ubuntu-base/releases/24.04/release/ubuntu-base-24.04.5-base-amd64.tar.gz", + "rootfs_sha256": "e77b6f10c2590cef872b33ee9f635a0e3fd1f57fb074c0e52b5c7f56147a0c86", + "rootfs_verification": "Official HTTPS SHA256SUMS match; publisher signature not independently verified.", + "glibc": "2.39-0ubuntu8.8", + "python_embedding_library": "libpython3.12t64 3.12.3-1ubuntu0.17", + "scope": "User-mapped private Ubuntu Base runtime and Xvfb; normal official demo activation. Actual rea setup/vendor postinst not executed. Targets not executed." + }, + "targets": [ + { + "name": "authored control.elf", + "bytes": 17360, + "sha256": "1a6f55303712e096ff5196c7fd6e8acf75473fbf0f7467d7b315f96f50b3f841", + "oracle": "Source input*7+13 and independent ELF symbol." + }, + { + "name": "tests/conformance/c/fixture.c ELF", + "bytes": 22584, + "sha256": "2f4b65b4cf32cd14a3d4e683204ec3e854fa92bbb15bb49897c6c94405f852c7", + "source_sha256": "e80b3a9754e19d8e1a837823555435c99a7eb3b614dcc0d71c4fc30cb4999320", + "compiler": "gcc 12.2.0-14+deb12u1", + "flags": [ + "-O0", + "-g", + "-fno-inline", + "-fno-omit-frame-pointer", + "-fno-pie", + "-no-pie" + ] + }, + { + "name": "libsqlite3.so.0.8.6", + "package": "3.40.1-2+deb12u2", + "bytes": 1437848, + "sha256": "2e6eef9a727f081f0d453b4e5e6cbd8b9ef8b6f86cbf7681cbad444d3b0b55c8", + "oracle": "Independent GNU objdump: sqlite3_libversion_number at 0xa1d50 returns 0x2e6301 = 3040001." + } + ], + "measurement_scope": "Fresh native process per task; OS caches unflushed. SDK timings include catalog and correctness assertions. Guarded family includes namespace, native provider, MCP client/runtime and cleanup/file cache. Outer guard/systemd costs excluded. Admission waits separated when recorded; older waits unmeasured.", + "resource_policy": { + "family_hard_bytes": 8589934592, + "family_early_stop_bytes": 7516192768, + "swap_bytes": 0, + "cpu_quota_percent": 100, + "cpu_affinity": [15], + "tasks_max": 192, + "host_start_min_bytes": 17179869184, + "host_stop_below_bytes": 8589934592, + "later_long_run_admission": "24 GiB for 20 seconds, unchanged family guard", + "heap_mib": 3072, + "heavy_concurrency": 1, + "counter_scope": "Verified at payload exit while observer remains alive; disappeared final scope counters remain unknown. Payload zero checks cover memory low/high/max/oom/oom_kill/oom_group_kill and pids max." + }, + "comparison_cohorts": { + "baseline": { + "runs": [ + { + "run": "sqlite-compare-baseline-1", + "sdk_seconds": 88.444565055, + "bridge_sha256": "97ec963352ae396de2e2e331eac84738836a336b57288e0a2c0b2d93fbb82a45", + "runtime_sha256": "0809b5004f214efa4b79775d5fda796eaecf4ad5756c55ee3b9d7cf82d91b750", + "calls": 17, + "phase_wall_ms": { + "open": 10740.266843, + "binary_session": 617.2116429999987, + "search-cold": 71199.76941899999, + "search-warm-0": 16.126670000012382, + "search-warm-1": 14.496440000002622, + "search-warm-2": 12.707978999998886, + "search-warm-3": 12.797219999993104, + "search-warm-4": 28.872738999998546, + "search-warm-5": 15.201450000007753, + "validation-full-inventory": 176.08521499999915, + "validation-broad": 16.511150000005728, + "validation-regex": 127.31491699999606, + "validation-case-insensitive": 12.753620000003139, + "search-absent": 10.286540000000969, + "selected-procedure": 12.330710000009276, + "decompile": 14.403501000007964, + "close": 216.15989500000433 + }, + "matches": 19, + "inventory": 1437, + "broad_matches": 74, + "target_unchanged": true, + "transcript_sha256": "dd24479b740dd6fc3d74c17b429eede16315f46d9c8204666642e7fc1a9f7c7b", + "guarded_seconds": 91.176, + "family_cpu_seconds": 89.965411, + "kernel_family_peak_bytes": 1525731328, + "sampled_family_peak_bytes": 1524944896, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null, + "warm_query_median_ms": 14.848945000005187, + "warm_six_total_ms": 100.20249800001329, + "admission": null + }, + { + "run": "sqlite-compare-baseline-2", + "sdk_seconds": 82.885563456, + "bridge_sha256": "97ec963352ae396de2e2e331eac84738836a336b57288e0a2c0b2d93fbb82a45", + "runtime_sha256": "0809b5004f214efa4b79775d5fda796eaecf4ad5756c55ee3b9d7cf82d91b750", + "calls": 17, + "phase_wall_ms": { + "open": 11215.932496000001, + "binary_session": 242.544289999998, + "search-cold": 65233.896502999996, + "search-warm-0": 17.66227099999378, + "search-warm-1": 15.684191000007559, + "search-warm-2": 13.00242000000435, + "search-warm-3": 12.874109999989741, + "search-warm-4": 13.99822999999742, + "search-warm-5": 12.18788100000529, + "validation-full-inventory": 186.7607599999901, + "validation-broad": 17.700960999995004, + "validation-regex": 121.32057700000587, + "validation-case-insensitive": 11.944520999997621, + "search-absent": 6.157979999989038, + "selected-procedure": 8.546360000007553, + "decompile": 13.74452000000747, + "close": 181.42349000000104 + }, + "matches": 19, + "inventory": 1437, + "broad_matches": 74, + "target_unchanged": true, + "transcript_sha256": "e67639cbac0dedc4490136d179ca86ffff2c8a428955170974e73fecaab16d67", + "guarded_seconds": 85.916, + "family_cpu_seconds": 85.020511, + "kernel_family_peak_bytes": 1508261888, + "sampled_family_peak_bytes": 1508261888, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null, + "warm_query_median_ms": 13.500325000000885, + "warm_six_total_ms": 85.40910299999814, + "admission": { + "minimum_host_gib": 24.0, + "stable_seconds": 20.0, + "waited_seconds": 120.0298915640451, + "host_available_bytes_at_admission": 27561779200 + } + }, + { + "run": "sqlite-compare-baseline-3", + "sdk_seconds": 86.58250158700001, + "bridge_sha256": "97ec963352ae396de2e2e331eac84738836a336b57288e0a2c0b2d93fbb82a45", + "runtime_sha256": "0809b5004f214efa4b79775d5fda796eaecf4ad5756c55ee3b9d7cf82d91b750", + "calls": 17, + "phase_wall_ms": { + "open": 11529.659282, + "binary_session": 228.66470199999821, + "search-cold": 68494.13222500001, + "search-warm-0": 109.84550499999023, + "search-warm-1": 22.96904199999699, + "search-warm-2": 15.55612999999721, + "search-warm-3": 26.612552000005962, + "search-warm-4": 45.15687200000684, + "search-warm-5": 50.505151999997906, + "validation-full-inventory": 307.9009439999936, + "validation-broad": 16.880311000000802, + "validation-regex": 118.27183499999228, + "validation-case-insensitive": 11.566659999996773, + "search-absent": 5.354089999993448, + "selected-procedure": 10.171901000008802, + "decompile": 16.196369999990566, + "close": 196.55931799999962 + }, + "matches": 19, + "inventory": 1437, + "broad_matches": 74, + "target_unchanged": true, + "transcript_sha256": "c9074060517ad0288a4a980c98d99d19f89bb1fea01862e92b862ab2cb47f1c5", + "guarded_seconds": 89.384, + "family_cpu_seconds": 88.332561, + "kernel_family_peak_bytes": 1396899840, + "sampled_family_peak_bytes": 1396563968, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null, + "warm_query_median_ms": 35.8847120000064, + "warm_six_total_ms": 270.64525299999514, + "admission": { + "minimum_host_gib": 24.0, + "stable_seconds": 20.0, + "waited_seconds": 40.005464048008434, + "host_available_bytes_at_admission": 29188427776 + } + } + ], + "medians": { + "sdk_seconds": 86.58250158700001, + "guarded_seconds": 89.384, + "family_cpu_seconds": 88.332561, + "kernel_family_peak_bytes": 1508261888, + "warm_query_median_ms": 14.848945000005187, + "warm_six_total_ms": 100.20249800001329 + } + }, + "candidate": { + "runs": [ + { + "run": "sqlite-compare-candidate-1", + "sdk_seconds": 95.29856059200002, + "bridge_sha256": "3f5693a1260748a633416edee9206503dd5deda0a6e6ea0c8d15f2418d7b281c", + "runtime_sha256": "0809b5004f214efa4b79775d5fda796eaecf4ad5756c55ee3b9d7cf82d91b750", + "calls": 17, + "phase_wall_ms": { + "open": 11201.593252000002, + "binary_session": 290.0358049999995, + "search-cold": 77533.91172999999, + "search-warm-0": 14.937599000011687, + "search-warm-1": 15.433709999997518, + "search-warm-2": 11.831369999999879, + "search-warm-3": 11.02096999999776, + "search-warm-4": 12.175889000005554, + "search-warm-5": 10.16972999999416, + "validation-full-inventory": 200.7143170000054, + "validation-broad": 13.392359999998007, + "validation-regex": 118.1646589999873, + "validation-case-insensitive": 8.557058999998844, + "search-absent": 4.477880000005825, + "selected-procedure": 10.334149999995134, + "decompile": 16.46730999999272, + "close": 199.84632699999202 + }, + "matches": 19, + "inventory": 1437, + "broad_matches": 74, + "target_unchanged": true, + "transcript_sha256": "c9ddda89ac0ed7608c2174a4ab9aa9032cb4289ef15bc013cb7c530330d045ea", + "guarded_seconds": 97.998, + "family_cpu_seconds": 96.690084, + "kernel_family_peak_bytes": 1507926016, + "sampled_family_peak_bytes": 1507921920, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null, + "warm_query_median_ms": 12.003629500002717, + "warm_six_total_ms": 75.56926800000656, + "admission": null + }, + { + "run": "sqlite-compare-candidate-2-retry1", + "sdk_seconds": 90.785647571, + "bridge_sha256": "3f5693a1260748a633416edee9206503dd5deda0a6e6ea0c8d15f2418d7b281c", + "runtime_sha256": "0809b5004f214efa4b79775d5fda796eaecf4ad5756c55ee3b9d7cf82d91b750", + "calls": 17, + "phase_wall_ms": { + "open": 11116.49335, + "binary_session": 676.7291210000003, + "search-cold": 72990.031273, + "search-warm-0": 14.271951000002446, + "search-warm-1": 13.48707000000286, + "search-warm-2": 12.43949099999736, + "search-warm-3": 10.646190999992541, + "search-warm-4": 13.080590000012307, + "search-warm-5": 11.626370999991195, + "validation-full-inventory": 204.4668299999903, + "validation-broad": 13.550380000000587, + "validation-regex": 130.55598000000464, + "validation-case-insensitive": 9.541219999999157, + "search-absent": 6.128810000009253, + "selected-procedure": 9.3794399999897, + "decompile": 14.78522000000521, + "close": 209.40417999999772 + }, + "matches": 19, + "inventory": 1437, + "broad_matches": 74, + "target_unchanged": true, + "transcript_sha256": "8a90700342abfdfebe713f96f13b548809c178b4d5e47f7fff5a68fc4e96625c", + "guarded_seconds": 93.63, + "family_cpu_seconds": 92.589971, + "kernel_family_peak_bytes": 1511862272, + "sampled_family_peak_bytes": 1509240832, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null, + "warm_query_median_ms": 12.760040500004834, + "warm_six_total_ms": 75.55166399999871, + "admission": { + "minimum_host_gib": 24.0, + "stable_seconds": 20.0, + "waited_seconds": 20.002894186065532, + "host_available_bytes_at_admission": 29044015104 + } + }, + { + "run": "sqlite-compare-candidate-3", + "sdk_seconds": 90.114488291, + "bridge_sha256": "3f5693a1260748a633416edee9206503dd5deda0a6e6ea0c8d15f2418d7b281c", + "runtime_sha256": "0809b5004f214efa4b79775d5fda796eaecf4ad5756c55ee3b9d7cf82d91b750", + "calls": 17, + "phase_wall_ms": { + "open": 11352.972276999997, + "binary_session": 240.12044999999853, + "search-cold": 72336.02312299999, + "search-warm-0": 15.078139999997802, + "search-warm-1": 15.000960999997915, + "search-warm-2": 13.200239999991027, + "search-warm-3": 9.891371000005165, + "search-warm-4": 12.056620000003022, + "search-warm-5": 7.5958709999977145, + "validation-full-inventory": 160.8151760000037, + "validation-broad": 13.689031000001705, + "validation-regex": 122.58544500000426, + "validation-case-insensitive": 7.607250000000931, + "search-absent": 5.273239999994985, + "selected-procedure": 9.556061000010232, + "decompile": 13.460630999994464, + "close": 206.01013799999782 + }, + "matches": 19, + "inventory": 1437, + "broad_matches": 74, + "target_unchanged": true, + "transcript_sha256": "bb391d389696b5dec2dfc3eb6889cde0ba3431eaa532cdeb1d4a194168e0bd45", + "guarded_seconds": 93.048, + "family_cpu_seconds": 92.203072, + "kernel_family_peak_bytes": 1509961728, + "sampled_family_peak_bytes": 1508675584, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null, + "warm_query_median_ms": 12.628429999997024, + "warm_six_total_ms": 72.82320299999265, + "admission": { + "minimum_host_gib": 24.0, + "stable_seconds": 20.0, + "waited_seconds": 20.00928954104893, + "host_available_bytes_at_admission": 28308754432 + } + } + ], + "medians": { + "sdk_seconds": 90.785647571, + "guarded_seconds": 93.63, + "family_cpu_seconds": 92.589971, + "kernel_family_peak_bytes": 1509961728, + "warm_query_median_ms": 12.628429999997024, + "warm_six_total_ms": 75.55166399999871 + } + } + }, + "evidence_equality_fields": [ + "evidence_id", + "subject", + "provider", + "analysis_profile", + "predicate_type", + "operation", + "parameters", + "raw_result", + "normalized_result", + "confidence", + "authority", + "limitations", + "relationship_coverage", + "locations", + "evidence_links" + ], + "paired_evidence_proofs": [ + { + "runs": ["sqlite-compare-baseline-1", "sqlite-compare-candidate-1"], + "same_named_requests_by_phase": true, + "equal_evidence_payloads": 14, + "environment_note": "Live environment observations are excluded from cross-session equality." + }, + { + "runs": [ + "sqlite-compare-baseline-2", + "sqlite-compare-candidate-2-retry1" + ], + "same_named_requests_by_phase": true, + "equal_evidence_payloads": 14, + "environment_note": "Live environment observations are excluded from cross-session equality." + }, + { + "runs": ["sqlite-compare-baseline-3", "sqlite-compare-candidate-3"], + "same_named_requests_by_phase": true, + "equal_evidence_payloads": 14, + "environment_note": "Live environment observations are excluded from cross-session equality." + } + ], + "separate_phase_control": { + "description": "Same 17 calls with the existing procedure_address moved before first search. Do not pool with main cohorts; unchanged lookup takes most preparation cost. No overall causal improvement established.", + "runs": [ + { + "run": "sqlite-phase-baseline-1", + "sdk_seconds": 86.371790142, + "bridge_sha256": "97ec963352ae396de2e2e331eac84738836a336b57288e0a2c0b2d93fbb82a45", + "runtime_sha256": "0809b5004f214efa4b79775d5fda796eaecf4ad5756c55ee3b9d7cf82d91b750", + "calls": 17, + "phase_wall_ms": { + "open": 11561.459470000002, + "binary_session": 258.15329500000007, + "selected-procedure": 67948.694085, + "search-cold": 329.2540949999966, + "search-warm-0": 95.90607099999033, + "search-warm-1": 41.74717000000237, + "search-warm-2": 48.135931999990134, + "search-warm-3": 44.47922100000142, + "search-warm-4": 49.83681000000797, + "search-warm-5": 80.34632100000454, + "validation-full-inventory": 194.61038300000655, + "validation-broad": 16.250620000006165, + "validation-regex": 123.2775619999884, + "validation-case-insensitive": 11.964999999996508, + "search-absent": 6.091730000000098, + "decompile": 17.39658000000054, + "close": 193.4717030000029 + }, + "matches": 19, + "inventory": 1437, + "broad_matches": 74, + "target_unchanged": true, + "transcript_sha256": "c319b6e6dc081c92a3f04bfac91cdd24df00ae7ac634c743b902755f9cc397db", + "guarded_seconds": 89.096, + "family_cpu_seconds": 88.065136, + "kernel_family_peak_bytes": 1410531328, + "sampled_family_peak_bytes": 1410342912, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null, + "warm_query_median_ms": 48.98637099999905, + "warm_six_total_ms": 360.45152499999676, + "admission": { + "minimum_host_gib": 24.0, + "stable_seconds": 20.0, + "waited_seconds": 140.03102767793462, + "host_available_bytes_at_admission": 28496936960 + } + }, + { + "run": "sqlite-phase-candidate-1", + "sdk_seconds": 111.251615902, + "bridge_sha256": "3f5693a1260748a633416edee9206503dd5deda0a6e6ea0c8d15f2418d7b281c", + "runtime_sha256": "0809b5004f214efa4b79775d5fda796eaecf4ad5756c55ee3b9d7cf82d91b750", + "calls": 17, + "phase_wall_ms": { + "open": 10239.382462, + "binary_session": 581.4653500000022, + "selected-procedure": 94357.583967, + "search-cold": 337.5330959999992, + "search-warm-0": 11.769040000013774, + "search-warm-1": 11.843040000007022, + "search-warm-2": 13.061471000008169, + "search-warm-3": 12.220719999997527, + "search-warm-4": 8.546900000001187, + "search-warm-5": 9.537460000006831, + "validation-full-inventory": 187.4057940000057, + "validation-broad": 13.757830999995349, + "validation-regex": 124.53969100001268, + "validation-case-insensitive": 7.283150000002934, + "search-absent": 5.436459000004106, + "decompile": 15.360469999999623, + "close": 199.8891730000032 + }, + "matches": 19, + "inventory": 1437, + "broad_matches": 74, + "target_unchanged": true, + "transcript_sha256": "6aaeae379f8e36920c40c645cf87d2086ee56fc4ca6e22346f264c5b8e786870", + "guarded_seconds": 113.917, + "family_cpu_seconds": 108.29572, + "kernel_family_peak_bytes": 1605947392, + "sampled_family_peak_bytes": 1602543616, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null, + "warm_query_median_ms": 11.806040000010398, + "warm_six_total_ms": 66.97863100003451, + "admission": { + "minimum_host_gib": 24.0, + "stable_seconds": 20.0, + "waited_seconds": 410.0758441670332, + "host_available_bytes_at_admission": 27743223808 + } + } + ], + "equality": { + "runs": ["sqlite-phase-baseline-1", "sqlite-phase-candidate-1"], + "same_named_requests_by_phase": true, + "equal_evidence_payloads": 14, + "environment_note": "Live environment observations are excluded from cross-session equality." + } + }, + "real_semantic_checks": [ + { + "run": "conformance-baseline-semantic-1", + "status": "passed", + "calls": 70, + "expected_errors": 6, + "coverage": { + "unicodeIdentity": true, + "literalMetacharacters": true, + "longValuePreserved": true, + "symbolUtf16LimitValidatedBeforeMutation": true, + "regexAndCaseModes": true, + "annotationCacheInvalidation": true, + "stringCoverage": { + "fullLiteralSourceBytes": 4120, + "fragmentedLiteralSourceBytes": 549, + "fullLiteralRecognition": { + "address": "0x402020", + "state": "not_observed", + "decodedObject": "unverified" + }, + "sourceBytesVerified": true, + "recognition": [ + { + "value": "REA_UTF8_é_😀", + "address": "0x403270", + "state": "not_observed", + "decodedObject": "unverified" + }, + { + "value": "REA_ESCAPED_\"\\line\nend\t\r", + "address": "0x4032ad", + "state": "observed", + "decodedObject": "verified" + }, + { + "value": "REA_LITERAL_BACKSLASH_\\n", + "address": "0x4032c6", + "state": "observed", + "decodedObject": "verified" + }, + { + "value": "REA_LITERAL_…", + "address": "0x403290", + "state": "not_observed", + "decodedObject": "unverified" + }, + { + "value": "REA_LATIN1_ÿ", + "address": "0x4032a0", + "state": "not_observed", + "decodedObject": "unverified" + } + ], + "displayEscapesExcludedFromMatching": true + } + }, + "target_unchanged": true + }, + { + "run": "conformance-candidate-semantic-1", + "status": "passed", + "calls": 70, + "expected_errors": 6, + "coverage": { + "unicodeIdentity": true, + "literalMetacharacters": true, + "longValuePreserved": true, + "symbolUtf16LimitValidatedBeforeMutation": true, + "regexAndCaseModes": true, + "annotationCacheInvalidation": true, + "stringCoverage": { + "fullLiteralSourceBytes": 4120, + "fragmentedLiteralSourceBytes": 549, + "fullLiteralRecognition": { + "address": "0x402020", + "state": "not_observed", + "decodedObject": "unverified" + }, + "sourceBytesVerified": true, + "recognition": [ + { + "value": "REA_UTF8_é_😀", + "address": "0x403270", + "state": "not_observed", + "decodedObject": "unverified" + }, + { + "value": "REA_ESCAPED_\"\\line\nend\t\r", + "address": "0x4032ad", + "state": "observed", + "decodedObject": "verified" + }, + { + "value": "REA_LITERAL_BACKSLASH_\\n", + "address": "0x4032c6", + "state": "observed", + "decodedObject": "verified" + }, + { + "value": "REA_LITERAL_…", + "address": "0x403290", + "state": "not_observed", + "decodedObject": "unverified" + }, + { + "value": "REA_LATIN1_ÿ", + "address": "0x4032a0", + "state": "not_observed", + "decodedObject": "unverified" + } + ], + "displayEscapesExcludedFromMatching": true + } + }, + "target_unchanged": true + } + ], + "cli_mcp_parity": [ + { + "run": "conformance-baseline-cli-1", + "normalized_facts_and_evidence_id_equal": true, + "stdout_bytes": 2293 + }, + { + "run": "conformance-candidate-cli-1", + "normalized_facts_and_evidence_id_equal": true, + "stdout_bytes": 2293 + } + ], + "attempt_ledger": [ + { + "attempt": "bootstrap-demo-libraries", + "receipt_sha256": "fd656f49e6cfa311bdf3c3644a2c43cc4ce688fcf86fdfbcd9bf6c98c6c9182c", + "note": "Added libXtst6 to private image; production dependency list already includes it.", + "guarded_seconds": 3.83, + "family_cpu_seconds": 2.067831, + "kernel_family_peak_bytes": 308596736, + "sampled_family_peak_bytes": 283430912, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "bootstrap-python-library", + "receipt_sha256": "d9f4c8dc60dbf3e5cfccdc0a2d4a025fb85fa28445ba88db673faf1c7c792cc1", + "note": "Installed only libpython3.12t64 into private image; unchanged baseline then passed. Installer hypothesis tracked separately in #5015.", + "guarded_seconds": 4.728, + "family_cpu_seconds": 2.005671, + "kernel_family_peak_bytes": 184156160, + "sampled_family_peak_bytes": 126746624, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "bootstrap-rootfs-configure-2", + "receipt_sha256": "fb0ce4168f954bbe9d3ac478105a57df89f51a6d2a8e0ef4f56d4aa2dbaa0815", + "note": "Pending dpkg configured; audit empty.", + "guarded_seconds": 4.504, + "family_cpu_seconds": 1.955605, + "kernel_family_peak_bytes": 158830592, + "sampled_family_peak_bytes": 158482432, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "bootstrap-rootfs-configure", + "receipt_sha256": "e249cb5add895e98157161482bcf83b20364710b615de01b4f38bb8c2bff78e0", + "note": "Harness namespace PATH omitted sbin; dpkg exited 2.", + "guarded_seconds": 0.368, + "family_cpu_seconds": 0.117568, + "kernel_family_peak_bytes": 14082048, + "sampled_family_peak_bytes": 9383936, + "abort": null, + "payload_exit": 2, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "bootstrap-rootfs", + "receipt_sha256": "6a0f9ab4481deed86fd92d44f5fe1369bf81c52e27e9fea43f510d3dff7cc4a0", + "note": "Private rootfs fontconfig postinst chown unmapped staff GID; ordinary pending configuration retry resolved it.", + "guarded_seconds": 50.672, + "family_cpu_seconds": 25.375248, + "kernel_family_peak_bytes": 996171776, + "sampled_family_peak_bytes": 976297984, + "abort": null, + "payload_exit": 100, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "check-fast-1", + "receipt_sha256": "ae5239241814e9f24dbf45ffc41687000bf1f8758526495a109c365a4c0d1815", + "note": "Separate setup, probe or verification work; not pooled into the comparison cohort.", + "guarded_seconds": 1.422, + "family_cpu_seconds": 0.8855, + "kernel_family_peak_bytes": 133808128, + "sampled_family_peak_bytes": 127303680, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "compatible-runtime-probe-2", + "receipt_sha256": "da03fcf954ae2904468a3fcaee56f6b5eb16b74d20c23f2ab52fbff9a13d8cef", + "note": "Separate setup, probe or verification work; not pooled into the comparison cohort.", + "guarded_seconds": 0.916, + "family_cpu_seconds": 0.235133, + "kernel_family_peak_bytes": 59285504, + "sampled_family_peak_bytes": 58499072, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "compile-conformance-1", + "receipt_sha256": "be0f33b7e88538a537e6785b789d8a2d94c3d67c198e449e03405409d8d07e8a", + "note": "passed-authored-source-fixture-compile", + "guarded_seconds": 0.666, + "family_cpu_seconds": 0.1918, + "kernel_family_peak_bytes": 49487872, + "sampled_family_peak_bytes": 43655168, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "conformance-baseline-cli-1", + "receipt_sha256": "d78cf08b31e85730ad7eea6d4649be331751470626a45f519b33a37cb36b7384", + "note": "passed-CLI-MCP-facts-and-identity-parity", + "guarded_seconds": 10.57, + "family_cpu_seconds": 9.683791, + "kernel_family_peak_bytes": 653844480, + "sampled_family_peak_bytes": 653340672, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "conformance-baseline-semantic-1", + "receipt_sha256": "88b2a7609ccda57835d23a1381695362cbe8689f57be323faa54aea5439431f5", + "note": "passed-real-public-semantic-workflow", + "guarded_seconds": 30.0, + "family_cpu_seconds": 27.373044, + "kernel_family_peak_bytes": 1385873408, + "sampled_family_peak_bytes": 1383739392, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "conformance-candidate-cli-1", + "receipt_sha256": "b1ee81382358652720fcb10577328d1f0cd8dd45247039c3bbda6f3be424e222", + "note": "passed-CLI-MCP-facts-and-identity-parity", + "guarded_seconds": 10.095, + "family_cpu_seconds": 9.335617, + "kernel_family_peak_bytes": 642347008, + "sampled_family_peak_bytes": 642084864, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "conformance-candidate-semantic-1", + "receipt_sha256": "76b23141f0e341c7f725fa539172bd66aea40be6ec8032dcf107308bb1382778", + "note": "passed-real-public-semantic-workflow", + "guarded_seconds": 26.5, + "family_cpu_seconds": 25.503541, + "kernel_family_peak_bytes": 1275756544, + "sampled_family_peak_bytes": 1273135104, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "control-baseline-1", + "receipt_sha256": "54f6bd05cd87b5a0f49e12f6c3ff6da2090d985b07080ca3e049f0a199509d74", + "note": "Test script passed relative path; public MCP correctly requires absolute filesystem path. No engine analysis executed; fixed harness resolve boundary.", + "guarded_seconds": 20.783, + "family_cpu_seconds": 19.227438, + "kernel_family_peak_bytes": 1073250304, + "sampled_family_peak_bytes": 1069842432, + "abort": null, + "payload_exit": 1, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "control-baseline-2", + "receipt_sha256": "7e18c0e7ba3b8dc1b154ad6aa8bcada6aa45f2c9a70bbb48c67c0edd99e57134", + "note": "demo helper lacks libXtst.so.6; provider typed runtime_dependency_unavailable exit79; production native installer already includes libxtst6, omitted in custom rootfs bootstrap. No completed analysis; no product bug claim.", + "guarded_seconds": 22.304, + "family_cpu_seconds": 21.458552, + "kernel_family_peak_bytes": 1198403584, + "sampled_family_peak_bytes": 1197539328, + "abort": null, + "payload_exit": 1, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "control-baseline-3", + "receipt_sha256": "d87ac40a9d12120df1496cc1751e712be8387d9203a9b0cf32a1dd19d30887ad", + "note": "Unchanged baseline bridge readiness timed out at 120 seconds.", + "guarded_seconds": 139.776, + "family_cpu_seconds": 28.248985, + "kernel_family_peak_bytes": 1340440576, + "sampled_family_peak_bytes": 1340391424, + "abort": null, + "payload_exit": 1, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "control-baseline-4-diagnostic", + "receipt_sha256": "1e9c1e03c62916ae8cf64b86cd1dcd1cc5f1fe2003fc878ae9523ed613eac49a", + "note": "Unchanged baseline readiness timeout; owned display showed completed analysis but Cannot launch script; private image lacked embedded Python library.", + "guarded_seconds": 139.365, + "family_cpu_seconds": 27.880969, + "kernel_family_peak_bytes": 1360601088, + "sampled_family_peak_bytes": 1360338944, + "abort": null, + "payload_exit": 1, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "control-baseline-5-python", + "receipt_sha256": "465c6f6e2f04fca945e88e40c677d26ed885d4f9e486602e7dea146c43602a93", + "note": "Adding only libpython3.12t64 made unchanged baseline bridge analyze/decompile/search/cleanup succeed; record separate private-runtime dependency omission. First completed real native Hopper task.", + "guarded_seconds": 25.779, + "family_cpu_seconds": 24.852892, + "kernel_family_peak_bytes": 1295712256, + "sampled_family_peak_bytes": 1295183872, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "docs-check-1", + "receipt_sha256": "46ed16b0ed40e4ab88634b227821484c77ea6d3dd1ac8ea5e341ab4a159c15b2", + "note": "Separate setup, probe or verification work; not pooled into the comparison cohort.", + "guarded_seconds": 3.53, + "family_cpu_seconds": 2.835321, + "kernel_family_peak_bytes": 294744064, + "sampled_family_peak_bytes": 292909056, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "doctor-control-1", + "receipt_sha256": "4ffe2b60ea0e952926c15dd1bfd4cef1dec3a46aa6f6d366232eb0abb11c91ed", + "note": "Separate setup, probe or verification work; not pooled into the comparison cohort.", + "guarded_seconds": 8.877, + "family_cpu_seconds": 6.300761, + "kernel_family_peak_bytes": 683057152, + "sampled_family_peak_bytes": 678879232, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "fetch-package", + "receipt_sha256": "102001fdd2221e460eb94522b5f306e7ab187920ae1687d5c90791d0173b7510", + "note": "Separate setup, probe or verification work; not pooled into the comparison cohort.", + "guarded_seconds": 11.831, + "family_cpu_seconds": 5.050212, + "kernel_family_peak_bytes": 382066688, + "sampled_family_peak_bytes": 381501440, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "fetch-rootfs", + "receipt_sha256": "2d8bd42287ac1c5e6c5ccf2d20609fea7b72335bdcddb9d81628653c24dc6299", + "note": "Separate setup, probe or verification work; not pooled into the comparison cohort.", + "guarded_seconds": 8.955, + "family_cpu_seconds": 2.563958, + "kernel_family_peak_bytes": 149532672, + "sampled_family_peak_bytes": 148639744, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "focused-checks-1", + "receipt_sha256": "89a5d051c45941029a7e6168e9252b63f56e10935634c5785bd8630d1ec66be0", + "note": "test:local rejected built boundary projects; zero tests ran. Rerun with test:focused passed.", + "guarded_seconds": 0.779, + "family_cpu_seconds": 0.366989, + "kernel_family_peak_bytes": 50671616, + "sampled_family_peak_bytes": 34418688, + "abort": null, + "payload_exit": 1, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "focused-checks-2", + "receipt_sha256": "8e141da838d04581067d0231d86b0b637d34f0eac26382c7141f9e44962ab79a", + "note": "Separate setup, probe or verification work; not pooled into the comparison cohort.", + "guarded_seconds": 15.808, + "family_cpu_seconds": 14.230238, + "kernel_family_peak_bytes": 759738368, + "sampled_family_peak_bytes": 756330496, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "isolation-probe", + "receipt_sha256": "45ce9c804b5d42fd4898901ab2f4f8d3f69c13757af0eb37b116452b4383eef4", + "note": "Separate setup, probe or verification work; not pooled into the comparison cohort.", + "guarded_seconds": 1.353, + "family_cpu_seconds": 0.269949, + "kernel_family_peak_bytes": 133095424, + "sampled_family_peak_bytes": 125976576, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "prepare", + "receipt_sha256": "fefe2c454223b492b181bd9245ceda7f3f98442e83bf7e0505f346a0b16d5c3d", + "note": "Separate setup, probe or verification work; not pooled into the comparison cohort.", + "guarded_seconds": 7.532, + "family_cpu_seconds": 4.530148, + "kernel_family_peak_bytes": 213495808, + "sampled_family_peak_bytes": 213233664, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "rootfs-probe", + "receipt_sha256": "e086fa06f6d3bdeaafa84b622a07cfbbe58e13d9036ee0815bdee9929fbb0f6f", + "note": "Separate setup, probe or verification work; not pooled into the comparison cohort.", + "guarded_seconds": 0.368, + "family_cpu_seconds": 0.095602, + "kernel_family_peak_bytes": 14295040, + "sampled_family_peak_bytes": 12546048, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "sqlite-baseline-1", + "receipt_sha256": "b82c25219ad7dbac26fb74d9c1ce52f1195225df31bb4a49686d075c7e87cd94", + "note": "Initial smoke harness; not paired with later comparison workload. Full task121.65s CPU115.420944s family1465540608B, cleanup verified; no OOM at payload exit.", + "guarded_seconds": 121.65, + "family_cpu_seconds": 115.420944, + "kernel_family_peak_bytes": 1465540608, + "sampled_family_peak_bytes": 1465540608, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "sqlite-compare-baseline-1", + "receipt_sha256": "dd6d672ec56ca0da12ba8d5be12c0a5ca67d7b9f20e4d57beded3926d3f1c4a4", + "note": "passed-paired-task", + "guarded_seconds": 91.176, + "family_cpu_seconds": 89.965411, + "kernel_family_peak_bytes": 1525731328, + "sampled_family_peak_bytes": 1524944896, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "sqlite-compare-baseline-2", + "receipt_sha256": "4c91d4d54fc7119d82a02b1146b193c3f768393174adeb75f739ee2ac4eeddcf", + "note": "passed-paired-task", + "guarded_seconds": 85.916, + "family_cpu_seconds": 85.020511, + "kernel_family_peak_bytes": 1508261888, + "sampled_family_peak_bytes": 1508261888, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "sqlite-compare-baseline-3", + "receipt_sha256": "78af861171bc4471c275cca2fe6d9b17d077d4015966ad9e1f108798adebaed3", + "note": "passed-paired-task", + "guarded_seconds": 89.384, + "family_cpu_seconds": 88.332561, + "kernel_family_peak_bytes": 1396899840, + "sampled_family_peak_bytes": 1396563968, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "sqlite-compare-candidate-1", + "receipt_sha256": "b6d45ad0ebf3040eaeea495ec210dabca72c44903e99be34498fcda73034faf8", + "note": "passed-paired-task", + "guarded_seconds": 97.998, + "family_cpu_seconds": 96.690084, + "kernel_family_peak_bytes": 1507926016, + "sampled_family_peak_bytes": 1507921920, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "sqlite-compare-candidate-2-retry1", + "receipt_sha256": "65d8b004ec6102eca5ef19d88ef67aff178ef8afcf6122f544adcac3b44e74da", + "note": "passed-paired-task", + "guarded_seconds": 93.63, + "family_cpu_seconds": 92.589971, + "kernel_family_peak_bytes": 1511862272, + "sampled_family_peak_bytes": 1509240832, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "sqlite-compare-candidate-2", + "receipt_sha256": "a0a8e119a971305af648fa4cba465c19bd4c0419561ea80f98d794749f273e73", + "note": "Host available memory early stop; incomplete, not scored. No payload-exit observation; sampled peak and last CPU are lower bounds.", + "guarded_seconds": 82.607, + "family_cpu_seconds": null, + "kernel_family_peak_bytes": null, + "sampled_family_peak_bytes": 1517019136, + "abort": "host_available_memory_early_stop", + "payload_exit": null, + "payload_cleanup_verified": null, + "payload_limit_events_all_zero": null, + "last_observed_cpu_seconds_if_no_payload_exit": 78.254736, + "last_observed_limit_events_if_no_payload_exit": { + "memory.events": { + "low": 0, + "high": 0, + "max": 0, + "oom": 0, + "oom_kill": 0, + "oom_group_kill": 0 + }, + "pids.events": { + "max": 0 + } + }, + "final_scope_limit_events": null + }, + { + "attempt": "sqlite-compare-candidate-3", + "receipt_sha256": "a53cd3e4510c733593b42698620fbca3ad94aaf10d446bf66d60e243c9f4e2d6", + "note": "passed-paired-task", + "guarded_seconds": 93.048, + "family_cpu_seconds": 92.203072, + "kernel_family_peak_bytes": 1509961728, + "sampled_family_peak_bytes": 1508675584, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "sqlite-phase-baseline-1", + "receipt_sha256": "a75f04939d0bc50f01ebc1fac1104f8c806b6f979da7309d60209734098ca313", + "note": "Separate setup, probe or verification work; not pooled into the comparison cohort.", + "guarded_seconds": 89.096, + "family_cpu_seconds": 88.065136, + "kernel_family_peak_bytes": 1410531328, + "sampled_family_peak_bytes": 1410342912, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "sqlite-phase-candidate-1", + "receipt_sha256": "07f5c844874d0ba051c05e5f18d3c29579d8f65f9f39f65e58cd60cc5e905477", + "note": "Separate setup, probe or verification work; not pooled into the comparison cohort.", + "guarded_seconds": 113.917, + "family_cpu_seconds": 108.29572, + "kernel_family_peak_bytes": 1605947392, + "sampled_family_peak_bytes": 1602543616, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "token-estimates-1", + "receipt_sha256": "c97826d13ffb4d6c4f7e587f146f82964bf4c572bc9df232956f44d2a4df2a70", + "note": "17 recorded task rows; final estimate must include later phase controls and preserve partial failed rows.", + "guarded_seconds": 15.733, + "family_cpu_seconds": 15.067083, + "kernel_family_peak_bytes": 181436416, + "sampled_family_peak_bytes": 181174272, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "token-estimates-final-1", + "receipt_sha256": "798459095db53dcad0d61f02116f8e33f1147a2544f682f75cd373b84a8b8d35", + "note": "Separate setup, probe or verification work; not pooled into the comparison cohort.", + "guarded_seconds": 30.132, + "family_cpu_seconds": 27.802364, + "kernel_family_peak_bytes": 141283328, + "sampled_family_peak_bytes": 140222464, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + }, + { + "attempt": "tokenizer-install-1", + "receipt_sha256": "a593624af25b72b132c8d0e459710ce8ece0a2f9ddede6e05073b044fa1c623d", + "note": "Separate setup, probe or verification work; not pooled into the comparison cohort.", + "guarded_seconds": 5.624, + "family_cpu_seconds": 4.150901, + "kernel_family_peak_bytes": 101011456, + "sampled_family_peak_bytes": 98983936, + "abort": null, + "payload_exit": 0, + "payload_cleanup_verified": true, + "payload_limit_events_all_zero": true, + "last_observed_cpu_seconds_if_no_payload_exit": null, + "last_observed_limit_events_if_no_payload_exit": null, + "final_scope_limit_events": null + } + ], + "before_scope_refusals": [ + { + "attempt": "compatible-runtime-probe", + "reason": "Host available memory below 16 GiB; no payload created." + } + ], + "executed_checks": { + "focused": { + "status": "passed", + "files": 5, + "tests": 35 + }, + "check:fast": { + "status": "passed", + "typecheck_lint": "both cached" + }, + "docs:check": { + "status": "passed", + "scope": "generated metadata checks and README consistency" + } + }, + "limits": [ + "Warm-path work removed; main whole-task elapsed and CPU were worse, with no memory or token benefit.", + "Native typed-string coverage is not binary-global source literal coverage; real UTF-16 decoding unqualified.", + "Minimal private runtime test does not qualify actual rea setup, native Debian compatibility, other operating systems, full Hopper parity or complex agent investigations.", + "Existing catalog dominates raw estimates; model usage and client filtering/replay are unknown.", + "Raw transcripts/assets remain private; this receipt contains curated facts, hashes, costs and failed-attempt outcomes." + ], + "context_estimates": { + "tokenizer": "tiktoken", + "version": "0.12.0", + "regex_version": "2026.9.29", + "encoding": "cl100k_base", + "scope": "Offline serialization estimates, one full catalog, named tool arguments and raw replies once each; raw replies may contain both text and structured copies. CLI rows add selected Node/CLI arguments and stdout. Model usage, initialization/protocol framing, prompts, rendering/filtering/replay and independent oracle context are unknown. Interrupted rows only cover persisted exchanges; in-flight invocation/reply and forced-exit stderr are unmeasured. This is provider validation, not model investigation performance.", + "runs": [ + { + "run": "conformance-baseline-semantic-1", + "status": "passed", + "coverage": "recorded_exchanges", + "catalog_tokens": 805330, + "named_request_tokens": 12312, + "raw_reply_tokens": 140985, + "text_reply_tokens": 67735, + "recorded_calls": 70, + "transcript_sha256": "96426d9a2870b7f1ee2f2038a7e0ba0ddde333302e5c6d3976721a5db64d6c08", + "raw_total_tokens": 958627 + }, + { + "run": "conformance-candidate-semantic-1", + "status": "passed", + "coverage": "recorded_exchanges", + "catalog_tokens": 805330, + "named_request_tokens": 12312, + "raw_reply_tokens": 140985, + "text_reply_tokens": 67735, + "recorded_calls": 70, + "transcript_sha256": "d97e7372c5f26c8ec9fd1a176070cb50d9354d89d981889b85efb2dd03d63fd3", + "raw_total_tokens": 958627 + }, + { + "run": "control-baseline-1", + "status": "failed", + "coverage": "recorded_exchanges", + "catalog_tokens": 805330, + "named_request_tokens": 32, + "raw_reply_tokens": 114, + "text_reply_tokens": 42, + "recorded_calls": 2, + "transcript_sha256": "5e25beedd76b3a0be71eb58ee93e7fc134438694623971321ccb7bb801864915", + "raw_total_tokens": 805476 + }, + { + "run": "control-baseline-2", + "status": "failed", + "coverage": "recorded_exchanges", + "catalog_tokens": 805330, + "named_request_tokens": 79, + "raw_reply_tokens": 27296, + "text_reply_tokens": 12759, + "recorded_calls": 4, + "transcript_sha256": "2d6dc350bbf6c0973ce4af7904cf2ad7bddd4fae0378caa13d643bd824b4ba66", + "raw_total_tokens": 832705 + }, + { + "run": "control-baseline-3", + "status": "failed", + "coverage": "recorded_exchanges", + "catalog_tokens": 805330, + "named_request_tokens": 79, + "raw_reply_tokens": 27519, + "text_reply_tokens": 13044, + "recorded_calls": 4, + "transcript_sha256": "199c69b32e641e36a9104906567aa01438963d21534719728cc9df32b26a7e00", + "raw_total_tokens": 832928 + }, + { + "run": "control-baseline-4-diagnostic", + "status": "failed", + "coverage": "recorded_exchanges", + "catalog_tokens": 805330, + "named_request_tokens": 79, + "raw_reply_tokens": 27435, + "text_reply_tokens": 12966, + "recorded_calls": 4, + "transcript_sha256": "7c73b0be748cad0405721963ddb5af43fe4e884a08fe8c08e3641aa4b3fd5e06", + "raw_total_tokens": 832844 + }, + { + "run": "control-baseline-5-python", + "status": "passed", + "coverage": "recorded_exchanges", + "catalog_tokens": 805330, + "named_request_tokens": 265, + "raw_reply_tokens": 38645, + "text_reply_tokens": 17975, + "recorded_calls": 13, + "transcript_sha256": "03d29954af5a0e27f4571ab5b1dc395e485aac1e506c54d80e466075e2939b2f", + "raw_total_tokens": 844240 + }, + { + "run": "sqlite-baseline-1", + "status": "passed", + "coverage": "recorded_exchanges", + "catalog_tokens": 805330, + "named_request_tokens": 280, + "raw_reply_tokens": 254311, + "text_reply_tokens": 120308, + "recorded_calls": 13, + "transcript_sha256": "f68cfce01e857836ccbb90ba3b2f280e60797bd1b62129e85d6ebe9f8c7a1c5f", + "raw_total_tokens": 1059921 + }, + { + "run": "sqlite-compare-baseline-1", + "status": "passed", + "coverage": "recorded_exchanges", + "catalog_tokens": 805330, + "named_request_tokens": 361, + "raw_reply_tokens": 259809, + "text_reply_tokens": 126832, + "recorded_calls": 17, + "transcript_sha256": "dd24479b740dd6fc3d74c17b429eede16315f46d9c8204666642e7fc1a9f7c7b", + "raw_total_tokens": 1065500 + }, + { + "run": "sqlite-compare-baseline-2", + "status": "passed", + "coverage": "recorded_exchanges", + "catalog_tokens": 805330, + "named_request_tokens": 361, + "raw_reply_tokens": 259809, + "text_reply_tokens": 126832, + "recorded_calls": 17, + "transcript_sha256": "e67639cbac0dedc4490136d179ca86ffff2c8a428955170974e73fecaab16d67", + "raw_total_tokens": 1065500 + }, + { + "run": "sqlite-compare-baseline-3", + "status": "passed", + "coverage": "recorded_exchanges", + "catalog_tokens": 805330, + "named_request_tokens": 361, + "raw_reply_tokens": 259805, + "text_reply_tokens": 126830, + "recorded_calls": 17, + "transcript_sha256": "c9074060517ad0288a4a980c98d99d19f89bb1fea01862e92b862ab2cb47f1c5", + "raw_total_tokens": 1065496 + }, + { + "run": "sqlite-compare-candidate-1", + "status": "passed", + "coverage": "recorded_exchanges", + "catalog_tokens": 805330, + "named_request_tokens": 361, + "raw_reply_tokens": 259809, + "text_reply_tokens": 126832, + "recorded_calls": 17, + "transcript_sha256": "c9ddda89ac0ed7608c2174a4ab9aa9032cb4289ef15bc013cb7c530330d045ea", + "raw_total_tokens": 1065500 + }, + { + "run": "sqlite-compare-candidate-2", + "status": "interrupted_no_report", + "coverage": "partial_lower_bound", + "catalog_tokens": 805330, + "named_request_tokens": 55, + "raw_reply_tokens": 26836, + "text_reply_tokens": 12419, + "recorded_calls": 2, + "transcript_sha256": "e3589467f593e2bad9d0eb98890388dc2ce1d93d50973fc78874f2ad29762137", + "raw_total_tokens": 832221 + }, + { + "run": "sqlite-compare-candidate-2-retry1", + "status": "passed", + "coverage": "recorded_exchanges", + "catalog_tokens": 805330, + "named_request_tokens": 361, + "raw_reply_tokens": 259815, + "text_reply_tokens": 126835, + "recorded_calls": 17, + "transcript_sha256": "8a90700342abfdfebe713f96f13b548809c178b4d5e47f7fff5a68fc4e96625c", + "raw_total_tokens": 1065506 + }, + { + "run": "sqlite-compare-candidate-3", + "status": "passed", + "coverage": "recorded_exchanges", + "catalog_tokens": 805330, + "named_request_tokens": 361, + "raw_reply_tokens": 259811, + "text_reply_tokens": 126833, + "recorded_calls": 17, + "transcript_sha256": "bb391d389696b5dec2dfc3eb6889cde0ba3431eaa532cdeb1d4a194168e0bd45", + "raw_total_tokens": 1065502 + }, + { + "run": "sqlite-phase-baseline-1", + "status": "passed", + "coverage": "recorded_exchanges", + "catalog_tokens": 805330, + "named_request_tokens": 361, + "raw_reply_tokens": 259815, + "text_reply_tokens": 126835, + "recorded_calls": 17, + "transcript_sha256": "c319b6e6dc081c92a3f04bfac91cdd24df00ae7ac634c743b902755f9cc397db", + "raw_total_tokens": 1065506 + }, + { + "run": "sqlite-phase-candidate-1", + "status": "passed", + "coverage": "recorded_exchanges", + "catalog_tokens": 805330, + "named_request_tokens": 361, + "raw_reply_tokens": 259811, + "text_reply_tokens": 126833, + "recorded_calls": 17, + "transcript_sha256": "6aaeae379f8e36920c40c645cf87d2086ee56fc4ca6e22346f264c5b8e786870", + "raw_total_tokens": 1065502 + }, + { + "run": "conformance-baseline-cli-1", + "status": "passed", + "interface": "CLI", + "selected_arguments_tokens": 70, + "stdout_tokens": 614, + "raw_total_tokens": 684 + }, + { + "run": "conformance-candidate-cli-1", + "status": "passed", + "interface": "CLI", + "selected_arguments_tokens": 70, + "stdout_tokens": 614, + "raw_total_tokens": 684 + } + ] + } +} diff --git a/experiments/analysis-efficiency/hopper-string-records.md b/experiments/analysis-efficiency/hopper-string-records.md new file mode 100644 index 0000000000..b6413ffa0b --- /dev/null +++ b/experiments/analysis-efficiency/hopper-string-records.md @@ -0,0 +1,236 @@ +# Hopper literal search: reuse decoded records + +For [#5010](https://github.com/morluto/rea/issues/5010), the bridge now keeps +the existing decoded records in its sorted search inventory and projects only +matches. Previously, every positive literal search called `_string_inventory` +again, sorting all native object addresses and building another complete map +before selecting records. The value-only `_strings` representation and the +intermediate match list have been removed. + +This removes repeated work at its owner. Three paired SQLite workflows found +shorter warm queries, but **did not demonstrate an overall task speedup**: +whole elapsed and CPU were worse. Token estimates were effectively unchanged; no memory +improvement was established. The [curated receipt](hopper-string-records.json) +keeps individual measurements, equality checks and failed attempts. + +## Representation and reuse boundary + +The existing `(id(document), kind)` search cache now holds references to full +records rather than just their decoded values. The existing string-record +owner still performs byte decoding, encoding inference and unknown handling. +Matching reads `record.value`; output copies the selected record with its +address, retaining native display, encoding/extent metadata and decoding +failures. The original document lifetime and annotation invalidation apply. +This does not add another cache or duplicate decoded payloads. Both CLI and +MCP use the same bridge through the existing application workflow. Regex keeps +its supervised worker implementation. + +The lookup remains a linear scan of recognized native strings. A complete +inventory is still decoded for the first exhaustive search, and returned hits +still consume output and retained Evidence. This change does not introduce +pagination, broader binary string discovery or a new context contract. + +## Pinned real-provider scope + +Baseline: REA 6.7.0 at `c12d9eda87aa5a022ee7e18ea96d8d03bfd12e96`, Node +24.18.0, Debian 12 x86-64, Linux `6.1.0-53-cloud-amd64`. All paired tasks used +the same compiled runtime SHA-256 +`0809b5004f214efa4b79775d5fda796eaecf4ad5756c55ee3b9d7cf82d91b750`. +Baseline and candidate bridge digests are recorded separately in the receipt; +the baseline file was mounted read-only over the bridge for baseline runs. + +The official [Hopper 6.6.0 Linux demo package](https://www.hopperapp.com:443/downloader/public/Hopper-6.6.0-Linux-demo.deb) +was authenticated through REA's production publisher verifier: 35,748,722 +bytes, SHA-256 +`59705e50f7ebf9104abe1e72ac3949bed1a8e43ec67a8c9510a1c5056c200d27`, +primary signing fingerprint `7C2D73BAFAF5B9A710A69DA1A26DF815099B1BCD`. +The extracted launcher digest was +`b2271008fb82bf14cfae5c3644cdde2b11fc6804bd7c99d67032ec1b1942d53c`. +Normal demo activation used an owned private Xvfb display. + +The host's glibc 2.36 could not satisfy this package's GLIBC_2.38 requirement. +Analysis therefore ran in a user-mapped private Ubuntu Base 24.04.5 namespace +with glibc 2.39, Qt6/X11 and embedded Python 3.12. The +[rootfs archive](https://cdimage.ubuntu.com/ubuntu-base/releases/24.04/release/ubuntu-base-24.04.5-base-amd64.tar.gz) +matched its official HTTPS checksum, SHA-256 +`e77b6f10c2590cef872b33ee9f635a0e3fd1f57fb074c0e52b5c7f56147a0c86`; +its publisher signature was not independently verified. Related libraries were +installed only into that private image. Actual `rea setup` and vendor package +post-install scripts were not executed. Native Debian installation and other +operating systems remain unqualified. + +| Artifact | Bytes | Independent oracle | +| ------------------------------------ | --------: | -------------------------------------------------------- | +| Small authored C ELF | 17,360 | Source `input * 7 + 13` and ELF symbol | +| Source-built conformance ELF | 22,584 | Fixture source, known literal bytes and symbol addresses | +| Shipped Debian `libsqlite3.so.0.8.6` | 1,437,848 | GNU objdump: version function returns `3040001` | + +Artifact digests, compiler and flags are in the receipt. No target was executed. +The small ELF was a baseline setup/decompilation smoke test; the conformance +ELF had baseline/candidate semantic and CLI/MCP checks. SQLite supplied the +repeated performance cohort. A constant-return oracle does not establish +whole-program correctness, ABI inference or complex-investigation effectiveness. + +## Completed task and correctness + +Each of three baseline/candidate pairs used a fresh Hopper process and the +same public stdio MCP workflow, including catalog discovery and 17 tool calls: + +1. Open the SQLite copy and inspect `binary_session`. +2. Search `sqlite_master` in case-sensitive literal mode, then repeat six times. +3. Read `list_strings` as a full-inventory validation oracle; compare all + matching records, including their metadata, with the selected result. +4. Check broader `sqlite` search, regex, case-insensitive and absent-pattern + queries against their expected native inventory scope. +5. Resolve `sqlite3_libversion_number`, decompile it, check its constant against + independent objdump, and close the binary and MCP transport. +6. Verify unchanged target bytes and cleanup of the owned process family. + +There were 1,437 recognized string objects, 19 `sqlite_master` matches and 74 +broader `sqlite` matches. Applicable decoded hit bytes were also checked +against file contents. All 42 Evidence payloads across the three pairs matched +on normalized and raw facts, identity, subject, provider/profile, authority, +confidence, limitations, relationship coverage, locations and evidence links. +Live environment observations were excluded from cross-session equality. + +The real conformance workflows each completed 70 public calls, including six +expected errors. They exercised Unicode symbol identity, literal metacharacters, +case/regex modes, annotation cache invalidation, typed string fragments, +display escaping and source-byte checks using the existing real-provider +verifier. CLI `fragment_tail` search returned the same normalized facts and +Evidence ID as MCP on both revisions (2,293 stdout bytes each). + +Native recognition remains qualified: the full 4,120-byte source literal and +some known UTF-8, ellipsis and Latin-1 literals were not recognized. Their file +bytes were verified, while decoded native-object coverage remains unverified. +The fragmented 549-byte source literal was recognized as separate objects; +matching does not pretend to join those fragments. Escaped text and literal +backslash-n were distinguished. Real UTF-16 provider decoding was not qualified; +the focused boundary fixture covers that representation separately. Empty +search results establish absence only within the native typed-object inventory. + +## Measurements + +Runs were serial and ordered B1/C1, C2/B2, B3/C3. C2 first stopped under the +host memory guard; its separate completed retry supplies the C2 row. OS caches +were not flushed. “Cold” means the first string query in a fresh provider; +“warm” means six repeats within that provider. + +Medians across the three completed sessions per revision: + +| Cost | Baseline | Candidate | +| ---------------------------------------- | ----------: | ----------: | +| Per-run median warm query | 14.849 ms | 12.628 ms | +| Six warm queries, total | 100.202 ms | 75.552 ms | +| Whole SDK workflow | 86.583 s | 90.786 s | +| Whole guarded workflow | 89.384 s | 93.630 s | +| Family CPU at payload exit | 88.333 s | 92.590 s | +| Kernel family peak, including file cache | 1,438.4 MiB | 1,440.0 MiB | + +Warm-query medians were lower in all three pairs, although one baseline run +was particularly variable. Whole guarded elapsed was about 4.7% worse and CPU +about 4.8% worse. The first query accounted for 65.2–71.2 seconds on baseline +and 72.3–77.5 seconds on candidate; that interval includes native preparation. +Do not interpret this cohort as end-to-end acceleration or dismiss the +unfavorable rows. + +A separate pair moved the **existing** `procedure_address` call before the +first string search, keeping the same 17 calls and equivalent facts/Evidence. +Most preparation cost then occurred in that unchanged lookup: 67.949 seconds +baseline versus 94.358 seconds candidate, before the changed search path ran. +The subsequent first string query took 329.254 versus 337.533 ms; total guarded +elapsed was 89.096 versus 113.917 seconds. This locates substantial cost outside +literal projection and preserves the negative whole-task result. It does not +prove its cause, establish a cold-query improvement or justify pooling the +reordered pair with the original cohort. + +## Resource and context accounting + +Every heavy task used an owned cgroup v2 family: 8 GiB hard limit, 7 GiB early +stop, zero swap, one CPU quota and affinity, 192 tasks, 3,072 MiB Node heap, +and one heavy job at a time. Host admission required 16 GiB available; tasks +stopped below 8 GiB. Later long runs added admission at 24 GiB sustained for +20 seconds without relaxing the family guard. + +Family counters include the namespace, native engine, MCP client/runtime, +validation and cleanup, including charged file cache. SDK wall time includes +catalog discovery and correctness assertions. Admission wait is outside the +guarded time and recorded separately where available; older waits and outer +guard/systemd costs were not measured. Payload-exit CPU, kernel memory peak, +limit counters and owned cleanup were verified while an observer remained. +Final counters after the scope disappeared remain unknown. Completed paired +payloads observed zero memory-limit, OOM/OOM-kill and task-limit events. + +The unchanged catalog advertised 207 tools, 3,158,927 serialized bytes, +SHA-256 `a1851aa1bfe3d72d6be6bec130519dbdfc1ee49288320cfac0144ced77755c61`. +Offline tiktoken 0.12.0 `cl100k_base` estimates for every completed SQLite +comparison were effectively unchanged (medians across three runs): + +| Serialization counted once | Baseline estimate | Candidate estimate | +| -------------------------------------------------- | ----------------: | -----------------: | +| Full catalog page array | 805,330 | 805,330 | +| Named tool requests | 361 | 361 | +| Raw replies | 259,809 | 259,811 | +| Total above | 1,065,500 | 1,065,502 | +| Text reply blocks alone, overlapping the raw count | 126,832 | 126,833 | + +Whole raw estimates ranged from 1,065,496–1,065,500 on baseline and +1,065,500–1,065,506 on candidate. The small variations include live environment +observations; there is no context reduction. + +The full inventory and extra controls are validation overhead included in +these totals. Raw replies may contain text and structured copies. Actual model +usage, prompts, initialization/protocol framing, client filtering/rendering or +replay and independent-oracle context were not measured. The receipt separates +other tasks and the interrupted transcript's persisted lower bound; it does +not count an unrecorded in-flight request or reply as observed. No model-token +saving or agent-reasoning advantage is claimed. + +## Failed attempts and verification + +The receipt preserves setup, diagnostic, failed and completed costs separately: + +- Initial private-rootfs configuration failed on an unmapped package GID, then + on the harness PATH; ordinary pending configuration completed successfully. +- One low-host-memory probe was refused before creating a payload. +- The first MCP harness used a relative path, correctly refused by the public + input boundary. A later custom-runtime launch lacked libXtst6, which the + production dependency list already includes. Both were harness/setup errors. +- Two unchanged-baseline launches timed out at bridge startup. An owned-display + observation showed completed native analysis but a Python bootstrap failure. + Adding only the missing embedded Python library made the baseline pass; + [#5015](https://github.com/morluto/rea/issues/5015) records the separate + installer dependency hypothesis and its unverified actual-installer scope. +- C2 stopped after 82.607 seconds when host available memory crossed the + unchanged stop threshold. Its sampled family peak was 1,517,019,136 bytes + and last observed CPU 78.255 seconds. Payload-exit/final counters are unknown; + it is incomplete and excluded from completed-run statistics. +- An initial `test:local` invocation rejected built boundary projects before + running tests. The documented `test:focused` invocation then passed all + 35 tests in five files, covering string search, regex, bridge facade and + real-provider verifier behavior. Those local tests do not themselves qualify + a native engine; the public native workflows above provide that evidence. + +`npm run check:fast` also passed (typecheck and lint both cached). Build and +metadata preparation passed before the native workflows. `npm run docs:check` +passed its cached metadata checks and all five README consistency tests. + +Reproduction uses `node scripts/rea.mjs --mcp` with an explicitly configured +Hopper launcher and private display. Drive the named calls above with absolute +target paths, retain full replies and compare independently verified bytes, +facts and Evidence. For the source fixture, compile +`tests/conformance/c/fixture.c` with +`gcc -O0 -g -fno-inline -fno-omit-frame-pointer -fno-pie -no-pie`. +Debug build paths can change rebuilt artifact digests. The shipped SQLite +digest pins the measured artifact rather than every build of its package. + +The focused local gate was: + +```sh +npm run test:focused -- \ + tests/boundary/providers/hopper/hopperStringSearch.test.ts \ + src/hopper/HopperRegexSearch.test.ts \ + tests/boundary/providers/hopper/hopperBridgeFacade.test.ts \ + tests/conformance/providers/hopper/realHopperSearch.test.ts \ + tests/conformance/providers/hopper/realHopperExhaustiveSearch.test.ts +```