From 7f780ef530f0d9bdcc3ce1ddaedf709a3728ecaf Mon Sep 17 00:00:00 2001 From: Garry Tan Date: Tue, 25 Aug 2026 16:03:53 +0000 Subject: [PATCH] chore(test): carve floors/ceilings recomputed; baseline + ratchet follow Phase 2 (OV9) All 9 carved skills re-anchored to post-Phase-2 measurements (cso's union had tripped its 72,000 floor at 71,379; design-consultation had 252B of margin). maxSkeletonBytes ceilings tightened to measured+~600B. Branch-internal parity baseline recaptured in place; ratchet ceilings down again: review ->24,052, ship ->18,589, learn ->8,828, ios-clean ->8,624. Co-Authored-By: Claude Fable 5 --- test/fixtures/context-budget.json | 98 ++--- test/fixtures/parity-baseline-v1.69.1.0.json | 354 +++++++++---------- test/helpers/carve-guards.ts | 36 +- 3 files changed, 244 insertions(+), 244 deletions(-) diff --git a/test/fixtures/context-budget.json b/test/fixtures/context-budget.json index 78ac32b5d..5ecbf41d6 100644 --- a/test/fixtures/context-budget.json +++ b/test/fixtures/context-budget.json @@ -2,64 +2,64 @@ "_comment": "Context-budget ratchet ceilings (~tokens). Regenerate: bun test/helpers/capture-context-budget.ts. Headroom: alwaysOnTotal x1.05, eagerPerInvocation x1.1. Graded by test/context-budget-ratchet.test.ts via lib/context-bill.ts checkBudget.", "alwaysOnTotal": 6344, "eagerPerInvocation": { - "autoplan": 24328, - "benchmark": 6997, - "benchmark-models": 5969, - "browse": 12352, + "autoplan": 22188, + "benchmark": 4856, + "benchmark-models": 3829, + "browse": 10212, "browser-skills/hackernews-frontpage": 371, - "canary": 12375, + "canary": 10235, "careful": 919, - "codex": 23643, - "context-restore": 11389, - "context-save": 12006, - "cso": 16979, - "design-consultation": 15440, - "design-html": 17456, - "design-review": 26021, - "design-shotgun": 16486, - "devex-review": 17379, - "diagram": 5987, - "document-generate": 14134, - "document-release": 11931, + "codex": 21502, + "context-restore": 9249, + "context-save": 9865, + "cso": 14838, + "design-consultation": 13300, + "design-html": 15315, + "design-review": 23881, + "design-shotgun": 14346, + "devex-review": 15239, + "diagram": 3847, + "document-generate": 11993, + "document-release": 9790, "freeze": 990, - "gstack": 5689, + "gstack": 3548, "gstack-upgrade": 3981, "guard": 889, - "health": 12587, - "investigate": 13240, - "ios-clean": 10764, - "ios-design-review": 10945, - "ios-fix": 10718, - "ios-qa": 13014, - "ios-sync": 10888, - "land-and-deploy": 26253, - "landing-report": 11299, - "learn": 10969, - "make-pdf": 7097, - "office-hours": 23646, - "open-gstack-browser": 6569, + "health": 10447, + "investigate": 11100, + "ios-clean": 8624, + "ios-design-review": 8805, + "ios-fix": 8577, + "ios-qa": 10874, + "ios-sync": 8748, + "land-and-deploy": 24113, + "landing-report": 9158, + "learn": 8828, + "make-pdf": 4957, + "office-hours": 21505, + "open-gstack-browser": 4429, "openclaw/skills/gstack-openclaw-ceo-review": 2764, "openclaw/skills/gstack-openclaw-investigate": 1429, "openclaw/skills/gstack-openclaw-office-hours": 4433, "openclaw/skills/gstack-openclaw-retro": 2542, - "pair-agent": 13319, - "plan-ceo-review": 21314, - "plan-design-review": 20755, - "plan-devex-review": 18596, - "plan-eng-review": 15535, - "plan-tune": 16543, - "qa": 20328, - "qa-only": 14809, - "retro": 21767, - "review": 26192, - "scrape": 6079, - "setup-browser-cookies": 5260, - "setup-deploy": 12129, - "setup-gbrain": 22133, - "ship": 20730, - "skillify": 13974, - "spec": 19544, - "sync-gbrain": 15757, + "pair-agent": 11178, + "plan-ceo-review": 19174, + "plan-design-review": 18615, + "plan-devex-review": 16455, + "plan-eng-review": 13395, + "plan-tune": 14402, + "qa": 18187, + "qa-only": 12669, + "retro": 19627, + "review": 24052, + "scrape": 3939, + "setup-browser-cookies": 3119, + "setup-deploy": 9989, + "setup-gbrain": 19992, + "ship": 18589, + "skillify": 11834, + "spec": 17404, + "sync-gbrain": 13616, "unfreeze": 393 } } diff --git a/test/fixtures/parity-baseline-v1.69.1.0.json b/test/fixtures/parity-baseline-v1.69.1.0.json index 484a58139..c36c36255 100644 --- a/test/fixtures/parity-baseline-v1.69.1.0.json +++ b/test/fixtures/parity-baseline-v1.69.1.0.json @@ -1,17 +1,17 @@ { "tag": "v1.69.1.0", - "capturedAt": "2026-08-25T15:19:35.204Z", - "capturedFromCommit": "6ed09078", + "capturedAt": "2026-08-25T15:47:29.737Z", + "capturedFromCommit": "135bf5d4", "capturedFromBranch": "prompt-token-load-reduction", "totalSkills": 53, - "totalCorpusBytes": 3234749, + "totalCorpusBytes": 2841533, "estTotalCatalogTokens": 4195, "topHeaviest": [ { "skill": "ship", - "skillMdBytes": 209656, - "skillMdLines": 1286, - "estTokens": 52414, + "skillMdBytes": 201464, + "skillMdLines": 1102, + "estTokens": 50366, "tmplBytes": 29080, "descriptionLen": 293, "hasGateEval": true, @@ -19,9 +19,9 @@ }, { "skill": "plan-ceo-review", - "skillMdBytes": 145538, - "skillMdLines": 1310, - "estTokens": 36385, + "skillMdBytes": 137346, + "skillMdLines": 1126, + "estTokens": 34337, "tmplBytes": 29028, "descriptionLen": 764, "hasGateEval": true, @@ -29,9 +29,9 @@ }, { "skill": "office-hours", - "skillMdBytes": 126047, - "skillMdLines": 1521, - "estTokens": 31512, + "skillMdBytes": 117855, + "skillMdLines": 1337, + "estTokens": 29464, "tmplBytes": 30566, "descriptionLen": 860, "hasGateEval": true, @@ -39,9 +39,9 @@ }, { "skill": "plan-eng-review", - "skillMdBytes": 119102, - "skillMdLines": 892, - "estTokens": 29776, + "skillMdBytes": 110910, + "skillMdLines": 708, + "estTokens": 27728, "tmplBytes": 13955, "descriptionLen": 201, "hasGateEval": true, @@ -49,9 +49,9 @@ }, { "skill": "plan-devex-review", - "skillMdBytes": 119025, - "skillMdLines": 1294, - "estTokens": 29756, + "skillMdBytes": 110833, + "skillMdLines": 1110, + "estTokens": 27708, "tmplBytes": 18598, "descriptionLen": 220, "hasGateEval": true, @@ -59,9 +59,9 @@ }, { "skill": "plan-design-review", - "skillMdBytes": 118485, - "skillMdLines": 1331, - "estTokens": 29621, + "skillMdBytes": 110293, + "skillMdLines": 1147, + "estTokens": 27573, "tmplBytes": 18463, "descriptionLen": 218, "hasGateEval": true, @@ -69,9 +69,9 @@ }, { "skill": "land-and-deploy", - "skillMdBytes": 100476, - "skillMdLines": 1818, - "estTokens": 25119, + "skillMdBytes": 92284, + "skillMdLines": 1634, + "estTokens": 23071, "tmplBytes": 54352, "descriptionLen": 160, "hasGateEval": true, @@ -79,9 +79,9 @@ }, { "skill": "review", - "skillMdBytes": 100242, - "skillMdLines": 1684, - "estTokens": 25061, + "skillMdBytes": 92050, + "skillMdLines": 1500, + "estTokens": 23013, "tmplBytes": 14141, "descriptionLen": 205, "hasGateEval": true, @@ -89,9 +89,9 @@ }, { "skill": "design-review", - "skillMdBytes": 99589, - "skillMdLines": 1835, - "estTokens": 24897, + "skillMdBytes": 91397, + "skillMdLines": 1651, + "estTokens": 22849, "tmplBytes": 11674, "descriptionLen": 306, "hasGateEval": true, @@ -99,9 +99,9 @@ }, { "skill": "autoplan", - "skillMdBytes": 93109, - "skillMdLines": 1686, - "estTokens": 23277, + "skillMdBytes": 84917, + "skillMdLines": 1502, + "estTokens": 21229, "tmplBytes": 46355, "descriptionLen": 336, "hasGateEval": true, @@ -111,9 +111,9 @@ "skills": { "autoplan": { "skill": "autoplan", - "skillMdBytes": 93109, - "skillMdLines": 1686, - "estTokens": 23277, + "skillMdBytes": 84917, + "skillMdLines": 1502, + "estTokens": 21229, "tmplBytes": 46355, "descriptionLen": 336, "hasGateEval": true, @@ -121,9 +121,9 @@ }, "benchmark": { "skill": "benchmark", - "skillMdBytes": 26777, - "skillMdLines": 591, - "estTokens": 6694, + "skillMdBytes": 18585, + "skillMdLines": 407, + "estTokens": 4646, "tmplBytes": 9378, "descriptionLen": 213, "hasGateEval": true, @@ -131,9 +131,9 @@ }, "benchmark-models": { "skill": "benchmark-models", - "skillMdBytes": 22844, - "skillMdLines": 466, - "estTokens": 5711, + "skillMdBytes": 14652, + "skillMdLines": 282, + "estTokens": 3663, "tmplBytes": 6631, "descriptionLen": 217, "hasGateEval": false, @@ -141,9 +141,9 @@ }, "browse": { "skill": "browse", - "skillMdBytes": 47273, - "skillMdLines": 862, - "estTokens": 11818, + "skillMdBytes": 39081, + "skillMdLines": 678, + "estTokens": 9770, "tmplBytes": 15864, "descriptionLen": 181, "hasGateEval": true, @@ -151,9 +151,9 @@ }, "canary": { "skill": "canary", - "skillMdBytes": 47362, - "skillMdLines": 862, - "estTokens": 11841, + "skillMdBytes": 39170, + "skillMdLines": 678, + "estTokens": 9793, "tmplBytes": 8033, "descriptionLen": 180, "hasGateEval": true, @@ -171,9 +171,9 @@ }, "codex": { "skill": "codex", - "skillMdBytes": 90485, - "skillMdLines": 1555, - "estTokens": 22621, + "skillMdBytes": 82293, + "skillMdLines": 1371, + "estTokens": 20573, "tmplBytes": 43658, "descriptionLen": 187, "hasGateEval": true, @@ -181,9 +181,9 @@ }, "context-restore": { "skill": "context-restore", - "skillMdBytes": 43587, - "skillMdLines": 753, - "estTokens": 10897, + "skillMdBytes": 35395, + "skillMdLines": 569, + "estTokens": 8849, "tmplBytes": 7092, "descriptionLen": 238, "hasGateEval": true, @@ -191,9 +191,9 @@ }, "context-save": { "skill": "context-save", - "skillMdBytes": 45947, - "skillMdLines": 838, - "estTokens": 11487, + "skillMdBytes": 37755, + "skillMdLines": 654, + "estTokens": 9439, "tmplBytes": 9293, "descriptionLen": 168, "hasGateEval": true, @@ -201,9 +201,9 @@ }, "cso": { "skill": "cso", - "skillMdBytes": 79571, - "skillMdLines": 1095, - "estTokens": 19893, + "skillMdBytes": 71379, + "skillMdLines": 911, + "estTokens": 17845, "tmplBytes": 21724, "descriptionLen": 196, "hasGateEval": true, @@ -211,9 +211,9 @@ }, "design-consultation": { "skill": "design-consultation", - "skillMdBytes": 80444, - "skillMdLines": 1040, - "estTokens": 20111, + "skillMdBytes": 72252, + "skillMdLines": 856, + "estTokens": 18063, "tmplBytes": 9554, "descriptionLen": 890, "hasGateEval": true, @@ -221,9 +221,9 @@ }, "design-html": { "skill": "design-html", - "skillMdBytes": 66806, - "skillMdLines": 1321, - "estTokens": 16702, + "skillMdBytes": 58614, + "skillMdLines": 1137, + "estTokens": 14654, "tmplBytes": 22567, "descriptionLen": 235, "hasGateEval": true, @@ -231,9 +231,9 @@ }, "design-review": { "skill": "design-review", - "skillMdBytes": 99589, - "skillMdLines": 1835, - "estTokens": 24897, + "skillMdBytes": 91397, + "skillMdLines": 1651, + "estTokens": 22849, "tmplBytes": 11674, "descriptionLen": 306, "hasGateEval": true, @@ -241,9 +241,9 @@ }, "design-shotgun": { "skill": "design-shotgun", - "skillMdBytes": 63095, - "skillMdLines": 1183, - "estTokens": 15774, + "skillMdBytes": 54903, + "skillMdLines": 999, + "estTokens": 13726, "tmplBytes": 13331, "descriptionLen": 788, "hasGateEval": true, @@ -251,9 +251,9 @@ }, "devex-review": { "skill": "devex-review", - "skillMdBytes": 66513, - "skillMdLines": 1116, - "estTokens": 16628, + "skillMdBytes": 58321, + "skillMdLines": 932, + "estTokens": 14580, "tmplBytes": 7984, "descriptionLen": 201, "hasGateEval": false, @@ -261,9 +261,9 @@ }, "diagram": { "skill": "diagram", - "skillMdBytes": 22912, - "skillMdLines": 466, - "estTokens": 5728, + "skillMdBytes": 14720, + "skillMdLines": 282, + "estTokens": 3680, "tmplBytes": 6732, "descriptionLen": 359, "hasGateEval": true, @@ -271,9 +271,9 @@ }, "document-generate": { "skill": "document-generate", - "skillMdBytes": 54091, - "skillMdLines": 1062, - "estTokens": 13523, + "skillMdBytes": 45899, + "skillMdLines": 878, + "estTokens": 11475, "tmplBytes": 15940, "descriptionLen": 334, "hasGateEval": false, @@ -281,9 +281,9 @@ }, "document-release": { "skill": "document-release", - "skillMdBytes": 71210, - "skillMdLines": 771, - "estTokens": 17803, + "skillMdBytes": 63018, + "skillMdLines": 587, + "estTokens": 15755, "tmplBytes": 6688, "descriptionLen": 192, "hasGateEval": true, @@ -321,9 +321,9 @@ }, "health": { "skill": "health", - "skillMdBytes": 48173, - "skillMdLines": 886, - "estTokens": 12043, + "skillMdBytes": 39981, + "skillMdLines": 702, + "estTokens": 9995, "tmplBytes": 11617, "descriptionLen": 184, "hasGateEval": true, @@ -331,9 +331,9 @@ }, "investigate": { "skill": "investigate", - "skillMdBytes": 50671, - "skillMdLines": 886, - "estTokens": 12668, + "skillMdBytes": 42479, + "skillMdLines": 702, + "estTokens": 10620, "tmplBytes": 11566, "descriptionLen": 1241, "hasGateEval": true, @@ -341,9 +341,9 @@ }, "ios-clean": { "skill": "ios-clean", - "skillMdBytes": 41196, - "skillMdLines": 684, - "estTokens": 10299, + "skillMdBytes": 33004, + "skillMdLines": 500, + "estTokens": 8251, "tmplBytes": 3743, "descriptionLen": 254, "hasGateEval": false, @@ -351,9 +351,9 @@ }, "ios-design-review": { "skill": "ios-design-review", - "skillMdBytes": 41888, - "skillMdLines": 687, - "estTokens": 10472, + "skillMdBytes": 33696, + "skillMdLines": 503, + "estTokens": 8424, "tmplBytes": 4417, "descriptionLen": 209, "hasGateEval": false, @@ -361,9 +361,9 @@ }, "ios-fix": { "skill": "ios-fix", - "skillMdBytes": 41017, - "skillMdLines": 683, - "estTokens": 10254, + "skillMdBytes": 32825, + "skillMdLines": 499, + "estTokens": 8206, "tmplBytes": 3574, "descriptionLen": 187, "hasGateEval": false, @@ -371,9 +371,9 @@ }, "ios-qa": { "skill": "ios-qa", - "skillMdBytes": 49808, - "skillMdLines": 840, - "estTokens": 12452, + "skillMdBytes": 41616, + "skillMdLines": 656, + "estTokens": 10404, "tmplBytes": 12370, "descriptionLen": 223, "hasGateEval": true, @@ -381,9 +381,9 @@ }, "ios-sync": { "skill": "ios-sync", - "skillMdBytes": 41670, - "skillMdLines": 681, - "estTokens": 10418, + "skillMdBytes": 33478, + "skillMdLines": 497, + "estTokens": 8370, "tmplBytes": 4220, "descriptionLen": 269, "hasGateEval": true, @@ -391,9 +391,9 @@ }, "land-and-deploy": { "skill": "land-and-deploy", - "skillMdBytes": 100476, - "skillMdLines": 1818, - "estTokens": 25119, + "skillMdBytes": 92284, + "skillMdLines": 1634, + "estTokens": 23071, "tmplBytes": 54352, "descriptionLen": 160, "hasGateEval": true, @@ -401,9 +401,9 @@ }, "landing-report": { "skill": "landing-report", - "skillMdBytes": 43242, - "skillMdLines": 729, - "estTokens": 10811, + "skillMdBytes": 35050, + "skillMdLines": 545, + "estTokens": 8763, "tmplBytes": 6847, "descriptionLen": 195, "hasGateEval": false, @@ -411,9 +411,9 @@ }, "learn": { "skill": "learn", - "skillMdBytes": 41979, - "skillMdLines": 763, - "estTokens": 10495, + "skillMdBytes": 33787, + "skillMdLines": 579, + "estTokens": 8447, "tmplBytes": 5594, "descriptionLen": 178, "hasGateEval": true, @@ -421,9 +421,9 @@ }, "make-pdf": { "skill": "make-pdf", - "skillMdBytes": 27161, - "skillMdLines": 593, - "estTokens": 6790, + "skillMdBytes": 18969, + "skillMdLines": 409, + "estTokens": 4742, "tmplBytes": 9306, "descriptionLen": 177, "hasGateEval": false, @@ -431,9 +431,9 @@ }, "office-hours": { "skill": "office-hours", - "skillMdBytes": 126047, - "skillMdLines": 1521, - "estTokens": 31512, + "skillMdBytes": 117855, + "skillMdLines": 1337, + "estTokens": 29464, "tmplBytes": 30566, "descriptionLen": 860, "hasGateEval": true, @@ -441,9 +441,9 @@ }, "open-gstack-browser": { "skill": "open-gstack-browser", - "skillMdBytes": 25141, - "skillMdLines": 560, - "estTokens": 6285, + "skillMdBytes": 16949, + "skillMdLines": 376, + "estTokens": 4237, "tmplBytes": 7719, "descriptionLen": 204, "hasGateEval": false, @@ -451,9 +451,9 @@ }, "pair-agent": { "skill": "pair-agent", - "skillMdBytes": 50973, - "skillMdLines": 963, - "estTokens": 12743, + "skillMdBytes": 42781, + "skillMdLines": 779, + "estTokens": 10695, "tmplBytes": 13368, "descriptionLen": 167, "hasGateEval": false, @@ -461,9 +461,9 @@ }, "plan-ceo-review": { "skill": "plan-ceo-review", - "skillMdBytes": 145538, - "skillMdLines": 1310, - "estTokens": 36385, + "skillMdBytes": 137346, + "skillMdLines": 1126, + "estTokens": 34337, "tmplBytes": 29028, "descriptionLen": 764, "hasGateEval": true, @@ -471,9 +471,9 @@ }, "plan-design-review": { "skill": "plan-design-review", - "skillMdBytes": 118485, - "skillMdLines": 1331, - "estTokens": 29621, + "skillMdBytes": 110293, + "skillMdLines": 1147, + "estTokens": 27573, "tmplBytes": 18463, "descriptionLen": 218, "hasGateEval": true, @@ -481,9 +481,9 @@ }, "plan-devex-review": { "skill": "plan-devex-review", - "skillMdBytes": 119025, - "skillMdLines": 1294, - "estTokens": 29756, + "skillMdBytes": 110833, + "skillMdLines": 1110, + "estTokens": 27708, "tmplBytes": 18598, "descriptionLen": 220, "hasGateEval": true, @@ -491,9 +491,9 @@ }, "plan-eng-review": { "skill": "plan-eng-review", - "skillMdBytes": 119102, - "skillMdLines": 892, - "estTokens": 29776, + "skillMdBytes": 110910, + "skillMdLines": 708, + "estTokens": 27728, "tmplBytes": 13955, "descriptionLen": 201, "hasGateEval": true, @@ -501,9 +501,9 @@ }, "plan-tune": { "skill": "plan-tune", - "skillMdBytes": 63312, - "skillMdLines": 1223, - "estTokens": 15828, + "skillMdBytes": 55120, + "skillMdLines": 1039, + "estTokens": 13780, "tmplBytes": 26922, "descriptionLen": 327, "hasGateEval": true, @@ -511,9 +511,9 @@ }, "qa": { "skill": "qa", - "skillMdBytes": 77798, - "skillMdLines": 1525, - "estTokens": 19450, + "skillMdBytes": 69606, + "skillMdLines": 1341, + "estTokens": 17402, "tmplBytes": 12701, "descriptionLen": 218, "hasGateEval": true, @@ -521,9 +521,9 @@ }, "qa-only": { "skill": "qa-only", - "skillMdBytes": 56678, - "skillMdLines": 1066, - "estTokens": 14170, + "skillMdBytes": 48486, + "skillMdLines": 882, + "estTokens": 12122, "tmplBytes": 3851, "descriptionLen": 165, "hasGateEval": true, @@ -531,9 +531,9 @@ }, "retro": { "skill": "retro", - "skillMdBytes": 83308, - "skillMdLines": 1625, - "estTokens": 20827, + "skillMdBytes": 75116, + "skillMdLines": 1441, + "estTokens": 18779, "tmplBytes": 42589, "descriptionLen": 838, "hasGateEval": true, @@ -541,9 +541,9 @@ }, "review": { "skill": "review", - "skillMdBytes": 100242, - "skillMdLines": 1684, - "estTokens": 25061, + "skillMdBytes": 92050, + "skillMdLines": 1500, + "estTokens": 23013, "tmplBytes": 14141, "descriptionLen": 205, "hasGateEval": true, @@ -551,9 +551,9 @@ }, "scrape": { "skill": "scrape", - "skillMdBytes": 23264, - "skillMdLines": 504, - "estTokens": 5816, + "skillMdBytes": 15072, + "skillMdLines": 320, + "estTokens": 3768, "tmplBytes": 5336, "descriptionLen": 167, "hasGateEval": true, @@ -561,9 +561,9 @@ }, "setup-browser-cookies": { "skill": "setup-browser-cookies", - "skillMdBytes": 20129, - "skillMdLines": 438, - "estTokens": 5032, + "skillMdBytes": 11937, + "skillMdLines": 254, + "estTokens": 2984, "tmplBytes": 2724, "descriptionLen": 222, "hasGateEval": false, @@ -571,9 +571,9 @@ }, "setup-deploy": { "skill": "setup-deploy", - "skillMdBytes": 46420, - "skillMdLines": 805, - "estTokens": 11605, + "skillMdBytes": 38228, + "skillMdLines": 621, + "estTokens": 9557, "tmplBytes": 7805, "descriptionLen": 197, "hasGateEval": true, @@ -581,9 +581,9 @@ }, "setup-gbrain": { "skill": "setup-gbrain", - "skillMdBytes": 84706, - "skillMdLines": 1700, - "estTokens": 21177, + "skillMdBytes": 76514, + "skillMdLines": 1516, + "estTokens": 19129, "tmplBytes": 48298, "descriptionLen": 325, "hasGateEval": true, @@ -591,9 +591,9 @@ }, "ship": { "skill": "ship", - "skillMdBytes": 209656, - "skillMdLines": 1286, - "estTokens": 52414, + "skillMdBytes": 201464, + "skillMdLines": 1102, + "estTokens": 50366, "tmplBytes": 29080, "descriptionLen": 293, "hasGateEval": true, @@ -601,9 +601,9 @@ }, "skillify": { "skill": "skillify", - "skillMdBytes": 53482, - "skillMdLines": 1036, - "estTokens": 13371, + "skillMdBytes": 45290, + "skillMdLines": 852, + "estTokens": 11323, "tmplBytes": 15338, "descriptionLen": 233, "hasGateEval": true, @@ -611,9 +611,9 @@ }, "spec": { "skill": "spec", - "skillMdBytes": 74800, - "skillMdLines": 1428, - "estTokens": 18700, + "skillMdBytes": 66608, + "skillMdLines": 1244, + "estTokens": 16652, "tmplBytes": 32196, "descriptionLen": 282, "hasGateEval": true, @@ -621,9 +621,9 @@ }, "sync-gbrain": { "skill": "sync-gbrain", - "skillMdBytes": 60303, - "skillMdLines": 1064, - "estTokens": 15076, + "skillMdBytes": 52111, + "skillMdLines": 880, + "estTokens": 13028, "tmplBytes": 23886, "descriptionLen": 246, "hasGateEval": false, diff --git a/test/helpers/carve-guards.ts b/test/helpers/carve-guards.ts index f665248c2..ad4d95523 100644 --- a/test/helpers/carve-guards.ts +++ b/test/helpers/carve-guards.ts @@ -126,8 +126,8 @@ export const CARVE_GUARDS: Record = { }, behavioral: 'external', externalTest: 'test/skill-e2e-ship-section-loading.test.ts', - maxSkeletonBytes: 91_600, // v1.68 fix wave: unconditional learnings capture (#2402, ~450B/skill); measured 91,061 - minUnionBytes: 120_000, + maxSkeletonBytes: 71_750, // token-reduction Phases 1-2 (v1.69.x branch): preamble bash -> bin/gstack-skill-start, onboarding -> gated emission; measured 71,145 + minUnionBytes: 181_000, // token-reduction Phases 1-2 (v1.69.x branch): preamble bash -> bin/gstack-skill-start, onboarding -> gated emission; measured union 201,464 mustContain: ['VERSION', 'CHANGELOG', 'review', 'merge', 'PR'], // v1.58.5.0: pre-push-guard install (#2077) stacks on the shared first-run-guidance preamble. // Fork port wave 2: multi-ecosystem test-detection evidence (Django/JVM @@ -157,8 +157,8 @@ export const CARVE_GUARDS: Record = { // v1.65 merge: provisional larger-of-both-waves budget; re-measured below. // Fork port wave 2 (#703): the repo-doc-preference block in the design // check grew every plan-review skeleton ~0.7KB. Measured values noted. - maxSkeletonBytes: 93_900, // v1.68 fix wave: #2402 learnings capture + spool queue-depth lines; measured 93,345 - minUnionBytes: 80_000, + maxSkeletonBytes: 73_980, // token-reduction Phases 1-2 (v1.69.x branch): preamble bash -> bin/gstack-skill-start, onboarding -> gated emission; measured 73,381 + minUnionBytes: 123_600, // token-reduction Phases 1-2 (v1.69.x branch): preamble bash -> bin/gstack-skill-start, onboarding -> gated emission; measured union 137,346 mustContain: ['SCOPE EXPANSION', 'SELECTIVE EXPANSION', 'HOLD SCOPE', 'SCOPE REDUCTION'], // Default-on Codex outside-voice (codexPreflight block + CODEX_MODE branch // prose replacing the smaller opt-in question) lands this ~5.2% over baseline. @@ -183,8 +183,8 @@ export const CARVE_GUARDS: Record = { // check grew every plan-review skeleton ~0.7KB. Measured values noted. // #2499 project-scope MCP jq in the brain-sync block grew every tier-2+ // skeleton ~1.5KB (entry resolution emitted once per SKILL.md). - maxSkeletonBytes: 71_800, // v1.68 fix wave (#2402); measured 71,228 - minUnionBytes: 70_000, + maxSkeletonBytes: 51_860, // token-reduction Phases 1-2 (v1.69.x branch): preamble bash -> bin/gstack-skill-start, onboarding -> gated emission; measured 51,264 + minUnionBytes: 99_800, // token-reduction Phases 1-2 (v1.69.x branch); measured union 110,910 mustContain: ['Architecture', 'Code Quality', 'Test', 'Performance'], // Cross-cutting preamble growth (v1.57.2.0 AUQ-failure prose fallback + the // decision-memory nudge + the v1.57.4.0 Boil-the-Ocean rename) plus the @@ -216,8 +216,8 @@ export const CARVE_GUARDS: Record = { // tier-2+ skeleton (measured 89,184). Main's v1.64.0.0 adds ~340 B more // (telemetry --error-message/--failed-step preamble prose, PR #769). // Budget covers the sum of both waves. - maxSkeletonBytes: 91_700, // v1.68 fix wave (#2402); measured 91,176 - minUnionBytes: 70_000, + maxSkeletonBytes: 71_840, // token-reduction Phases 1-2 (v1.69.x branch): preamble bash -> bin/gstack-skill-start, onboarding -> gated emission; measured 71,242 + minUnionBytes: 99_200, // token-reduction Phases 1-2 (v1.69.x branch); measured union 110,293 mustContain: ['design', 'visual'], maxSizeRatio: 1.12, // D1 1.104 + main's ~0.008 }, @@ -240,8 +240,8 @@ export const CARVE_GUARDS: Record = { // check grew every plan-review skeleton ~0.7KB. Measured values noted. // #2499 project-scope MCP jq in the brain-sync block grew every tier-2+ // skeleton ~1.5KB (entry resolution emitted once per SKILL.md). - maxSkeletonBytes: 83_500, // v1.68 fix wave (#2402); measured 82,941 - minUnionBytes: 70_000, + maxSkeletonBytes: 63_580, // token-reduction Phases 1-2 (v1.69.x branch): preamble bash -> bin/gstack-skill-start, onboarding -> gated emission; measured 62,977 + minUnionBytes: 99_700, // token-reduction Phases 1-2 (v1.69.x branch); measured union 110,833 mustContain: ['developer experience', 'Getting Started'], // Default-on Codex outside-voice (codexPreflight block + CODEX_MODE branch // prose replacing the smaller opt-in question) lands this ~5.7% over baseline. @@ -270,8 +270,8 @@ export const CARVE_GUARDS: Record = { // the #538 opt-out + D1 evidence directive — ratio 1.104 measured. // #2499 project-scope MCP jq in the brain-sync block grew every tier-2+ // skeleton ~1.5KB (entry resolution emitted once per SKILL.md). - maxSkeletonBytes: 102_800, // v1.68 fix wave (#2402); measured 102,220 - minUnionBytes: 70_000, + maxSkeletonBytes: 82_900, // token-reduction Phases 1-2 (v1.69.x branch): preamble bash -> bin/gstack-skill-start, onboarding -> gated emission; measured 82,304 + minUnionBytes: 106_000, // token-reduction Phases 1-2 (v1.69.x branch); measured union 117,855 mustContain: ['design doc', 'problem statement'], maxSizeRatio: 1.12, }, @@ -291,8 +291,8 @@ export const CARVE_GUARDS: Record = { // +Conductor AUQ-default-prose rule + one-way/continuation safety in the // always-loaded AskUserQuestion Format section. // v1.2.0 activation lift: first-run-guidance section in the shared preamble. - maxSkeletonBytes: 57_900, // v1.68 fix wave (#2402); measured 57,385 - minUnionBytes: 55_000, + maxSkeletonBytes: 38_070, // token-reduction Phases 1-2 (v1.69.x branch): preamble bash -> bin/gstack-skill-start, onboarding -> gated emission; measured 37,469 + minUnionBytes: 56_700, // token-reduction Phases 1-2 (v1.69.x branch): preamble bash -> bin/gstack-skill-start, onboarding -> gated emission; measured union 63,018 mustContain: ['CHANGELOG', 'Diataxis', 'coverage'], // Two intentional additions stack on this small skill: the AUQ-failure prose // fallback (v1.57.2.0, ~2KB to every preamble) AND the new default-on Codex @@ -322,8 +322,8 @@ export const CARVE_GUARDS: Record = { // v1.65 merge: provisional larger-of-both-waves budget; re-measured below. // v1.64.1.0: shared-preamble prose from the two parallel v1.64 waves lands // the skeleton at 69,022 B; +~1 KB headroom. - maxSkeletonBytes: 71_400, // v1.68 fix wave (#2402); measured 70,815 - minUnionBytes: 72_000, + maxSkeletonBytes: 51_500, // token-reduction Phases 1-2 (v1.69.x branch): preamble bash -> bin/gstack-skill-start, onboarding -> gated emission; measured 50,899 + minUnionBytes: 65_000, // token-reduction Phases 1-2 (v1.69.x branch): preamble bash -> bin/gstack-skill-start, onboarding -> gated emission; measured union 72,252 mustContain: ['Typography', 'Color', 'Aesthetic Direction'], // Cross-cutting preamble growth (v1.57.2.0 AUQ-failure prose fallback ~2KB + // the cross-session decision-memory nudge) lands this carved skeleton just over @@ -362,8 +362,8 @@ export const CARVE_GUARDS: Record = { // +Conductor AUQ-default-prose rule + one-way/continuation safety in the // always-loaded AskUserQuestion Format section. // v1.2.0 activation lift: first-run-guidance section in the shared preamble. - maxSkeletonBytes: 77_300, // v1.68 fix wave (#2402); measured 76,705 - minUnionBytes: 72_000, + maxSkeletonBytes: 57_390, // token-reduction Phases 1-2 (v1.69.x branch); measured 56,789 + minUnionBytes: 64_200, // token-reduction Phases 1-2 (v1.69.x branch); measured union 71,379 mustContain: ['OWASP', 'STRIDE', 'daily', 'comprehensive', 'verif'], // cso keeps its mode-dispatch + FP-filtering phases always-loaded, so the // cross-cutting preamble growth (v1.57.2.0 AUQ-failure prose fallback ~2KB + the