chore(test): baseline + ratchet follow Phase 3 (OV9); OV8 evaluated — shrink floor stays

Branch-internal baseline recaptured; ratchet ceilings down again. OV8's
floor-retirement question, evaluated as planned after Phase 3: the 80% shrink
floor stays — it uniquely catches accidental body deletion in non-carved
skills BETWEEN ratchet recaptures, and the capture command has amortized the
fixture-refresh cost that motivated retiring it.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This commit is contained in:
Garry Tan
2026-08-25 16:09:08 +00:00
co-authored by Claude Fable 5
parent a3343beef0
commit c7488f7e38
2 changed files with 193 additions and 193 deletions
+40 -40
View File
@@ -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": 22188,
"autoplan": 21874,
"benchmark": 4856,
"benchmark-models": 3829,
"browse": 10212,
"browser-skills/hackernews-frontpage": 371,
"canary": 10235,
"canary": 9921,
"careful": 919,
"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,
"codex": 21188,
"context-restore": 8934,
"context-save": 9551,
"cso": 14524,
"design-consultation": 12985,
"design-html": 15001,
"design-review": 23567,
"design-shotgun": 14031,
"devex-review": 14924,
"diagram": 3847,
"document-generate": 11993,
"document-release": 9790,
"document-generate": 11679,
"document-release": 9476,
"freeze": 990,
"gstack": 3548,
"gstack-upgrade": 3981,
"guard": 889,
"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,
"health": 10133,
"investigate": 10785,
"ios-clean": 8310,
"ios-design-review": 8490,
"ios-fix": 8263,
"ios-qa": 10560,
"ios-sync": 8433,
"land-and-deploy": 23798,
"landing-report": 8844,
"learn": 8514,
"make-pdf": 4957,
"office-hours": 21505,
"office-hours": 21191,
"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": 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,
"pair-agent": 10864,
"plan-ceo-review": 18859,
"plan-design-review": 18300,
"plan-devex-review": 16141,
"plan-eng-review": 13081,
"plan-tune": 14088,
"qa": 17873,
"qa-only": 12355,
"retro": 19313,
"review": 23737,
"scrape": 3939,
"setup-browser-cookies": 3119,
"setup-deploy": 9989,
"setup-gbrain": 19992,
"ship": 18589,
"skillify": 11834,
"spec": 17404,
"sync-gbrain": 13616,
"setup-deploy": 9674,
"setup-gbrain": 19678,
"ship": 18275,
"skillify": 11520,
"spec": 17090,
"sync-gbrain": 13302,
"unfreeze": 393
}
}
+153 -153
View File
@@ -1,17 +1,17 @@
{
"tag": "v1.69.1.0",
"capturedAt": "2026-08-25T15:47:29.737Z",
"capturedFromCommit": "135bf5d4",
"capturedAt": "2026-08-25T16:05:43.563Z",
"capturedFromCommit": "7f780ef5",
"capturedFromBranch": "prompt-token-load-reduction",
"totalSkills": 53,
"totalCorpusBytes": 2841533,
"totalCorpusBytes": 2791573,
"estTotalCatalogTokens": 4195,
"topHeaviest": [
{
"skill": "ship",
"skillMdBytes": 201464,
"skillMdLines": 1102,
"estTokens": 50366,
"skillMdBytes": 200215,
"skillMdLines": 1087,
"estTokens": 50054,
"tmplBytes": 29080,
"descriptionLen": 293,
"hasGateEval": true,
@@ -19,9 +19,9 @@
},
{
"skill": "plan-ceo-review",
"skillMdBytes": 137346,
"skillMdLines": 1126,
"estTokens": 34337,
"skillMdBytes": 136097,
"skillMdLines": 1111,
"estTokens": 34024,
"tmplBytes": 29028,
"descriptionLen": 764,
"hasGateEval": true,
@@ -29,9 +29,9 @@
},
{
"skill": "office-hours",
"skillMdBytes": 117855,
"skillMdLines": 1337,
"estTokens": 29464,
"skillMdBytes": 116606,
"skillMdLines": 1322,
"estTokens": 29152,
"tmplBytes": 30566,
"descriptionLen": 860,
"hasGateEval": true,
@@ -39,9 +39,9 @@
},
{
"skill": "plan-eng-review",
"skillMdBytes": 110910,
"skillMdLines": 708,
"estTokens": 27728,
"skillMdBytes": 109661,
"skillMdLines": 693,
"estTokens": 27415,
"tmplBytes": 13955,
"descriptionLen": 201,
"hasGateEval": true,
@@ -49,9 +49,9 @@
},
{
"skill": "plan-devex-review",
"skillMdBytes": 110833,
"skillMdLines": 1110,
"estTokens": 27708,
"skillMdBytes": 109584,
"skillMdLines": 1095,
"estTokens": 27396,
"tmplBytes": 18598,
"descriptionLen": 220,
"hasGateEval": true,
@@ -59,9 +59,9 @@
},
{
"skill": "plan-design-review",
"skillMdBytes": 110293,
"skillMdLines": 1147,
"estTokens": 27573,
"skillMdBytes": 109044,
"skillMdLines": 1132,
"estTokens": 27261,
"tmplBytes": 18463,
"descriptionLen": 218,
"hasGateEval": true,
@@ -69,9 +69,9 @@
},
{
"skill": "land-and-deploy",
"skillMdBytes": 92284,
"skillMdLines": 1634,
"estTokens": 23071,
"skillMdBytes": 91035,
"skillMdLines": 1619,
"estTokens": 22759,
"tmplBytes": 54352,
"descriptionLen": 160,
"hasGateEval": true,
@@ -79,9 +79,9 @@
},
{
"skill": "review",
"skillMdBytes": 92050,
"skillMdLines": 1500,
"estTokens": 23013,
"skillMdBytes": 90801,
"skillMdLines": 1485,
"estTokens": 22700,
"tmplBytes": 14141,
"descriptionLen": 205,
"hasGateEval": true,
@@ -89,9 +89,9 @@
},
{
"skill": "design-review",
"skillMdBytes": 91397,
"skillMdLines": 1651,
"estTokens": 22849,
"skillMdBytes": 90148,
"skillMdLines": 1636,
"estTokens": 22537,
"tmplBytes": 11674,
"descriptionLen": 306,
"hasGateEval": true,
@@ -99,9 +99,9 @@
},
{
"skill": "autoplan",
"skillMdBytes": 84917,
"skillMdLines": 1502,
"estTokens": 21229,
"skillMdBytes": 83668,
"skillMdLines": 1487,
"estTokens": 20917,
"tmplBytes": 46355,
"descriptionLen": 336,
"hasGateEval": true,
@@ -111,9 +111,9 @@
"skills": {
"autoplan": {
"skill": "autoplan",
"skillMdBytes": 84917,
"skillMdLines": 1502,
"estTokens": 21229,
"skillMdBytes": 83668,
"skillMdLines": 1487,
"estTokens": 20917,
"tmplBytes": 46355,
"descriptionLen": 336,
"hasGateEval": true,
@@ -151,9 +151,9 @@
},
"canary": {
"skill": "canary",
"skillMdBytes": 39170,
"skillMdLines": 678,
"estTokens": 9793,
"skillMdBytes": 37921,
"skillMdLines": 663,
"estTokens": 9480,
"tmplBytes": 8033,
"descriptionLen": 180,
"hasGateEval": true,
@@ -171,9 +171,9 @@
},
"codex": {
"skill": "codex",
"skillMdBytes": 82293,
"skillMdLines": 1371,
"estTokens": 20573,
"skillMdBytes": 81044,
"skillMdLines": 1356,
"estTokens": 20261,
"tmplBytes": 43658,
"descriptionLen": 187,
"hasGateEval": true,
@@ -181,9 +181,9 @@
},
"context-restore": {
"skill": "context-restore",
"skillMdBytes": 35395,
"skillMdLines": 569,
"estTokens": 8849,
"skillMdBytes": 34146,
"skillMdLines": 554,
"estTokens": 8537,
"tmplBytes": 7092,
"descriptionLen": 238,
"hasGateEval": true,
@@ -191,9 +191,9 @@
},
"context-save": {
"skill": "context-save",
"skillMdBytes": 37755,
"skillMdLines": 654,
"estTokens": 9439,
"skillMdBytes": 36506,
"skillMdLines": 639,
"estTokens": 9127,
"tmplBytes": 9293,
"descriptionLen": 168,
"hasGateEval": true,
@@ -201,9 +201,9 @@
},
"cso": {
"skill": "cso",
"skillMdBytes": 71379,
"skillMdLines": 911,
"estTokens": 17845,
"skillMdBytes": 70130,
"skillMdLines": 896,
"estTokens": 17533,
"tmplBytes": 21724,
"descriptionLen": 196,
"hasGateEval": true,
@@ -211,9 +211,9 @@
},
"design-consultation": {
"skill": "design-consultation",
"skillMdBytes": 72252,
"skillMdLines": 856,
"estTokens": 18063,
"skillMdBytes": 71003,
"skillMdLines": 841,
"estTokens": 17751,
"tmplBytes": 9554,
"descriptionLen": 890,
"hasGateEval": true,
@@ -221,9 +221,9 @@
},
"design-html": {
"skill": "design-html",
"skillMdBytes": 58614,
"skillMdLines": 1137,
"estTokens": 14654,
"skillMdBytes": 57365,
"skillMdLines": 1122,
"estTokens": 14341,
"tmplBytes": 22567,
"descriptionLen": 235,
"hasGateEval": true,
@@ -231,9 +231,9 @@
},
"design-review": {
"skill": "design-review",
"skillMdBytes": 91397,
"skillMdLines": 1651,
"estTokens": 22849,
"skillMdBytes": 90148,
"skillMdLines": 1636,
"estTokens": 22537,
"tmplBytes": 11674,
"descriptionLen": 306,
"hasGateEval": true,
@@ -241,9 +241,9 @@
},
"design-shotgun": {
"skill": "design-shotgun",
"skillMdBytes": 54903,
"skillMdLines": 999,
"estTokens": 13726,
"skillMdBytes": 53654,
"skillMdLines": 984,
"estTokens": 13414,
"tmplBytes": 13331,
"descriptionLen": 788,
"hasGateEval": true,
@@ -251,9 +251,9 @@
},
"devex-review": {
"skill": "devex-review",
"skillMdBytes": 58321,
"skillMdLines": 932,
"estTokens": 14580,
"skillMdBytes": 57072,
"skillMdLines": 917,
"estTokens": 14268,
"tmplBytes": 7984,
"descriptionLen": 201,
"hasGateEval": false,
@@ -271,9 +271,9 @@
},
"document-generate": {
"skill": "document-generate",
"skillMdBytes": 45899,
"skillMdLines": 878,
"estTokens": 11475,
"skillMdBytes": 44650,
"skillMdLines": 863,
"estTokens": 11163,
"tmplBytes": 15940,
"descriptionLen": 334,
"hasGateEval": false,
@@ -281,9 +281,9 @@
},
"document-release": {
"skill": "document-release",
"skillMdBytes": 63018,
"skillMdLines": 587,
"estTokens": 15755,
"skillMdBytes": 61769,
"skillMdLines": 572,
"estTokens": 15442,
"tmplBytes": 6688,
"descriptionLen": 192,
"hasGateEval": true,
@@ -321,9 +321,9 @@
},
"health": {
"skill": "health",
"skillMdBytes": 39981,
"skillMdLines": 702,
"estTokens": 9995,
"skillMdBytes": 38732,
"skillMdLines": 687,
"estTokens": 9683,
"tmplBytes": 11617,
"descriptionLen": 184,
"hasGateEval": true,
@@ -331,9 +331,9 @@
},
"investigate": {
"skill": "investigate",
"skillMdBytes": 42479,
"skillMdLines": 702,
"estTokens": 10620,
"skillMdBytes": 41230,
"skillMdLines": 687,
"estTokens": 10308,
"tmplBytes": 11566,
"descriptionLen": 1241,
"hasGateEval": true,
@@ -341,9 +341,9 @@
},
"ios-clean": {
"skill": "ios-clean",
"skillMdBytes": 33004,
"skillMdLines": 500,
"estTokens": 8251,
"skillMdBytes": 31755,
"skillMdLines": 485,
"estTokens": 7939,
"tmplBytes": 3743,
"descriptionLen": 254,
"hasGateEval": false,
@@ -351,9 +351,9 @@
},
"ios-design-review": {
"skill": "ios-design-review",
"skillMdBytes": 33696,
"skillMdLines": 503,
"estTokens": 8424,
"skillMdBytes": 32447,
"skillMdLines": 488,
"estTokens": 8112,
"tmplBytes": 4417,
"descriptionLen": 209,
"hasGateEval": false,
@@ -361,9 +361,9 @@
},
"ios-fix": {
"skill": "ios-fix",
"skillMdBytes": 32825,
"skillMdLines": 499,
"estTokens": 8206,
"skillMdBytes": 31576,
"skillMdLines": 484,
"estTokens": 7894,
"tmplBytes": 3574,
"descriptionLen": 187,
"hasGateEval": false,
@@ -371,9 +371,9 @@
},
"ios-qa": {
"skill": "ios-qa",
"skillMdBytes": 41616,
"skillMdLines": 656,
"estTokens": 10404,
"skillMdBytes": 40367,
"skillMdLines": 641,
"estTokens": 10092,
"tmplBytes": 12370,
"descriptionLen": 223,
"hasGateEval": true,
@@ -381,9 +381,9 @@
},
"ios-sync": {
"skill": "ios-sync",
"skillMdBytes": 33478,
"skillMdLines": 497,
"estTokens": 8370,
"skillMdBytes": 32229,
"skillMdLines": 482,
"estTokens": 8057,
"tmplBytes": 4220,
"descriptionLen": 269,
"hasGateEval": true,
@@ -391,9 +391,9 @@
},
"land-and-deploy": {
"skill": "land-and-deploy",
"skillMdBytes": 92284,
"skillMdLines": 1634,
"estTokens": 23071,
"skillMdBytes": 91035,
"skillMdLines": 1619,
"estTokens": 22759,
"tmplBytes": 54352,
"descriptionLen": 160,
"hasGateEval": true,
@@ -401,9 +401,9 @@
},
"landing-report": {
"skill": "landing-report",
"skillMdBytes": 35050,
"skillMdLines": 545,
"estTokens": 8763,
"skillMdBytes": 33801,
"skillMdLines": 530,
"estTokens": 8450,
"tmplBytes": 6847,
"descriptionLen": 195,
"hasGateEval": false,
@@ -411,9 +411,9 @@
},
"learn": {
"skill": "learn",
"skillMdBytes": 33787,
"skillMdLines": 579,
"estTokens": 8447,
"skillMdBytes": 32538,
"skillMdLines": 564,
"estTokens": 8135,
"tmplBytes": 5594,
"descriptionLen": 178,
"hasGateEval": true,
@@ -431,9 +431,9 @@
},
"office-hours": {
"skill": "office-hours",
"skillMdBytes": 117855,
"skillMdLines": 1337,
"estTokens": 29464,
"skillMdBytes": 116606,
"skillMdLines": 1322,
"estTokens": 29152,
"tmplBytes": 30566,
"descriptionLen": 860,
"hasGateEval": true,
@@ -451,9 +451,9 @@
},
"pair-agent": {
"skill": "pair-agent",
"skillMdBytes": 42781,
"skillMdLines": 779,
"estTokens": 10695,
"skillMdBytes": 41532,
"skillMdLines": 764,
"estTokens": 10383,
"tmplBytes": 13368,
"descriptionLen": 167,
"hasGateEval": false,
@@ -461,9 +461,9 @@
},
"plan-ceo-review": {
"skill": "plan-ceo-review",
"skillMdBytes": 137346,
"skillMdLines": 1126,
"estTokens": 34337,
"skillMdBytes": 136097,
"skillMdLines": 1111,
"estTokens": 34024,
"tmplBytes": 29028,
"descriptionLen": 764,
"hasGateEval": true,
@@ -471,9 +471,9 @@
},
"plan-design-review": {
"skill": "plan-design-review",
"skillMdBytes": 110293,
"skillMdLines": 1147,
"estTokens": 27573,
"skillMdBytes": 109044,
"skillMdLines": 1132,
"estTokens": 27261,
"tmplBytes": 18463,
"descriptionLen": 218,
"hasGateEval": true,
@@ -481,9 +481,9 @@
},
"plan-devex-review": {
"skill": "plan-devex-review",
"skillMdBytes": 110833,
"skillMdLines": 1110,
"estTokens": 27708,
"skillMdBytes": 109584,
"skillMdLines": 1095,
"estTokens": 27396,
"tmplBytes": 18598,
"descriptionLen": 220,
"hasGateEval": true,
@@ -491,9 +491,9 @@
},
"plan-eng-review": {
"skill": "plan-eng-review",
"skillMdBytes": 110910,
"skillMdLines": 708,
"estTokens": 27728,
"skillMdBytes": 109661,
"skillMdLines": 693,
"estTokens": 27415,
"tmplBytes": 13955,
"descriptionLen": 201,
"hasGateEval": true,
@@ -501,9 +501,9 @@
},
"plan-tune": {
"skill": "plan-tune",
"skillMdBytes": 55120,
"skillMdLines": 1039,
"estTokens": 13780,
"skillMdBytes": 53871,
"skillMdLines": 1024,
"estTokens": 13468,
"tmplBytes": 26922,
"descriptionLen": 327,
"hasGateEval": true,
@@ -511,9 +511,9 @@
},
"qa": {
"skill": "qa",
"skillMdBytes": 69606,
"skillMdLines": 1341,
"estTokens": 17402,
"skillMdBytes": 68357,
"skillMdLines": 1326,
"estTokens": 17089,
"tmplBytes": 12701,
"descriptionLen": 218,
"hasGateEval": true,
@@ -521,9 +521,9 @@
},
"qa-only": {
"skill": "qa-only",
"skillMdBytes": 48486,
"skillMdLines": 882,
"estTokens": 12122,
"skillMdBytes": 47237,
"skillMdLines": 867,
"estTokens": 11809,
"tmplBytes": 3851,
"descriptionLen": 165,
"hasGateEval": true,
@@ -531,9 +531,9 @@
},
"retro": {
"skill": "retro",
"skillMdBytes": 75116,
"skillMdLines": 1441,
"estTokens": 18779,
"skillMdBytes": 73867,
"skillMdLines": 1426,
"estTokens": 18467,
"tmplBytes": 42589,
"descriptionLen": 838,
"hasGateEval": true,
@@ -541,9 +541,9 @@
},
"review": {
"skill": "review",
"skillMdBytes": 92050,
"skillMdLines": 1500,
"estTokens": 23013,
"skillMdBytes": 90801,
"skillMdLines": 1485,
"estTokens": 22700,
"tmplBytes": 14141,
"descriptionLen": 205,
"hasGateEval": true,
@@ -571,9 +571,9 @@
},
"setup-deploy": {
"skill": "setup-deploy",
"skillMdBytes": 38228,
"skillMdLines": 621,
"estTokens": 9557,
"skillMdBytes": 36979,
"skillMdLines": 606,
"estTokens": 9245,
"tmplBytes": 7805,
"descriptionLen": 197,
"hasGateEval": true,
@@ -581,9 +581,9 @@
},
"setup-gbrain": {
"skill": "setup-gbrain",
"skillMdBytes": 76514,
"skillMdLines": 1516,
"estTokens": 19129,
"skillMdBytes": 75265,
"skillMdLines": 1501,
"estTokens": 18816,
"tmplBytes": 48298,
"descriptionLen": 325,
"hasGateEval": true,
@@ -591,9 +591,9 @@
},
"ship": {
"skill": "ship",
"skillMdBytes": 201464,
"skillMdLines": 1102,
"estTokens": 50366,
"skillMdBytes": 200215,
"skillMdLines": 1087,
"estTokens": 50054,
"tmplBytes": 29080,
"descriptionLen": 293,
"hasGateEval": true,
@@ -601,9 +601,9 @@
},
"skillify": {
"skill": "skillify",
"skillMdBytes": 45290,
"skillMdLines": 852,
"estTokens": 11323,
"skillMdBytes": 44041,
"skillMdLines": 837,
"estTokens": 11010,
"tmplBytes": 15338,
"descriptionLen": 233,
"hasGateEval": true,
@@ -611,9 +611,9 @@
},
"spec": {
"skill": "spec",
"skillMdBytes": 66608,
"skillMdLines": 1244,
"estTokens": 16652,
"skillMdBytes": 65359,
"skillMdLines": 1229,
"estTokens": 16340,
"tmplBytes": 32196,
"descriptionLen": 282,
"hasGateEval": true,
@@ -621,9 +621,9 @@
},
"sync-gbrain": {
"skill": "sync-gbrain",
"skillMdBytes": 52111,
"skillMdLines": 880,
"estTokens": 13028,
"skillMdBytes": 50862,
"skillMdLines": 865,
"estTokens": 12716,
"tmplBytes": 23886,
"descriptionLen": 246,
"hasGateEval": false,