@cleocode/skills 2026.5.82 → 2026.5.83

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (31) hide show
  1. package/README.md +0 -1
  2. package/package.json +1 -1
  3. package/profiles/recommended.json +1 -1
  4. package/skills/manifest.json +45 -1
  5. package/skills/ct-grade-v2-1/MIGRATION.md +0 -28
  6. package/skills/ct-grade-v2-1/SKILL.md +0 -235
  7. package/skills/ct-grade-v2-1/agents/analysis-reporter.md +0 -203
  8. package/skills/ct-grade-v2-1/agents/blind-comparator.md +0 -157
  9. package/skills/ct-grade-v2-1/agents/scenario-runner.md +0 -160
  10. package/skills/ct-grade-v2-1/evals/evals.json +0 -74
  11. package/skills/ct-grade-v2-1/grade-viewer/__pycache__/build_op_stats.cpython-314.pyc +0 -0
  12. package/skills/ct-grade-v2-1/grade-viewer/__pycache__/generate_grade_review.cpython-314.pyc +0 -0
  13. package/skills/ct-grade-v2-1/grade-viewer/build_op_stats.py +0 -174
  14. package/skills/ct-grade-v2-1/grade-viewer/eval-analysis.json +0 -41
  15. package/skills/ct-grade-v2-1/grade-viewer/eval-report.md +0 -37
  16. package/skills/ct-grade-v2-1/grade-viewer/generate_grade_review.py +0 -1023
  17. package/skills/ct-grade-v2-1/grade-viewer/generate_grade_viewer.py +0 -548
  18. package/skills/ct-grade-v2-1/grade-viewer/grade-review-eval.html +0 -613
  19. package/skills/ct-grade-v2-1/grade-viewer/grade-review.html +0 -1532
  20. package/skills/ct-grade-v2-1/grade-viewer/viewer.html +0 -620
  21. package/skills/ct-grade-v2-1/manifest-entry.json +0 -31
  22. package/skills/ct-grade-v2-1/references/ab-testing.md +0 -173
  23. package/skills/ct-grade-v2-1/references/domains-ssot.md +0 -156
  24. package/skills/ct-grade-v2-1/references/grade-spec-v2.md +0 -167
  25. package/skills/ct-grade-v2-1/references/playbook-v2.md +0 -325
  26. package/skills/ct-grade-v2-1/references/token-tracking.md +0 -200
  27. package/skills/ct-grade-v2-1/scripts/generate_report.py +0 -419
  28. package/skills/ct-grade-v2-1/scripts/run_ab_test.py +0 -493
  29. package/skills/ct-grade-v2-1/scripts/run_scenario.py +0 -396
  30. package/skills/ct-grade-v2-1/scripts/setup_run.py +0 -207
  31. package/skills/ct-grade-v2-1/scripts/token_tracker.py +0 -175
@@ -1,620 +0,0 @@
1
- <!DOCTYPE html>
2
- <html lang="en">
3
- <head>
4
- <meta charset="UTF-8">
5
- <meta name="viewport" content="width=device-width, initial-scale=1.0">
6
- <title>CLEO Grade Review</title>
7
- <link rel="preconnect" href="https://fonts.googleapis.com">
8
- <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
9
- <link href="https://fonts.googleapis.com/css2?family=Poppins:wght@500;600;700&family=Lora:wght@400;500&family=JetBrains+Mono:wght@400;500&display=swap" rel="stylesheet">
10
- <style>
11
- :root {
12
- --bg: #faf9f5;
13
- --surface: #ffffff;
14
- --surface2: #f4f2ec;
15
- --border: #e8e6dc;
16
- --text: #141413;
17
- --text-muted: #8a8880;
18
- --accent: #d97757;
19
- --header-bg: #141413;
20
- --header-text:#faf9f5;
21
- --radius: 6px;
22
- --mono: 'JetBrains Mono', 'Fira Code', monospace;
23
- --grade-a: #4a8c5c; --grade-a-bg: #eaf4ee;
24
- --grade-b: #3a7fa8; --grade-b-bg: #e8f2f9;
25
- --grade-c: #b07d2a; --grade-c-bg: #fdf4e3;
26
- --grade-d: #c45d30; --grade-d-bg: #fceee8;
27
- --grade-f: #c03030; --grade-f-bg: #fce8e8;
28
- --dim1: #7b68c8; --dim2: #4a8fa5; --dim3: #4a8c5c;
29
- --dim4: #c4923a; --dim5: #d97757;
30
- --mcp: #4a8fa5; --cli: #7b68c8; --tie: #8a8880;
31
- --win: #4a8c5c; --loss: #c03030;
32
- }
33
- * { box-sizing: border-box; margin: 0; padding: 0; }
34
- body { font-family: 'Lora', Georgia, serif; background: var(--bg); color: var(--text); height: 100vh; display: flex; flex-direction: column; font-size: 14px; }
35
- .header { background: var(--header-bg); color: var(--header-text); padding: .75rem 1.5rem; display: flex; justify-content: space-between; align-items: center; flex-shrink: 0; gap: 1rem; }
36
- .header-left { display: flex; align-items: center; gap: .75rem; }
37
- .header-logo { font-family: 'Poppins', sans-serif; font-size: .65rem; font-weight: 700; letter-spacing: .12em; text-transform: uppercase; background: var(--accent); color: #fff; padding: .2em .5em; border-radius: 3px; }
38
- .header-title { font-family: 'Poppins', sans-serif; font-size: 1rem; font-weight: 600; color: #fff; }
39
- .header-subtitle { font-size: .75rem; color: #888; }
40
- .header-meta { font-size: .7rem; color: #666; text-align: right; line-height: 1.6; }
41
- .tabs-bar { background: var(--surface); border-bottom: 1px solid var(--border); display: flex; padding: 0 1.5rem; gap: .25rem; flex-shrink: 0; }
42
- .tab-btn { font-family: 'Poppins', sans-serif; font-size: .75rem; font-weight: 500; padding: .6rem .9rem; border: none; background: transparent; color: var(--text-muted); cursor: pointer; border-bottom: 2px solid transparent; transition: color .15s, border-color .15s; letter-spacing: .02em; }
43
- .tab-btn:hover { color: var(--text); }
44
- .tab-btn.active { color: var(--accent); border-bottom-color: var(--accent); }
45
- .tab-badge { display: inline-block; background: var(--surface2); border-radius: 9px; font-size: .65rem; padding: .1em .45em; margin-left: .3em; color: var(--text-muted); }
46
- .tab-btn.active .tab-badge { background: #fde8df; color: var(--accent); }
47
- .main { display: flex; flex: 1; overflow: hidden; }
48
- .tab-panel { display: none; width: 100%; overflow: hidden; }
49
- .tab-panel.active { display: flex; }
50
- .sidebar { width: 240px; min-width: 180px; border-right: 1px solid var(--border); background: var(--surface); display: flex; flex-direction: column; flex-shrink: 0; overflow-y: auto; }
51
- .sidebar-header { font-family: 'Poppins', sans-serif; font-size: .65rem; font-weight: 600; letter-spacing: .08em; text-transform: uppercase; color: var(--text-muted); padding: .75rem 1rem .4rem; }
52
- .sidebar-item { padding: .55rem 1rem; cursor: pointer; display: flex; align-items: center; gap: .5rem; border-left: 3px solid transparent; transition: background .1s; font-size: .8rem; }
53
- .sidebar-item:hover { background: var(--bg); }
54
- .sidebar-item.active { background: var(--bg); border-left-color: var(--accent); }
55
- .item-label { flex: 1; min-width: 0; overflow: hidden; text-overflow: ellipsis; white-space: nowrap; }
56
- .item-score { font-family: var(--mono); font-size: .7rem; font-weight: 500; }
57
- .content { flex: 1; overflow-y: auto; padding: 1.5rem; display: flex; flex-direction: column; gap: 1.25rem; }
58
- .grade-card { background: var(--surface); border: 1px solid var(--border); border-radius: var(--radius); padding: 1.25rem; }
59
- .grade-card-header { display: flex; align-items: flex-start; justify-content: space-between; gap: 1rem; margin-bottom: 1rem; }
60
- .grade-score-block { display: flex; align-items: center; gap: 1rem; }
61
- .grade-letter-badge { font-family: 'Poppins', sans-serif; font-size: 2.5rem; font-weight: 700; width: 64px; height: 64px; display: flex; align-items: center; justify-content: center; border-radius: 8px; }
62
- .grade-letter-badge.A { background: var(--grade-a-bg); color: var(--grade-a); }
63
- .grade-letter-badge.B { background: var(--grade-b-bg); color: var(--grade-b); }
64
- .grade-letter-badge.C { background: var(--grade-c-bg); color: var(--grade-c); }
65
- .grade-letter-badge.D { background: var(--grade-d-bg); color: var(--grade-d); }
66
- .grade-letter-badge.F { background: var(--grade-f-bg); color: var(--grade-f); }
67
- .grade-score-details h2 { font-family: 'Poppins', sans-serif; font-size: 1.5rem; font-weight: 700; line-height: 1; }
68
- .grade-score-details .pct { font-size: .85rem; color: var(--text-muted); }
69
- .grade-session-id { font-family: var(--mono); font-size: .65rem; color: var(--text-muted); background: var(--surface2); padding: .2em .5em; border-radius: 3px; }
70
- .dimensions { display: flex; flex-direction: column; gap: .65rem; }
71
- .dimension-row { display: flex; align-items: center; gap: .75rem; }
72
- .dimension-name { font-family: 'Poppins', sans-serif; font-size: .7rem; font-weight: 500; width: 145px; flex-shrink: 0; }
73
- .dimension-bar-wrap { flex: 1; height: 8px; background: var(--surface2); border-radius: 4px; overflow: hidden; }
74
- .dimension-bar { height: 100%; border-radius: 4px; transition: width .4s ease; }
75
- .dimension-score-label { font-family: var(--mono); font-size: .7rem; width: 42px; text-align: right; flex-shrink: 0; }
76
- .flags-section h3 { font-family: 'Poppins', sans-serif; font-size: .75rem; font-weight: 600; color: var(--text-muted); text-transform: uppercase; letter-spacing: .07em; margin-bottom: .6rem; }
77
- .flags-list { display: flex; flex-direction: column; gap: .4rem; }
78
- .flag-item { display: flex; align-items: flex-start; gap: .5rem; font-size: .78rem; padding: .4rem .6rem; background: var(--grade-f-bg); border-left: 3px solid var(--grade-f); border-radius: 0 var(--radius) var(--radius) 0; line-height: 1.4; }
79
- .flag-icon { flex-shrink: 0; }
80
- .no-flags { font-size: .78rem; color: var(--win); background: var(--grade-a-bg); padding: .4rem .6rem; border-left: 3px solid var(--win); border-radius: 0 var(--radius) var(--radius) 0; }
81
- .token-meta-grid { display: grid; grid-template-columns: repeat(auto-fill, minmax(130px, 1fr)); gap: .6rem; }
82
- .token-stat { background: var(--surface2); border-radius: var(--radius); padding: .6rem .75rem; }
83
- .token-stat-label { font-family: 'Poppins', sans-serif; font-size: .6rem; font-weight: 600; text-transform: uppercase; letter-spacing: .07em; color: var(--text-muted); margin-bottom: .2rem; }
84
- .token-stat-value { font-family: var(--mono); font-size: .95rem; font-weight: 500; }
85
- .ab-panel, .token-panel, .history-panel { width: 100%; padding: 1.5rem; overflow-y: auto; display: flex; flex-direction: column; gap: 1.25rem; }
86
- .section-title { font-family: 'Poppins', sans-serif; font-size: .75rem; font-weight: 600; text-transform: uppercase; letter-spacing: .08em; color: var(--text-muted); margin-bottom: .75rem; }
87
- .ab-summary-grid { display: grid; grid-template-columns: repeat(auto-fill, minmax(150px, 1fr)); gap: .75rem; }
88
- .ab-stat-card { background: var(--surface); border: 1px solid var(--border); border-radius: var(--radius); padding: .85rem 1rem; }
89
- .ab-stat-label { font-family: 'Poppins', sans-serif; font-size: .65rem; font-weight: 600; text-transform: uppercase; letter-spacing: .07em; color: var(--text-muted); margin-bottom: .3rem; }
90
- .ab-stat-value { font-family: var(--mono); font-size: 1.3rem; font-weight: 500; }
91
- .ab-stat-value.win { color: var(--win); } .ab-stat-value.loss { color: var(--loss); } .ab-stat-value.neutral { color: var(--text-muted); }
92
- .ab-winner-banner { display: flex; align-items: center; gap: .6rem; padding: .6rem 1rem; border-radius: var(--radius); font-family: 'Poppins', sans-serif; font-size: .8rem; font-weight: 600; }
93
- .ab-winner-banner.mcp { background: #e8f2f9; color: var(--mcp); }
94
- .ab-winner-banner.cli { background: #ede8f9; color: var(--cli); }
95
- .ab-winner-banner.tie { background: var(--surface2); color: var(--text-muted); }
96
- .ab-table-wrap { overflow-x: auto; }
97
- table.ab-table { width: 100%; border-collapse: collapse; font-size: .76rem; }
98
- .ab-table th { font-family: 'Poppins', sans-serif; font-size: .65rem; font-weight: 600; text-transform: uppercase; letter-spacing: .07em; color: var(--text-muted); padding: .4rem .75rem; text-align: left; background: var(--surface); border-bottom: 1px solid var(--border); white-space: nowrap; }
99
- .ab-table td { padding: .45rem .75rem; border-bottom: 1px solid var(--border); vertical-align: middle; }
100
- .ab-table tr:hover td { background: var(--bg); }
101
- .op-name { font-family: var(--mono); font-size: .72rem; }
102
- .win-pill { display: inline-block; padding: .15em .5em; border-radius: 9px; font-family: 'Poppins', sans-serif; font-size: .65rem; font-weight: 600; text-transform: uppercase; }
103
- .win-pill.mcp { background: #d0e8f4; color: var(--mcp); } .win-pill.cli { background: #ddd8f4; color: var(--cli); } .win-pill.tie { background: var(--surface2); color: var(--text-muted); }
104
- .delta-pos { color: var(--loss); font-family: var(--mono); font-size: .72rem; }
105
- .delta-neg { color: var(--win); font-family: var(--mono); font-size: .72rem; }
106
- .delta-zero { color: var(--text-muted); font-family: var(--mono); font-size: .72rem; }
107
- .bar-chart { display: flex; flex-direction: column; gap: .5rem; }
108
- .bar-row { display: flex; align-items: center; gap: .75rem; }
109
- .bar-label { font-family: var(--mono); font-size: .72rem; width: 130px; flex-shrink: 0; text-align: right; }
110
- .bar-track { flex: 1; height: 16px; background: var(--surface2); border-radius: 3px; overflow: hidden; }
111
- .bar-fill { height: 100%; border-radius: 3px; transition: width .4s ease; }
112
- .bar-val { font-family: var(--mono); font-size: .7rem; width: 70px; flex-shrink: 0; }
113
- .chart-card { background: var(--surface); border: 1px solid var(--border); border-radius: var(--radius); padding: 1rem 1.25rem; }
114
- .chart-card-title { font-family: 'Poppins', sans-serif; font-size: .8rem; font-weight: 600; margin-bottom: 1rem; }
115
- .history-list { display: flex; flex-direction: column; gap: .5rem; }
116
- .history-item { background: var(--surface); border: 1px solid var(--border); border-radius: var(--radius); padding: .75rem 1rem; display: flex; align-items: center; gap: 1rem; cursor: pointer; transition: border-color .15s; }
117
- .history-item:hover { border-color: var(--accent); }
118
- .history-grade-badge { font-family: 'Poppins', sans-serif; font-size: 1.1rem; font-weight: 700; width: 36px; height: 36px; display: flex; align-items: center; justify-content: center; border-radius: 5px; flex-shrink: 0; }
119
- .history-grade-badge.A { background: var(--grade-a-bg); color: var(--grade-a); } .history-grade-badge.B { background: var(--grade-b-bg); color: var(--grade-b); }
120
- .history-grade-badge.C { background: var(--grade-c-bg); color: var(--grade-c); } .history-grade-badge.D { background: var(--grade-d-bg); color: var(--grade-d); }
121
- .history-grade-badge.F { background: var(--grade-f-bg); color: var(--grade-f); }
122
- .history-item-main { flex: 1; min-width: 0; }
123
- .history-session-id { font-family: var(--mono); font-size: .7rem; color: var(--text-muted); overflow: hidden; text-overflow: ellipsis; white-space: nowrap; }
124
- .history-item-meta { display: flex; gap: .6rem; font-size: .7rem; color: var(--text-muted); margin-top: .2rem; flex-wrap: wrap; }
125
- .history-score-bar { width: 120px; height: 5px; background: var(--surface2); border-radius: 3px; overflow: hidden; flex-shrink: 0; }
126
- .history-score-fill { height: 100%; border-radius: 3px; }
127
- canvas.trend { width: 100%; height: 90px; display: block; }
128
- .empty-state { display: flex; flex-direction: column; align-items: center; justify-content: center; gap: .75rem; padding: 3rem 2rem; color: var(--text-muted); text-align: center; }
129
- .empty-state .empty-icon { font-size: 2.5rem; }
130
- .empty-state h3 { font-family: 'Poppins', sans-serif; font-size: .9rem; color: var(--text); }
131
- .empty-state p { font-size: .78rem; max-width: 380px; line-height: 1.6; }
132
- .empty-cmd { font-family: var(--mono); background: var(--surface2); padding: .25em .5em; border-radius: 3px; font-size: .74rem; display: inline-block; margin-top: .25rem; }
133
- .ev-title { font-family: 'Poppins', sans-serif; font-size: .75rem; font-weight: 600; text-transform: uppercase; letter-spacing: .07em; color: var(--text-muted); margin-bottom: .75rem; }
134
- .ev-row { margin-bottom: .5rem; }
135
- .ev-dim { font-family: 'Poppins', sans-serif; font-size: .68rem; font-weight: 600; margin-bottom: .2rem; }
136
- .ev-bullet { font-size: .74rem; padding: .15rem 0 .15rem .75rem; color: var(--text-muted); }
137
- .ab-note { font-size: .72rem; color: var(--text-muted); background: var(--grade-c-bg); border-left: 3px solid var(--grade-c); padding: .5rem .75rem; border-radius: 0 4px 4px 0; }
138
- ::-webkit-scrollbar { width: 6px; height: 6px; }
139
- ::-webkit-scrollbar-track { background: transparent; }
140
- ::-webkit-scrollbar-thumb { background: var(--border); border-radius: 3px; }
141
- ::-webkit-scrollbar-thumb:hover { background: var(--text-muted); }
142
- </style>
143
- </head>
144
- <body>
145
-
146
- <div class="header">
147
- <div class="header-left">
148
- <span class="header-logo">CLEO</span>
149
- <div>
150
- <div class="header-title" id="hdr-title">Grade Review</div>
151
- <div class="header-subtitle" id="hdr-sub"></div>
152
- </div>
153
- </div>
154
- <div class="header-meta" id="hdr-meta"></div>
155
- </div>
156
-
157
- <div class="tabs-bar">
158
- <button class="tab-btn active" data-tab="grades">Grades <span class="tab-badge" id="tb-grades">0</span></button>
159
- <button class="tab-btn" data-tab="ab">A/B Comparison <span class="tab-badge" id="tb-ab">0</span></button>
160
- <button class="tab-btn" data-tab="tokens">Token Analysis</button>
161
- <button class="tab-btn" data-tab="history">History <span class="tab-badge" id="tb-history">0</span></button>
162
- </div>
163
-
164
- <div class="main">
165
- <div class="tab-panel active" id="panel-grades">
166
- <div class="sidebar" id="grade-sidebar"></div>
167
- <div class="content" id="grade-content"></div>
168
- </div>
169
- <div class="tab-panel" id="panel-ab">
170
- <div class="ab-panel" id="ab-content"></div>
171
- </div>
172
- <div class="tab-panel" id="panel-tokens">
173
- <div class="token-panel" id="token-content"></div>
174
- </div>
175
- <div class="tab-panel" id="panel-history">
176
- <div class="history-panel" id="history-content"></div>
177
- </div>
178
- </div>
179
-
180
- <script>
181
- /*__EMBEDDED_GRADE_DATA__*/
182
-
183
- // ── Helpers ────────────────────────────────────────────────────────────────
184
-
185
- function scoreLetter(s) {
186
- if (s == null) return '?';
187
- if (s >= 90) return 'A'; if (s >= 75) return 'B';
188
- if (s >= 60) return 'C'; if (s >= 45) return 'D'; return 'F';
189
- }
190
- function scoreColor(s) {
191
- if (s >= 90) return 'var(--grade-a)'; if (s >= 75) return 'var(--grade-b)';
192
- if (s >= 60) return 'var(--grade-c)'; if (s >= 45) return 'var(--grade-d)';
193
- return 'var(--grade-f)';
194
- }
195
- const DIM_COLORS = ['var(--dim1)','var(--dim2)','var(--dim3)','var(--dim4)','var(--dim5)'];
196
- const DIM_NAMES = ['Session Discipline','Discovery Efficiency','Task Hygiene','Error Protocol','Progressive Disclosure'];
197
- const DIM_KEYS = ['sessionDiscipline','discoveryEfficiency','taskHygiene','errorProtocol','disclosureUse'];
198
-
199
- function fmtNum(n, fallback) {
200
- if (n == null) return fallback !== undefined ? fallback : '\u2014';
201
- return typeof n === 'number' && !Number.isInteger(n) ? n.toFixed(1) : String(n);
202
- }
203
- function fmtDate(s) {
204
- if (!s) return '';
205
- try { return new Date(s).toLocaleString(undefined, { dateStyle: 'short', timeStyle: 'short' }); } catch { return s; }
206
- }
207
-
208
- // Safe element builder — never uses innerHTML with data
209
- function mk(tag, cls, text) {
210
- var e = document.createElement(tag);
211
- if (cls) e.className = cls;
212
- if (text != null) e.textContent = String(text);
213
- return e;
214
- }
215
- function mks(tag, css, text) { // inline style variant
216
- var e = document.createElement(tag);
217
- if (css) e.style.cssText = css;
218
- if (text != null) e.textContent = String(text);
219
- return e;
220
- }
221
- function app(parent) {
222
- for (var i = 1; i < arguments.length; i++) {
223
- if (arguments[i] != null) parent.appendChild(arguments[i]);
224
- }
225
- return parent;
226
- }
227
-
228
- // ── Tab switching ──────────────────────────────────────────────────────────
229
- document.querySelectorAll('.tab-btn').forEach(function(btn) {
230
- btn.addEventListener('click', function() {
231
- document.querySelectorAll('.tab-btn').forEach(function(b) { b.classList.remove('active'); });
232
- document.querySelectorAll('.tab-panel').forEach(function(p) { p.classList.remove('active'); });
233
- btn.classList.add('active');
234
- document.getElementById('panel-' + btn.dataset.tab).classList.add('active');
235
- });
236
- });
237
-
238
- // ── Empty state ────────────────────────────────────────────────────────────
239
- function emptyState(icon, title, desc, cmd) {
240
- var d = mk('div', 'empty-state');
241
- app(d, mks('div', 'font-size:2.5rem;', icon), mk('h3', null, title), mk('p', null, desc));
242
- if (cmd) app(d, mk('span', 'empty-cmd', cmd));
243
- return d;
244
- }
245
-
246
- // ── Grades Tab ─────────────────────────────────────────────────────────────
247
- function renderGradesTab(grades) {
248
- var sidebar = document.getElementById('grade-sidebar');
249
- var content = document.getElementById('grade-content');
250
- sidebar.textContent = '';
251
- content.textContent = '';
252
- if (!grades || !grades.length) {
253
- content.appendChild(emptyState('📊', 'No grade results yet',
254
- 'Run a scenario to generate grades:', 'python scripts/run_scenario.py --scenario S1 --cleo cleo-dev'));
255
- return;
256
- }
257
- document.getElementById('tb-grades').textContent = grades.length;
258
- app(sidebar, mk('div', 'sidebar-header', 'Sessions'));
259
- grades.forEach(function(g, i) {
260
- var score = g.totalScore != null ? g.totalScore : g.score;
261
- var letter = scoreLetter(score);
262
- var sid = (g.sessionId || '').slice(0, 22) || ('Run ' + (i + 1));
263
- var item = mk('div', 'sidebar-item' + (i === 0 ? ' active' : ''));
264
- var ls = mks('span', 'font-family:Poppins,sans-serif;font-weight:700;font-size:.85rem;color:' + scoreColor(score) + ';flex-shrink:0;width:18px;', letter);
265
- var lab = mk('span', 'item-label', sid);
266
- var sc = mks('span', null, score != null ? (score + '/100') : '\u2014');
267
- sc.className = 'item-score'; sc.style.color = scoreColor(score);
268
- app(item, ls, lab, sc);
269
- item.addEventListener('click', (function(g2, item2) {
270
- return function() {
271
- sidebar.querySelectorAll('.sidebar-item').forEach(function(x) { x.classList.remove('active'); });
272
- item2.classList.add('active');
273
- renderGradeDetail(g2, content);
274
- };
275
- })(g, item));
276
- sidebar.appendChild(item);
277
- });
278
- renderGradeDetail(grades[0], content);
279
- }
280
-
281
- function renderGradeDetail(g, container) {
282
- container.textContent = '';
283
- var score = g.totalScore != null ? g.totalScore : (g.score || 0);
284
- var max = g.maxScore || 100;
285
- var letter = scoreLetter(score);
286
- var pct = Math.round((score / max) * 100);
287
- var dims = g.dimensions || {};
288
- var flags = g.flags || [];
289
- var tmeta = g._tokenMeta || null;
290
- var scenario = g._scenarioId || g.scenario || '';
291
-
292
- // Score card
293
- var card = mk('div', 'grade-card');
294
- var hdr = mk('div', 'grade-card-header');
295
- var sb = mk('div', 'grade-score-block');
296
- var badge = mk('div', 'grade-letter-badge ' + letter, letter);
297
- var sd = mk('div', 'grade-score-details');
298
- var h2 = mk('h2'); h2.textContent = score + '/' + max; h2.style.color = scoreColor(score);
299
- var pctEl = mk('div', 'pct', pct + '% \u2014 Grade ' + letter + (scenario ? ' \u2014 ' + scenario : ''));
300
- app(sd, h2, pctEl);
301
- if (g.timestamp) app(sd, mks('div', 'font-size:.68rem;color:var(--text-muted);margin-top:.25rem;', fmtDate(g.timestamp)));
302
- app(sb, badge, sd);
303
- app(hdr, sb);
304
- if (g.sessionId) app(hdr, mk('code', 'grade-session-id', g.sessionId));
305
- card.appendChild(hdr);
306
-
307
- // Dimension bars
308
- var dimsWrap = mk('div', 'dimensions');
309
- DIM_KEYS.forEach(function(key, i) {
310
- var d = dims[key]; if (!d) return;
311
- var ds = d.score || 0, dm = d.max || 20, dp = Math.round((ds / dm) * 100);
312
- var row = mk('div', 'dimension-row');
313
- var bar = mk('div', 'dimension-bar');
314
- bar.style.cssText = 'width:' + dp + '%;background:' + DIM_COLORS[i] + ';';
315
- var bw = mk('div', 'dimension-bar-wrap'); bw.appendChild(bar);
316
- var sl = mk('span', 'dimension-score-label', ds + '/' + dm);
317
- sl.style.color = DIM_COLORS[i];
318
- app(row, mk('span', 'dimension-name', DIM_NAMES[i]), bw, sl);
319
- if (d.evidence && d.evidence.length) row.title = d.evidence.join(' | ');
320
- dimsWrap.appendChild(row);
321
- });
322
- card.appendChild(dimsWrap);
323
- container.appendChild(card);
324
-
325
- // Flags
326
- var fs = mk('div', 'grade-card flags-section');
327
- var ft = mk('h3', null, 'Flags' + (flags.length ? ' (' + flags.length + ')' : ''));
328
- var fl = mk('div', 'flags-list');
329
- if (!flags.length) {
330
- fl.appendChild(mk('div', 'no-flags', '\u2713 No flags \u2014 all protocol checks passed'));
331
- } else {
332
- flags.forEach(function(f) {
333
- var item = mk('div', 'flag-item');
334
- app(item, mks('span', 'flex-shrink:0;', '\u26A0'), mk('span', null, f));
335
- fl.appendChild(item);
336
- });
337
- }
338
- app(fs, ft, fl);
339
- container.appendChild(fs);
340
-
341
- // Evidence
342
- var hasEvidence = DIM_KEYS.some(function(k) { return dims[k] && dims[k].evidence && dims[k].evidence.length; });
343
- if (hasEvidence) {
344
- var ec = mk('div', 'grade-card');
345
- ec.appendChild(mk('div', 'ev-title', 'Evidence'));
346
- DIM_KEYS.forEach(function(key, i) {
347
- var d = dims[key]; if (!d || !d.evidence || !d.evidence.length) return;
348
- var row = mk('div', 'ev-row');
349
- var dimLabel = mk('div', 'ev-dim', DIM_NAMES[i]); dimLabel.style.color = DIM_COLORS[i];
350
- row.appendChild(dimLabel);
351
- d.evidence.forEach(function(ev) { row.appendChild(mk('div', 'ev-bullet', '\u2022 ' + ev)); });
352
- ec.appendChild(row);
353
- });
354
- container.appendChild(ec);
355
- }
356
-
357
- // Token metadata
358
- var tokCard = mk('div', 'grade-card');
359
- tokCard.appendChild(mk('div', 'ev-title', 'Token Metadata'));
360
- var grid = mk('div', 'token-meta-grid');
361
- function addStat(label, value) {
362
- var s = mk('div', 'token-stat');
363
- app(s, mk('div', 'token-stat-label', label), mk('div', 'token-stat-value', fmtNum(value, '\u2014')));
364
- grid.appendChild(s);
365
- }
366
- if (tmeta) {
367
- addStat('Method', tmeta.estimationMethod || tmeta.estimation_method || '\u2014');
368
- if (tmeta.totalEstimatedTokens) addStat('Est. Tokens', tmeta.totalEstimatedTokens);
369
- if (tmeta.inputTokens) addStat('Input Tokens', tmeta.inputTokens);
370
- if (tmeta.outputTokens) addStat('Output Tokens', tmeta.outputTokens);
371
- if (tmeta.cacheReadTokens) addStat('Cache Read', tmeta.cacheReadTokens);
372
- } else if (g.entryCount) {
373
- addStat('Audit Entries', g.entryCount);
374
- addStat('Est. Tokens', '~' + (g.entryCount * 150));
375
- addStat('Method', 'entry_count_proxy');
376
- } else {
377
- addStat('Note', 'Enable OTEL or use run_scenario.py for token data');
378
- }
379
- app(tokCard, grid);
380
- container.appendChild(tokCard);
381
- }
382
-
383
- // ── A/B Tab ────────────────────────────────────────────────────────────────
384
- function renderABTab(ab) {
385
- var container = document.getElementById('ab-content');
386
- container.textContent = '';
387
- if (!ab || !ab.total_runs) {
388
- container.appendChild(emptyState('⚖️', 'No A/B results yet',
389
- 'Run a blind comparison:', 'python scripts/run_ab_test.py --domain tasks --runs 3 --cleo cleo-dev'));
390
- return;
391
- }
392
- document.getElementById('tb-ab').textContent = ab.total_runs || 0;
393
- var wins = ab.global_wins || {}, wr = ab.global_win_rate || {};
394
- var delta = ab.avg_token_delta_mcp_minus_cli, winner = ab.overall_winner || 'tie';
395
- var banner = mk('div', 'ab-winner-banner ' + winner);
396
- var wlabel = winner === 'mcp' ? '\u26A1 MCP wins overall' : winner === 'cli' ? '\u26A1 CLI wins overall' : '\u2194 Overall tie';
397
- banner.appendChild(mk('span', null, wlabel));
398
- if (delta != null) {
399
- var sign = delta > 0 ? '+' : '';
400
- banner.appendChild(mks('span', 'font-size:.7rem;font-weight:400;opacity:.8;',
401
- ' \u2014 MCP uses ' + sign + delta.toFixed(1) + ' tokens/op vs CLI'));
402
- }
403
- container.appendChild(banner);
404
-
405
- var sumTitle = mk('div', 'section-title', 'Summary'); container.appendChild(sumTitle);
406
- var grid = mk('div', 'ab-summary-grid');
407
- [
408
- ['Total Runs', ab.total_runs, ''],
409
- ['MCP Wins', wins.mcp || 0, 'win'],
410
- ['CLI Wins', wins.cli || 0, winner === 'cli' ? 'win' : ''],
411
- ['Ties', wins.tie || 0, 'neutral'],
412
- ['MCP Win %', ((wr.mcp || 0) * 100).toFixed(1) + '%', winner === 'mcp' ? 'win' : ''],
413
- ['CLI Win %', ((wr.cli || 0) * 100).toFixed(1) + '%', winner === 'cli' ? 'win' : ''],
414
- ['Token Delta', delta != null ? ((delta > 0 ? '+' : '') + delta.toFixed(1) + 't') : '\u2014',
415
- delta > 10 ? 'loss' : delta < -10 ? 'win' : 'neutral'],
416
- ].forEach(function(row) {
417
- var c = mk('div', 'ab-stat-card');
418
- app(c, mk('div', 'ab-stat-label', row[0]), mk('div', 'ab-stat-value ' + row[2], String(row[1])));
419
- grid.appendChild(c);
420
- });
421
- container.appendChild(grid);
422
-
423
- var rawPerOp = ab.per_operation;
424
- var perOp = null;
425
- if (Array.isArray(rawPerOp)) {
426
- perOp = {};
427
- rawPerOp.forEach(function(s) { if (s && s.operation) perOp[s.operation] = s; });
428
- } else if (rawPerOp && typeof rawPerOp === 'object') {
429
- perOp = rawPerOp;
430
- }
431
- if (perOp && Object.keys(perOp).length) {
432
- container.appendChild(mks('div', 'font-family:Poppins,sans-serif;font-size:.75rem;font-weight:600;text-transform:uppercase;letter-spacing:.08em;color:var(--text-muted);margin-top:.5rem;', 'Per-Operation Results'));
433
- var wrap = mk('div', 'ab-table-wrap');
434
- var tbl = document.createElement('table'); tbl.className = 'ab-table';
435
- var thead = document.createElement('thead'), hrow = document.createElement('tr');
436
- ['Operation','Winner','MCP Wins','CLI Wins','Ties','Token \u0394','MCP chars','CLI chars','MCP ms','CLI ms'].forEach(function(h) {
437
- hrow.appendChild(mk('th', null, h));
438
- });
439
- thead.appendChild(hrow); tbl.appendChild(thead);
440
- var tbody = document.createElement('tbody');
441
- Object.keys(perOp).sort().forEach(function(opKey) {
442
- var d = perOp[opKey], row = document.createElement('tr');
443
- var delta2 = d.avg_token_delta != null ? d.avg_token_delta : 0;
444
- var dcls = delta2 > 10 ? 'delta-pos' : delta2 < -10 ? 'delta-neg' : 'delta-zero';
445
- var opW = d.winner || (d.mcp_wins > d.cli_wins ? 'mcp' : d.cli_wins > d.mcp_wins ? 'cli' : 'tie');
446
- var cells = [
447
- mk('span', 'op-name', opKey),
448
- mk('span', 'win-pill ' + opW, opW.toUpperCase()),
449
- document.createTextNode(String(d.mcp_wins || 0)),
450
- document.createTextNode(String(d.cli_wins || 0)),
451
- document.createTextNode(String(d.ties || 0)),
452
- mk('span', dcls, (delta2 > 0 ? '+' : '') + delta2.toFixed(0) + 't'),
453
- document.createTextNode(fmtNum(d.avg_mcp_chars)),
454
- document.createTextNode(fmtNum(d.avg_cli_chars)),
455
- document.createTextNode(fmtNum(d.avg_mcp_ms) + 'ms'),
456
- document.createTextNode(fmtNum(d.avg_cli_ms) + 'ms'),
457
- ];
458
- cells.forEach(function(c) { var td = document.createElement('td'); td.appendChild(c); row.appendChild(td); });
459
- tbody.appendChild(row);
460
- });
461
- tbl.appendChild(tbody); wrap.appendChild(tbl); container.appendChild(wrap);
462
- }
463
- container.appendChild(mk('div', 'ab-note',
464
- '\u26A0 CLI-only sessions always score 0 on S5 (Progressive Disclosure) — metadata.gateway is not set by the CLI adapter. MCP earns +10 for query gateway usage automatically.'));
465
- }
466
-
467
- // ── Token Analysis Tab ─────────────────────────────────────────────────────
468
- function renderTokenTab(ta) {
469
- var container = document.getElementById('token-content');
470
- container.textContent = '';
471
- if (!ta || !Object.keys(ta).length) {
472
- container.appendChild(emptyState('\uD83D\uDD22', 'No token data yet',
473
- 'Generate token analysis:', 'python scripts/token_tracker.py --grades-file .cleo/metrics/GRADES.jsonl --breakdown-by domain'));
474
- return;
475
- }
476
- function barChart(title, entries, maxVal, colorFn) {
477
- var card = mk('div', 'chart-card');
478
- card.appendChild(mk('div', 'chart-card-title', title));
479
- var chart = mk('div', 'bar-chart');
480
- entries.forEach(function(row) {
481
- var label = row[0], val = row[1], color = colorFn ? colorFn(row) : 'var(--accent)';
482
- var pct = maxVal > 0 ? Math.min((Math.abs(val) / maxVal) * 90, 90) : 0;
483
- var r = mk('div', 'bar-row');
484
- var track = mk('div', 'bar-track');
485
- var fill = mk('div', 'bar-fill'); fill.style.cssText = 'width:' + pct + '%;background:' + color + ';';
486
- track.appendChild(fill);
487
- app(r, mk('span', 'bar-label', label), track, mk('span', 'bar-val', fmtNum(val) + 't'));
488
- chart.appendChild(r);
489
- });
490
- card.appendChild(chart);
491
- return card;
492
- }
493
- if (ta.breakdown_by_domain) {
494
- var entries = Object.entries(ta.breakdown_by_domain).map(function(kv) { return [kv[0], kv[1].mean || 0]; });
495
- var maxD = Math.max.apply(null, entries.map(function(e) { return e[1]; }).concat([1]));
496
- var colors2 = ['var(--dim1)','var(--dim2)','var(--dim3)','var(--dim4)','var(--dim5)','var(--accent)','var(--mcp)','var(--cli)','#9b7ec8','#5a9a6a'];
497
- container.appendChild(barChart('Estimated Tokens by Domain', entries, maxD, function(row, i) { return colors2[entries.indexOf(row) % colors2.length]; }));
498
- }
499
- if (ta.breakdown_by_gateway) {
500
- var gw = ta.breakdown_by_gateway;
501
- var gwEntries = [['query (MCP)', gw.mcp_query && gw.mcp_query.mean || 0], ['CLI', gw.cli && gw.cli.mean || 0]];
502
- var maxGw = Math.max.apply(null, gwEntries.map(function(e) { return e[1]; }).concat([1]));
503
- var gwColors = ['var(--mcp)', 'var(--cli)'];
504
- container.appendChild(barChart('MCP vs CLI Token Split', gwEntries, maxGw, function(row) { return gwColors[gwEntries.indexOf(row)]; }));
505
- if (gw.note) {
506
- var noteEl = mks('p', 'font-size:.7rem;color:var(--text-muted);margin-top:.5rem;line-height:1.5;', gw.note);
507
- container.lastChild.appendChild(noteEl);
508
- }
509
- }
510
- if (ta.ab_aggregation && ta.ab_aggregation.per_operation) {
511
- var opEntries = Object.entries(ta.ab_aggregation.per_operation).map(function(kv) { return [kv[0], kv[1].avg_delta || 0]; });
512
- opEntries.sort(function(a,b) { return b[1] - a[1]; });
513
- var maxOp = Math.max.apply(null, opEntries.map(function(e) { return Math.abs(e[1]); }).concat([1]));
514
- container.appendChild(barChart('Token Delta per Op (MCP \u2212 CLI)', opEntries, maxOp, function(row) {
515
- return row[1] > 0 ? 'var(--grade-d)' : 'var(--win)';
516
- }));
517
- }
518
- }
519
-
520
- // ── History Tab ────────────────────────────────────────────────────────────
521
- function renderHistoryTab(grades) {
522
- var container = document.getElementById('history-content');
523
- container.textContent = '';
524
- if (!grades || !grades.length) {
525
- container.appendChild(emptyState('\uD83D\uDCC8', 'No history yet', 'Run graded sessions to build history.', ''));
526
- return;
527
- }
528
- document.getElementById('tb-history').textContent = grades.length;
529
- if (grades.length > 1) {
530
- var sorted = grades.slice().sort(function(a,b) { return new Date(a.timestamp) - new Date(b.timestamp); });
531
- var scores = sorted.map(function(g) { return g.totalScore != null ? g.totalScore : (g.score || 0); });
532
- var trendCard = mk('div', 'chart-card');
533
- trendCard.appendChild(mk('div', 'chart-card-title', 'Score Trend'));
534
- var canvas = document.createElement('canvas');
535
- canvas.className = 'trend'; canvas.width = 800; canvas.height = 90;
536
- trendCard.appendChild(canvas);
537
- container.appendChild(trendCard);
538
- requestAnimationFrame(function() { drawSparkline(canvas, scores); });
539
- }
540
- var desc = grades.slice().sort(function(a,b) { return new Date(b.timestamp) - new Date(a.timestamp); });
541
- container.appendChild(mk('div', 'section-title', 'All Sessions'));
542
- var list = mk('div', 'history-list');
543
- desc.forEach(function(g, idx) {
544
- var score = g.totalScore != null ? g.totalScore : (g.score || 0);
545
- var letter = scoreLetter(score);
546
- var pct = Math.round((score / (g.maxScore || 100)) * 100);
547
- var flags = (g.flags || []).length;
548
- var item = mk('div', 'history-item');
549
- var badge = mk('div', 'history-grade-badge ' + letter, letter);
550
- var main = mk('div', 'history-item-main');
551
- var sid = mk('div', 'history-session-id', g.sessionId || '(no session id)');
552
- var meta = mk('div', 'history-item-meta');
553
- app(meta, mks('span', null, fmtDate(g.timestamp)), mks('span', null, score + '/100'));
554
- if (flags > 0) app(meta, mks('span', 'color:var(--loss);', flags + ' flag' + (flags !== 1 ? 's' : '')));
555
- if (g._scenarioId || g.scenario) app(meta, mks('span', 'color:var(--text-muted);', g._scenarioId || g.scenario));
556
- app(main, sid, meta);
557
- var barWrap = mk('div', 'history-score-bar');
558
- var fill = mk('div', 'history-score-fill');
559
- fill.style.cssText = 'width:' + pct + '%;background:' + scoreColor(score) + ';';
560
- barWrap.appendChild(fill);
561
- app(item, badge, main, barWrap);
562
- item.addEventListener('click', (function(g2) {
563
- return function() {
564
- document.querySelector('[data-tab="grades"]').click();
565
- var items = document.querySelectorAll('#grade-sidebar .sidebar-item');
566
- var allGrades = EMBEDDED_GRADE_DATA.grades || [];
567
- for (var j = 0; j < allGrades.length; j++) {
568
- if (allGrades[j].sessionId === g2.sessionId && items[j]) { items[j].click(); break; }
569
- }
570
- };
571
- })(g));
572
- list.appendChild(item);
573
- });
574
- container.appendChild(list);
575
- }
576
-
577
- function drawSparkline(canvas, scores) {
578
- var W = canvas.clientWidth || 600, H = canvas.clientHeight || 90;
579
- canvas.width = W; canvas.height = H;
580
- var ctx = canvas.getContext('2d');
581
- var pad = 10, uW = W - pad*2, uH = H - pad*2;
582
- var toX = function(i) { return pad + (i / (scores.length - 1)) * uW; };
583
- var toY = function(s) { return pad + (1 - s / 100) * uH; };
584
- var grad = ctx.createLinearGradient(0, pad, 0, H - pad);
585
- grad.addColorStop(0, 'rgba(73,143,165,.3)'); grad.addColorStop(1, 'rgba(73,143,165,.02)');
586
- ctx.beginPath();
587
- scores.forEach(function(s,i) { i===0 ? ctx.moveTo(toX(i),toY(s)) : ctx.lineTo(toX(i),toY(s)); });
588
- ctx.lineTo(toX(scores.length-1), H-pad); ctx.lineTo(toX(0), H-pad); ctx.closePath();
589
- ctx.fillStyle = grad; ctx.fill();
590
- ctx.beginPath();
591
- scores.forEach(function(s,i) { i===0 ? ctx.moveTo(toX(i),toY(s)) : ctx.lineTo(toX(i),toY(s)); });
592
- ctx.strokeStyle = '#4a8fa5'; ctx.lineWidth = 2; ctx.lineJoin = 'round'; ctx.stroke();
593
- [90,75,60,45].forEach(function(t) {
594
- ctx.beginPath(); ctx.moveTo(pad,toY(t)); ctx.lineTo(W-pad,toY(t));
595
- ctx.strokeStyle = 'rgba(0,0,0,.07)'; ctx.lineWidth = 1; ctx.setLineDash([3,4]); ctx.stroke(); ctx.setLineDash([]);
596
- });
597
- scores.forEach(function(s,i) {
598
- ctx.beginPath(); ctx.arc(toX(i),toY(s),3.5,0,Math.PI*2); ctx.fillStyle = scoreColor(s); ctx.fill();
599
- });
600
- }
601
-
602
- // ── Init ───────────────────────────────────────────────────────────────────
603
- (function() {
604
- var data = typeof EMBEDDED_GRADE_DATA !== 'undefined' ? EMBEDDED_GRADE_DATA : {};
605
- document.getElementById('hdr-title').textContent = data.title || 'ct-grade \u2014 Review';
606
- if (data.subtitle) document.getElementById('hdr-sub').textContent = data.subtitle;
607
- if (data.generated_at) {
608
- var m = document.getElementById('hdr-meta');
609
- m.appendChild(document.createTextNode('Generated ' + fmtDate(data.generated_at)));
610
- m.appendChild(document.createElement('br'));
611
- m.appendChild(document.createTextNode((data.grades || []).length + ' grade(s) \u00b7 ' + ((data.ab_results && data.ab_results.total_runs) || 0) + ' A/B run(s)'));
612
- }
613
- renderGradesTab(data.grades || []);
614
- renderABTab(data.ab_results || {});
615
- renderTokenTab(data.token_analysis || {});
616
- renderHistoryTab(data.grades || []);
617
- })();
618
- </script>
619
- </body>
620
- </html>
@@ -1,31 +0,0 @@
1
- {
2
- "_comment": "CLEO-only metadata — merge into packages/skills/skills/manifest.json",
3
- "name": "ct-grade",
4
- "version": "2.1.0",
5
- "tier": 2,
6
- "token_budget": 16000,
7
- "capabilities": {
8
- "inputs": ["SESSION_ID", "SCENARIO_ID", "DOMAIN", "OPERATION_LIST"],
9
- "outputs": ["grade-result", "ab-comparison", "token-report", "analysis-report"],
10
- "dispatch_triggers": ["grade", "ab-test", "blind-test", "token-efficiency", "scenario"],
11
- "compatible_subagent_types": ["general-purpose", "Explore"],
12
- "agents": ["scenario-runner", "blind-comparator", "analysis-reporter"],
13
- "chains_to": ["ct-cleo", "ct-research-agent"],
14
- "dispatch_keywords": {
15
- "primary": ["grade", "ab-test", "blind-test", "token-efficiency"],
16
- "secondary": ["mcp-cli-comparison", "behavioral-compliance", "session-quality", "playbook"]
17
- }
18
- },
19
- "constraints": {
20
- "max_context_tokens": 80000,
21
- "requires_session": false,
22
- "requires_epic": false
23
- },
24
- "references": [
25
- "skills/ct-grade-v2-1/references/grade-spec-v2.md",
26
- "skills/ct-grade-v2-1/references/playbook-v2.md",
27
- "skills/ct-grade-v2-1/references/ab-testing.md",
28
- "skills/ct-grade-v2-1/references/token-tracking.md",
29
- "skills/ct-grade-v2-1/references/domains-ssot.md"
30
- ]
31
- }