@sentry/warden 0.30.0 → 0.31.0
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.
- package/CHANGELOG.md +246 -0
- package/README.md +2 -1
- package/coverage/lcov-report/base.css +224 -0
- package/coverage/lcov-report/block-navigation.js +87 -0
- package/coverage/lcov-report/favicon.png +0 -0
- package/coverage/lcov-report/index.html +446 -0
- package/coverage/lcov-report/prettify.css +1 -0
- package/coverage/lcov-report/prettify.js +2 -0
- package/coverage/lcov-report/sort-arrow-sprite.png +0 -0
- package/coverage/lcov-report/sorter.js +210 -0
- package/coverage/lcov-report/src/action/checks/index.html +116 -0
- package/coverage/lcov-report/src/action/checks/manager.ts.html +499 -0
- package/coverage/lcov-report/src/action/error-reporting.ts.html +181 -0
- package/coverage/lcov-report/src/action/fix-evaluation/github.ts.html +508 -0
- package/coverage/lcov-report/src/action/fix-evaluation/index.html +161 -0
- package/coverage/lcov-report/src/action/fix-evaluation/index.ts.html +1243 -0
- package/coverage/lcov-report/src/action/fix-evaluation/judge.ts.html +817 -0
- package/coverage/lcov-report/src/action/fix-evaluation/types.ts.html +274 -0
- package/coverage/lcov-report/src/action/index.html +161 -0
- package/coverage/lcov-report/src/action/inputs.ts.html +529 -0
- package/coverage/lcov-report/src/action/pi-ncc-compat.ts.html +292 -0
- package/coverage/lcov-report/src/action/review/coordination.ts.html +208 -0
- package/coverage/lcov-report/src/action/review/index.html +131 -0
- package/coverage/lcov-report/src/action/review/poster.ts.html +1051 -0
- package/coverage/lcov-report/src/action/review-state.ts.html +280 -0
- package/coverage/lcov-report/src/action/triggers/executor.ts.html +856 -0
- package/coverage/lcov-report/src/action/triggers/index.html +116 -0
- package/coverage/lcov-report/src/action/workflow/base.ts.html +1135 -0
- package/coverage/lcov-report/src/action/workflow/index.html +146 -0
- package/coverage/lcov-report/src/action/workflow/pr-workflow.ts.html +2713 -0
- package/coverage/lcov-report/src/action/workflow/schedule.ts.html +1027 -0
- package/coverage/lcov-report/src/cli/args.ts.html +1657 -0
- package/coverage/lcov-report/src/cli/commands/add.ts.html +1567 -0
- package/coverage/lcov-report/src/cli/commands/build.ts.html +1618 -0
- package/coverage/lcov-report/src/cli/commands/index.html +191 -0
- package/coverage/lcov-report/src/cli/commands/init.ts.html +1357 -0
- package/coverage/lcov-report/src/cli/commands/runs.ts.html +2560 -0
- package/coverage/lcov-report/src/cli/commands/setup-app/browser.ts.html +214 -0
- package/coverage/lcov-report/src/cli/commands/setup-app/credentials.ts.html +220 -0
- package/coverage/lcov-report/src/cli/commands/setup-app/index.html +161 -0
- package/coverage/lcov-report/src/cli/commands/setup-app/manifest.ts.html +235 -0
- package/coverage/lcov-report/src/cli/commands/setup-app/server.ts.html +601 -0
- package/coverage/lcov-report/src/cli/commands/setup-app.ts.html +571 -0
- package/coverage/lcov-report/src/cli/commands/sync.ts.html +475 -0
- package/coverage/lcov-report/src/cli/context.ts.html +532 -0
- package/coverage/lcov-report/src/cli/diff-apply.ts.html +334 -0
- package/coverage/lcov-report/src/cli/files.ts.html +904 -0
- package/coverage/lcov-report/src/cli/fix.ts.html +1060 -0
- package/coverage/lcov-report/src/cli/git.ts.html +1201 -0
- package/coverage/lcov-report/src/cli/help.ts.html +1696 -0
- package/coverage/lcov-report/src/cli/index.html +281 -0
- package/coverage/lcov-report/src/cli/input.ts.html +397 -0
- package/coverage/lcov-report/src/cli/log-cleanup.ts.html +337 -0
- package/coverage/lcov-report/src/cli/main.ts.html +5107 -0
- package/coverage/lcov-report/src/cli/output/box.ts.html +790 -0
- package/coverage/lcov-report/src/cli/output/formatters.ts.html +1099 -0
- package/coverage/lcov-report/src/cli/output/icons.ts.html +142 -0
- package/coverage/lcov-report/src/cli/output/index.html +281 -0
- package/coverage/lcov-report/src/cli/output/index.ts.html +295 -0
- package/coverage/lcov-report/src/cli/output/ink-runner.tsx.html +1597 -0
- package/coverage/lcov-report/src/cli/output/jsonl-schema-gen.ts.html +355 -0
- package/coverage/lcov-report/src/cli/output/jsonl.ts.html +2137 -0
- package/coverage/lcov-report/src/cli/output/live-status.tsx.html +916 -0
- package/coverage/lcov-report/src/cli/output/reporter.ts.html +1630 -0
- package/coverage/lcov-report/src/cli/output/tasks.ts.html +3193 -0
- package/coverage/lcov-report/src/cli/output/tty.ts.html +310 -0
- package/coverage/lcov-report/src/cli/output/verbosity.ts.html +181 -0
- package/coverage/lcov-report/src/cli/signals.ts.html +199 -0
- package/coverage/lcov-report/src/cli/terminal.ts.html +1060 -0
- package/coverage/lcov-report/src/config/index.html +161 -0
- package/coverage/lcov-report/src/config/index.ts.html +94 -0
- package/coverage/lcov-report/src/config/loader.ts.html +1753 -0
- package/coverage/lcov-report/src/config/schema.ts.html +910 -0
- package/coverage/lcov-report/src/config/writer.ts.html +538 -0
- package/coverage/lcov-report/src/diff/apply.ts.html +241 -0
- package/coverage/lcov-report/src/diff/classify.ts.html +361 -0
- package/coverage/lcov-report/src/diff/coalesce.ts.html +1123 -0
- package/coverage/lcov-report/src/diff/context.ts.html +952 -0
- package/coverage/lcov-report/src/diff/index.html +191 -0
- package/coverage/lcov-report/src/diff/index.ts.html +100 -0
- package/coverage/lcov-report/src/diff/parser.ts.html +532 -0
- package/coverage/lcov-report/src/evals/harness.ts.html +700 -0
- package/coverage/lcov-report/src/evals/index.html +221 -0
- package/coverage/lcov-report/src/evals/index.ts.html +955 -0
- package/coverage/lcov-report/src/evals/names.ts.html +121 -0
- package/coverage/lcov-report/src/evals/runner.ts.html +664 -0
- package/coverage/lcov-report/src/evals/scaffold.ts.html +970 -0
- package/coverage/lcov-report/src/evals/types.ts.html +661 -0
- package/coverage/lcov-report/src/evals/usage.ts.html +178 -0
- package/coverage/lcov-report/src/evals/verify.ts.html +1054 -0
- package/coverage/lcov-report/src/event/context.ts.html +490 -0
- package/coverage/lcov-report/src/event/index.html +116 -0
- package/coverage/lcov-report/src/index.html +116 -0
- package/coverage/lcov-report/src/output/dedup.ts.html +2950 -0
- package/coverage/lcov-report/src/output/github-checks.ts.html +1579 -0
- package/coverage/lcov-report/src/output/index.html +161 -0
- package/coverage/lcov-report/src/output/renderer.ts.html +988 -0
- package/coverage/lcov-report/src/output/stale.ts.html +640 -0
- package/coverage/lcov-report/src/sdk/analyze.ts.html +3010 -0
- package/coverage/lcov-report/src/sdk/auth.ts.html +217 -0
- package/coverage/lcov-report/src/sdk/circuit-breaker.ts.html +304 -0
- package/coverage/lcov-report/src/sdk/errors.ts.html +658 -0
- package/coverage/lcov-report/src/sdk/extract.ts.html +1810 -0
- package/coverage/lcov-report/src/sdk/fix-quality.ts.html +841 -0
- package/coverage/lcov-report/src/sdk/haiku.ts.html +1456 -0
- package/coverage/lcov-report/src/sdk/index.html +416 -0
- package/coverage/lcov-report/src/sdk/json-output.ts.html +508 -0
- package/coverage/lcov-report/src/sdk/model-pricing.json.html +499 -0
- package/coverage/lcov-report/src/sdk/otel.ts.html +433 -0
- package/coverage/lcov-report/src/sdk/post-process.ts.html +412 -0
- package/coverage/lcov-report/src/sdk/prepare.ts.html +400 -0
- package/coverage/lcov-report/src/sdk/pricing.ts.html +778 -0
- package/coverage/lcov-report/src/sdk/prompt-sections.ts.html +595 -0
- package/coverage/lcov-report/src/sdk/prompt.ts.html +451 -0
- package/coverage/lcov-report/src/sdk/report-files.ts.html +166 -0
- package/coverage/lcov-report/src/sdk/retry.ts.html +202 -0
- package/coverage/lcov-report/src/sdk/runner.ts.html +400 -0
- package/coverage/lcov-report/src/sdk/runtimes/claude.ts.html +1693 -0
- package/coverage/lcov-report/src/sdk/runtimes/index.html +176 -0
- package/coverage/lcov-report/src/sdk/runtimes/index.ts.html +244 -0
- package/coverage/lcov-report/src/sdk/runtimes/model-selectors.ts.html +328 -0
- package/coverage/lcov-report/src/sdk/runtimes/pi.ts.html +2104 -0
- package/coverage/lcov-report/src/sdk/runtimes/types.ts.html +508 -0
- package/coverage/lcov-report/src/sdk/types.ts.html +730 -0
- package/coverage/lcov-report/src/sdk/usage.ts.html +502 -0
- package/coverage/lcov-report/src/sdk/verify.ts.html +1027 -0
- package/coverage/lcov-report/src/sentry.ts.html +994 -0
- package/coverage/lcov-report/src/skill-builder/agentic.ts.html +1063 -0
- package/coverage/lcov-report/src/skill-builder/authoring-provider.ts.html +334 -0
- package/coverage/lcov-report/src/skill-builder/definition.ts.html +748 -0
- package/coverage/lcov-report/src/skill-builder/index.html +236 -0
- package/coverage/lcov-report/src/skill-builder/outline-contract.ts.html +400 -0
- package/coverage/lcov-report/src/skill-builder/outline-state.ts.html +382 -0
- package/coverage/lcov-report/src/skill-builder/outline.ts.html +1402 -0
- package/coverage/lcov-report/src/skill-builder/skill-contract.ts.html +382 -0
- package/coverage/lcov-report/src/skill-builder/skill-prompts.ts.html +1180 -0
- package/coverage/lcov-report/src/skill-builder/skill.ts.html +2398 -0
- package/coverage/lcov-report/src/skills/index.html +131 -0
- package/coverage/lcov-report/src/skills/loader.ts.html +1693 -0
- package/coverage/lcov-report/src/skills/remote.ts.html +1909 -0
- package/coverage/lcov-report/src/triggers/index.html +116 -0
- package/coverage/lcov-report/src/triggers/matcher.ts.html +841 -0
- package/coverage/lcov-report/src/types/index.html +116 -0
- package/coverage/lcov-report/src/types/index.ts.html +1240 -0
- package/coverage/lcov-report/src/utils/async.ts.html +331 -0
- package/coverage/lcov-report/src/utils/exec.ts.html +496 -0
- package/coverage/lcov-report/src/utils/index.html +176 -0
- package/coverage/lcov-report/src/utils/index.ts.html +313 -0
- package/coverage/lcov-report/src/utils/path.ts.html +202 -0
- package/coverage/lcov-report/src/utils/version.ts.html +136 -0
- package/coverage/lcov.info +19134 -0
- package/dist/cli/commands/runs.d.ts +1 -1
- package/dist/cli/commands/runs.d.ts.map +1 -1
- package/dist/cli/commands/runs.js +1 -1
- package/dist/cli/commands/runs.js.map +1 -1
- package/dist/cli/files.d.ts.map +1 -1
- package/dist/cli/files.js +10 -11
- package/dist/cli/files.js.map +1 -1
- package/dist/sdk/verify.d.ts.map +1 -1
- package/dist/sdk/verify.js +3 -2
- package/dist/sdk/verify.js.map +1 -1
- package/dist/skill-builder/outline.d.ts.map +1 -1
- package/dist/skill-builder/outline.js.map +1 -1
- package/package.json +8 -3
- package/test-results.junit.xml +4329 -0
- package/dist/evals/index.d.ts +0 -22
- package/dist/evals/index.d.ts.map +0 -1
- package/dist/evals/index.js +0 -92
- package/dist/evals/index.js.map +0 -1
- package/dist/evals/judge.d.ts +0 -12
- package/dist/evals/judge.d.ts.map +0 -1
- package/dist/evals/judge.js +0 -171
- package/dist/evals/judge.js.map +0 -1
- package/dist/evals/runner.d.ts +0 -18
- package/dist/evals/runner.d.ts.map +0 -1
- package/dist/evals/runner.js +0 -149
- package/dist/evals/runner.js.map +0 -1
- package/dist/evals/setup.d.ts +0 -2
- package/dist/evals/setup.d.ts.map +0 -1
- package/dist/evals/setup.js +0 -22
- package/dist/evals/setup.js.map +0 -1
- package/dist/evals/types.d.ts +0 -160
- package/dist/evals/types.d.ts.map +0 -1
- package/dist/evals/types.js +0 -148
- package/dist/evals/types.js.map +0 -1
|
@@ -0,0 +1,661 @@
|
|
|
1
|
+
|
|
2
|
+
<!doctype html>
|
|
3
|
+
<html lang="en">
|
|
4
|
+
|
|
5
|
+
<head>
|
|
6
|
+
<title>Code coverage report for src/evals/types.ts</title>
|
|
7
|
+
<meta charset="utf-8" />
|
|
8
|
+
<link rel="stylesheet" href="../../prettify.css" />
|
|
9
|
+
<link rel="stylesheet" href="../../base.css" />
|
|
10
|
+
<link rel="shortcut icon" type="image/x-icon" href="../../favicon.png" />
|
|
11
|
+
<meta name="viewport" content="width=device-width, initial-scale=1" />
|
|
12
|
+
<style type='text/css'>
|
|
13
|
+
.coverage-summary .sorter {
|
|
14
|
+
background-image: url(../../sort-arrow-sprite.png);
|
|
15
|
+
}
|
|
16
|
+
</style>
|
|
17
|
+
</head>
|
|
18
|
+
|
|
19
|
+
<body>
|
|
20
|
+
<div class='wrapper'>
|
|
21
|
+
<div class='pad1'>
|
|
22
|
+
<h1><a href="../../index.html">All files</a> / <a href="index.html">src/evals</a> types.ts</h1>
|
|
23
|
+
<div class='clearfix'>
|
|
24
|
+
|
|
25
|
+
<div class='fl pad1y space-right2'>
|
|
26
|
+
<span class="strong">100% </span>
|
|
27
|
+
<span class="quiet">Statements</span>
|
|
28
|
+
<span class='fraction'>24/24</span>
|
|
29
|
+
</div>
|
|
30
|
+
|
|
31
|
+
|
|
32
|
+
<div class='fl pad1y space-right2'>
|
|
33
|
+
<span class="strong">100% </span>
|
|
34
|
+
<span class="quiet">Branches</span>
|
|
35
|
+
<span class='fraction'>16/16</span>
|
|
36
|
+
</div>
|
|
37
|
+
|
|
38
|
+
|
|
39
|
+
<div class='fl pad1y space-right2'>
|
|
40
|
+
<span class="strong">100% </span>
|
|
41
|
+
<span class="quiet">Functions</span>
|
|
42
|
+
<span class='fraction'>1/1</span>
|
|
43
|
+
</div>
|
|
44
|
+
|
|
45
|
+
|
|
46
|
+
<div class='fl pad1y space-right2'>
|
|
47
|
+
<span class="strong">100% </span>
|
|
48
|
+
<span class="quiet">Lines</span>
|
|
49
|
+
<span class='fraction'>23/23</span>
|
|
50
|
+
</div>
|
|
51
|
+
|
|
52
|
+
|
|
53
|
+
</div>
|
|
54
|
+
<p class="quiet">
|
|
55
|
+
Press <em>n</em> or <em>j</em> to go to the next uncovered block, <em>b</em>, <em>p</em> or <em>k</em> for the previous block.
|
|
56
|
+
</p>
|
|
57
|
+
<template id="filterTemplate">
|
|
58
|
+
<div class="quiet">
|
|
59
|
+
Filter:
|
|
60
|
+
<input type="search" id="fileSearch">
|
|
61
|
+
</div>
|
|
62
|
+
</template>
|
|
63
|
+
</div>
|
|
64
|
+
<div class='status-line high'></div>
|
|
65
|
+
<pre><table class="coverage">
|
|
66
|
+
<tr><td class="line-count quiet"><a name='L1'></a><a href='#L1'>1</a>
|
|
67
|
+
<a name='L2'></a><a href='#L2'>2</a>
|
|
68
|
+
<a name='L3'></a><a href='#L3'>3</a>
|
|
69
|
+
<a name='L4'></a><a href='#L4'>4</a>
|
|
70
|
+
<a name='L5'></a><a href='#L5'>5</a>
|
|
71
|
+
<a name='L6'></a><a href='#L6'>6</a>
|
|
72
|
+
<a name='L7'></a><a href='#L7'>7</a>
|
|
73
|
+
<a name='L8'></a><a href='#L8'>8</a>
|
|
74
|
+
<a name='L9'></a><a href='#L9'>9</a>
|
|
75
|
+
<a name='L10'></a><a href='#L10'>10</a>
|
|
76
|
+
<a name='L11'></a><a href='#L11'>11</a>
|
|
77
|
+
<a name='L12'></a><a href='#L12'>12</a>
|
|
78
|
+
<a name='L13'></a><a href='#L13'>13</a>
|
|
79
|
+
<a name='L14'></a><a href='#L14'>14</a>
|
|
80
|
+
<a name='L15'></a><a href='#L15'>15</a>
|
|
81
|
+
<a name='L16'></a><a href='#L16'>16</a>
|
|
82
|
+
<a name='L17'></a><a href='#L17'>17</a>
|
|
83
|
+
<a name='L18'></a><a href='#L18'>18</a>
|
|
84
|
+
<a name='L19'></a><a href='#L19'>19</a>
|
|
85
|
+
<a name='L20'></a><a href='#L20'>20</a>
|
|
86
|
+
<a name='L21'></a><a href='#L21'>21</a>
|
|
87
|
+
<a name='L22'></a><a href='#L22'>22</a>
|
|
88
|
+
<a name='L23'></a><a href='#L23'>23</a>
|
|
89
|
+
<a name='L24'></a><a href='#L24'>24</a>
|
|
90
|
+
<a name='L25'></a><a href='#L25'>25</a>
|
|
91
|
+
<a name='L26'></a><a href='#L26'>26</a>
|
|
92
|
+
<a name='L27'></a><a href='#L27'>27</a>
|
|
93
|
+
<a name='L28'></a><a href='#L28'>28</a>
|
|
94
|
+
<a name='L29'></a><a href='#L29'>29</a>
|
|
95
|
+
<a name='L30'></a><a href='#L30'>30</a>
|
|
96
|
+
<a name='L31'></a><a href='#L31'>31</a>
|
|
97
|
+
<a name='L32'></a><a href='#L32'>32</a>
|
|
98
|
+
<a name='L33'></a><a href='#L33'>33</a>
|
|
99
|
+
<a name='L34'></a><a href='#L34'>34</a>
|
|
100
|
+
<a name='L35'></a><a href='#L35'>35</a>
|
|
101
|
+
<a name='L36'></a><a href='#L36'>36</a>
|
|
102
|
+
<a name='L37'></a><a href='#L37'>37</a>
|
|
103
|
+
<a name='L38'></a><a href='#L38'>38</a>
|
|
104
|
+
<a name='L39'></a><a href='#L39'>39</a>
|
|
105
|
+
<a name='L40'></a><a href='#L40'>40</a>
|
|
106
|
+
<a name='L41'></a><a href='#L41'>41</a>
|
|
107
|
+
<a name='L42'></a><a href='#L42'>42</a>
|
|
108
|
+
<a name='L43'></a><a href='#L43'>43</a>
|
|
109
|
+
<a name='L44'></a><a href='#L44'>44</a>
|
|
110
|
+
<a name='L45'></a><a href='#L45'>45</a>
|
|
111
|
+
<a name='L46'></a><a href='#L46'>46</a>
|
|
112
|
+
<a name='L47'></a><a href='#L47'>47</a>
|
|
113
|
+
<a name='L48'></a><a href='#L48'>48</a>
|
|
114
|
+
<a name='L49'></a><a href='#L49'>49</a>
|
|
115
|
+
<a name='L50'></a><a href='#L50'>50</a>
|
|
116
|
+
<a name='L51'></a><a href='#L51'>51</a>
|
|
117
|
+
<a name='L52'></a><a href='#L52'>52</a>
|
|
118
|
+
<a name='L53'></a><a href='#L53'>53</a>
|
|
119
|
+
<a name='L54'></a><a href='#L54'>54</a>
|
|
120
|
+
<a name='L55'></a><a href='#L55'>55</a>
|
|
121
|
+
<a name='L56'></a><a href='#L56'>56</a>
|
|
122
|
+
<a name='L57'></a><a href='#L57'>57</a>
|
|
123
|
+
<a name='L58'></a><a href='#L58'>58</a>
|
|
124
|
+
<a name='L59'></a><a href='#L59'>59</a>
|
|
125
|
+
<a name='L60'></a><a href='#L60'>60</a>
|
|
126
|
+
<a name='L61'></a><a href='#L61'>61</a>
|
|
127
|
+
<a name='L62'></a><a href='#L62'>62</a>
|
|
128
|
+
<a name='L63'></a><a href='#L63'>63</a>
|
|
129
|
+
<a name='L64'></a><a href='#L64'>64</a>
|
|
130
|
+
<a name='L65'></a><a href='#L65'>65</a>
|
|
131
|
+
<a name='L66'></a><a href='#L66'>66</a>
|
|
132
|
+
<a name='L67'></a><a href='#L67'>67</a>
|
|
133
|
+
<a name='L68'></a><a href='#L68'>68</a>
|
|
134
|
+
<a name='L69'></a><a href='#L69'>69</a>
|
|
135
|
+
<a name='L70'></a><a href='#L70'>70</a>
|
|
136
|
+
<a name='L71'></a><a href='#L71'>71</a>
|
|
137
|
+
<a name='L72'></a><a href='#L72'>72</a>
|
|
138
|
+
<a name='L73'></a><a href='#L73'>73</a>
|
|
139
|
+
<a name='L74'></a><a href='#L74'>74</a>
|
|
140
|
+
<a name='L75'></a><a href='#L75'>75</a>
|
|
141
|
+
<a name='L76'></a><a href='#L76'>76</a>
|
|
142
|
+
<a name='L77'></a><a href='#L77'>77</a>
|
|
143
|
+
<a name='L78'></a><a href='#L78'>78</a>
|
|
144
|
+
<a name='L79'></a><a href='#L79'>79</a>
|
|
145
|
+
<a name='L80'></a><a href='#L80'>80</a>
|
|
146
|
+
<a name='L81'></a><a href='#L81'>81</a>
|
|
147
|
+
<a name='L82'></a><a href='#L82'>82</a>
|
|
148
|
+
<a name='L83'></a><a href='#L83'>83</a>
|
|
149
|
+
<a name='L84'></a><a href='#L84'>84</a>
|
|
150
|
+
<a name='L85'></a><a href='#L85'>85</a>
|
|
151
|
+
<a name='L86'></a><a href='#L86'>86</a>
|
|
152
|
+
<a name='L87'></a><a href='#L87'>87</a>
|
|
153
|
+
<a name='L88'></a><a href='#L88'>88</a>
|
|
154
|
+
<a name='L89'></a><a href='#L89'>89</a>
|
|
155
|
+
<a name='L90'></a><a href='#L90'>90</a>
|
|
156
|
+
<a name='L91'></a><a href='#L91'>91</a>
|
|
157
|
+
<a name='L92'></a><a href='#L92'>92</a>
|
|
158
|
+
<a name='L93'></a><a href='#L93'>93</a>
|
|
159
|
+
<a name='L94'></a><a href='#L94'>94</a>
|
|
160
|
+
<a name='L95'></a><a href='#L95'>95</a>
|
|
161
|
+
<a name='L96'></a><a href='#L96'>96</a>
|
|
162
|
+
<a name='L97'></a><a href='#L97'>97</a>
|
|
163
|
+
<a name='L98'></a><a href='#L98'>98</a>
|
|
164
|
+
<a name='L99'></a><a href='#L99'>99</a>
|
|
165
|
+
<a name='L100'></a><a href='#L100'>100</a>
|
|
166
|
+
<a name='L101'></a><a href='#L101'>101</a>
|
|
167
|
+
<a name='L102'></a><a href='#L102'>102</a>
|
|
168
|
+
<a name='L103'></a><a href='#L103'>103</a>
|
|
169
|
+
<a name='L104'></a><a href='#L104'>104</a>
|
|
170
|
+
<a name='L105'></a><a href='#L105'>105</a>
|
|
171
|
+
<a name='L106'></a><a href='#L106'>106</a>
|
|
172
|
+
<a name='L107'></a><a href='#L107'>107</a>
|
|
173
|
+
<a name='L108'></a><a href='#L108'>108</a>
|
|
174
|
+
<a name='L109'></a><a href='#L109'>109</a>
|
|
175
|
+
<a name='L110'></a><a href='#L110'>110</a>
|
|
176
|
+
<a name='L111'></a><a href='#L111'>111</a>
|
|
177
|
+
<a name='L112'></a><a href='#L112'>112</a>
|
|
178
|
+
<a name='L113'></a><a href='#L113'>113</a>
|
|
179
|
+
<a name='L114'></a><a href='#L114'>114</a>
|
|
180
|
+
<a name='L115'></a><a href='#L115'>115</a>
|
|
181
|
+
<a name='L116'></a><a href='#L116'>116</a>
|
|
182
|
+
<a name='L117'></a><a href='#L117'>117</a>
|
|
183
|
+
<a name='L118'></a><a href='#L118'>118</a>
|
|
184
|
+
<a name='L119'></a><a href='#L119'>119</a>
|
|
185
|
+
<a name='L120'></a><a href='#L120'>120</a>
|
|
186
|
+
<a name='L121'></a><a href='#L121'>121</a>
|
|
187
|
+
<a name='L122'></a><a href='#L122'>122</a>
|
|
188
|
+
<a name='L123'></a><a href='#L123'>123</a>
|
|
189
|
+
<a name='L124'></a><a href='#L124'>124</a>
|
|
190
|
+
<a name='L125'></a><a href='#L125'>125</a>
|
|
191
|
+
<a name='L126'></a><a href='#L126'>126</a>
|
|
192
|
+
<a name='L127'></a><a href='#L127'>127</a>
|
|
193
|
+
<a name='L128'></a><a href='#L128'>128</a>
|
|
194
|
+
<a name='L129'></a><a href='#L129'>129</a>
|
|
195
|
+
<a name='L130'></a><a href='#L130'>130</a>
|
|
196
|
+
<a name='L131'></a><a href='#L131'>131</a>
|
|
197
|
+
<a name='L132'></a><a href='#L132'>132</a>
|
|
198
|
+
<a name='L133'></a><a href='#L133'>133</a>
|
|
199
|
+
<a name='L134'></a><a href='#L134'>134</a>
|
|
200
|
+
<a name='L135'></a><a href='#L135'>135</a>
|
|
201
|
+
<a name='L136'></a><a href='#L136'>136</a>
|
|
202
|
+
<a name='L137'></a><a href='#L137'>137</a>
|
|
203
|
+
<a name='L138'></a><a href='#L138'>138</a>
|
|
204
|
+
<a name='L139'></a><a href='#L139'>139</a>
|
|
205
|
+
<a name='L140'></a><a href='#L140'>140</a>
|
|
206
|
+
<a name='L141'></a><a href='#L141'>141</a>
|
|
207
|
+
<a name='L142'></a><a href='#L142'>142</a>
|
|
208
|
+
<a name='L143'></a><a href='#L143'>143</a>
|
|
209
|
+
<a name='L144'></a><a href='#L144'>144</a>
|
|
210
|
+
<a name='L145'></a><a href='#L145'>145</a>
|
|
211
|
+
<a name='L146'></a><a href='#L146'>146</a>
|
|
212
|
+
<a name='L147'></a><a href='#L147'>147</a>
|
|
213
|
+
<a name='L148'></a><a href='#L148'>148</a>
|
|
214
|
+
<a name='L149'></a><a href='#L149'>149</a>
|
|
215
|
+
<a name='L150'></a><a href='#L150'>150</a>
|
|
216
|
+
<a name='L151'></a><a href='#L151'>151</a>
|
|
217
|
+
<a name='L152'></a><a href='#L152'>152</a>
|
|
218
|
+
<a name='L153'></a><a href='#L153'>153</a>
|
|
219
|
+
<a name='L154'></a><a href='#L154'>154</a>
|
|
220
|
+
<a name='L155'></a><a href='#L155'>155</a>
|
|
221
|
+
<a name='L156'></a><a href='#L156'>156</a>
|
|
222
|
+
<a name='L157'></a><a href='#L157'>157</a>
|
|
223
|
+
<a name='L158'></a><a href='#L158'>158</a>
|
|
224
|
+
<a name='L159'></a><a href='#L159'>159</a>
|
|
225
|
+
<a name='L160'></a><a href='#L160'>160</a>
|
|
226
|
+
<a name='L161'></a><a href='#L161'>161</a>
|
|
227
|
+
<a name='L162'></a><a href='#L162'>162</a>
|
|
228
|
+
<a name='L163'></a><a href='#L163'>163</a>
|
|
229
|
+
<a name='L164'></a><a href='#L164'>164</a>
|
|
230
|
+
<a name='L165'></a><a href='#L165'>165</a>
|
|
231
|
+
<a name='L166'></a><a href='#L166'>166</a>
|
|
232
|
+
<a name='L167'></a><a href='#L167'>167</a>
|
|
233
|
+
<a name='L168'></a><a href='#L168'>168</a>
|
|
234
|
+
<a name='L169'></a><a href='#L169'>169</a>
|
|
235
|
+
<a name='L170'></a><a href='#L170'>170</a>
|
|
236
|
+
<a name='L171'></a><a href='#L171'>171</a>
|
|
237
|
+
<a name='L172'></a><a href='#L172'>172</a>
|
|
238
|
+
<a name='L173'></a><a href='#L173'>173</a>
|
|
239
|
+
<a name='L174'></a><a href='#L174'>174</a>
|
|
240
|
+
<a name='L175'></a><a href='#L175'>175</a>
|
|
241
|
+
<a name='L176'></a><a href='#L176'>176</a>
|
|
242
|
+
<a name='L177'></a><a href='#L177'>177</a>
|
|
243
|
+
<a name='L178'></a><a href='#L178'>178</a>
|
|
244
|
+
<a name='L179'></a><a href='#L179'>179</a>
|
|
245
|
+
<a name='L180'></a><a href='#L180'>180</a>
|
|
246
|
+
<a name='L181'></a><a href='#L181'>181</a>
|
|
247
|
+
<a name='L182'></a><a href='#L182'>182</a>
|
|
248
|
+
<a name='L183'></a><a href='#L183'>183</a>
|
|
249
|
+
<a name='L184'></a><a href='#L184'>184</a>
|
|
250
|
+
<a name='L185'></a><a href='#L185'>185</a>
|
|
251
|
+
<a name='L186'></a><a href='#L186'>186</a>
|
|
252
|
+
<a name='L187'></a><a href='#L187'>187</a>
|
|
253
|
+
<a name='L188'></a><a href='#L188'>188</a>
|
|
254
|
+
<a name='L189'></a><a href='#L189'>189</a>
|
|
255
|
+
<a name='L190'></a><a href='#L190'>190</a>
|
|
256
|
+
<a name='L191'></a><a href='#L191'>191</a>
|
|
257
|
+
<a name='L192'></a><a href='#L192'>192</a>
|
|
258
|
+
<a name='L193'></a><a href='#L193'>193</a></td><td class="line-coverage quiet"><span class="cline-any cline-neutral"> </span>
|
|
259
|
+
<span class="cline-any cline-neutral"> </span>
|
|
260
|
+
<span class="cline-any cline-neutral"> </span>
|
|
261
|
+
<span class="cline-any cline-neutral"> </span>
|
|
262
|
+
<span class="cline-any cline-neutral"> </span>
|
|
263
|
+
<span class="cline-any cline-neutral"> </span>
|
|
264
|
+
<span class="cline-any cline-neutral"> </span>
|
|
265
|
+
<span class="cline-any cline-yes">5x</span>
|
|
266
|
+
<span class="cline-any cline-neutral"> </span>
|
|
267
|
+
<span class="cline-any cline-neutral"> </span>
|
|
268
|
+
<span class="cline-any cline-yes">5x</span>
|
|
269
|
+
<span class="cline-any cline-neutral"> </span>
|
|
270
|
+
<span class="cline-any cline-neutral"> </span>
|
|
271
|
+
<span class="cline-any cline-neutral"> </span>
|
|
272
|
+
<span class="cline-any cline-neutral"> </span>
|
|
273
|
+
<span class="cline-any cline-yes">5x</span>
|
|
274
|
+
<span class="cline-any cline-neutral"> </span>
|
|
275
|
+
<span class="cline-any cline-neutral"> </span>
|
|
276
|
+
<span class="cline-any cline-neutral"> </span>
|
|
277
|
+
<span class="cline-any cline-neutral"> </span>
|
|
278
|
+
<span class="cline-any cline-neutral"> </span>
|
|
279
|
+
<span class="cline-any cline-neutral"> </span>
|
|
280
|
+
<span class="cline-any cline-neutral"> </span>
|
|
281
|
+
<span class="cline-any cline-neutral"> </span>
|
|
282
|
+
<span class="cline-any cline-neutral"> </span>
|
|
283
|
+
<span class="cline-any cline-neutral"> </span>
|
|
284
|
+
<span class="cline-any cline-neutral"> </span>
|
|
285
|
+
<span class="cline-any cline-neutral"> </span>
|
|
286
|
+
<span class="cline-any cline-yes">5x</span>
|
|
287
|
+
<span class="cline-any cline-neutral"> </span>
|
|
288
|
+
<span class="cline-any cline-neutral"> </span>
|
|
289
|
+
<span class="cline-any cline-neutral"> </span>
|
|
290
|
+
<span class="cline-any cline-neutral"> </span>
|
|
291
|
+
<span class="cline-any cline-neutral"> </span>
|
|
292
|
+
<span class="cline-any cline-neutral"> </span>
|
|
293
|
+
<span class="cline-any cline-neutral"> </span>
|
|
294
|
+
<span class="cline-any cline-neutral"> </span>
|
|
295
|
+
<span class="cline-any cline-neutral"> </span>
|
|
296
|
+
<span class="cline-any cline-neutral"> </span>
|
|
297
|
+
<span class="cline-any cline-neutral"> </span>
|
|
298
|
+
<span class="cline-any cline-yes">5x</span>
|
|
299
|
+
<span class="cline-any cline-neutral"> </span>
|
|
300
|
+
<span class="cline-any cline-neutral"> </span>
|
|
301
|
+
<span class="cline-any cline-neutral"> </span>
|
|
302
|
+
<span class="cline-any cline-neutral"> </span>
|
|
303
|
+
<span class="cline-any cline-neutral"> </span>
|
|
304
|
+
<span class="cline-any cline-neutral"> </span>
|
|
305
|
+
<span class="cline-any cline-neutral"> </span>
|
|
306
|
+
<span class="cline-any cline-neutral"> </span>
|
|
307
|
+
<span class="cline-any cline-neutral"> </span>
|
|
308
|
+
<span class="cline-any cline-neutral"> </span>
|
|
309
|
+
<span class="cline-any cline-neutral"> </span>
|
|
310
|
+
<span class="cline-any cline-neutral"> </span>
|
|
311
|
+
<span class="cline-any cline-neutral"> </span>
|
|
312
|
+
<span class="cline-any cline-neutral"> </span>
|
|
313
|
+
<span class="cline-any cline-neutral"> </span>
|
|
314
|
+
<span class="cline-any cline-neutral"> </span>
|
|
315
|
+
<span class="cline-any cline-neutral"> </span>
|
|
316
|
+
<span class="cline-any cline-neutral"> </span>
|
|
317
|
+
<span class="cline-any cline-neutral"> </span>
|
|
318
|
+
<span class="cline-any cline-neutral"> </span>
|
|
319
|
+
<span class="cline-any cline-neutral"> </span>
|
|
320
|
+
<span class="cline-any cline-neutral"> </span>
|
|
321
|
+
<span class="cline-any cline-neutral"> </span>
|
|
322
|
+
<span class="cline-any cline-yes">5x</span>
|
|
323
|
+
<span class="cline-any cline-neutral"> </span>
|
|
324
|
+
<span class="cline-any cline-neutral"> </span>
|
|
325
|
+
<span class="cline-any cline-neutral"> </span>
|
|
326
|
+
<span class="cline-any cline-neutral"> </span>
|
|
327
|
+
<span class="cline-any cline-neutral"> </span>
|
|
328
|
+
<span class="cline-any cline-neutral"> </span>
|
|
329
|
+
<span class="cline-any cline-neutral"> </span>
|
|
330
|
+
<span class="cline-any cline-neutral"> </span>
|
|
331
|
+
<span class="cline-any cline-neutral"> </span>
|
|
332
|
+
<span class="cline-any cline-neutral"> </span>
|
|
333
|
+
<span class="cline-any cline-neutral"> </span>
|
|
334
|
+
<span class="cline-any cline-neutral"> </span>
|
|
335
|
+
<span class="cline-any cline-neutral"> </span>
|
|
336
|
+
<span class="cline-any cline-neutral"> </span>
|
|
337
|
+
<span class="cline-any cline-neutral"> </span>
|
|
338
|
+
<span class="cline-any cline-neutral"> </span>
|
|
339
|
+
<span class="cline-any cline-neutral"> </span>
|
|
340
|
+
<span class="cline-any cline-neutral"> </span>
|
|
341
|
+
<span class="cline-any cline-yes">5x</span>
|
|
342
|
+
<span class="cline-any cline-neutral"> </span>
|
|
343
|
+
<span class="cline-any cline-neutral"> </span>
|
|
344
|
+
<span class="cline-any cline-neutral"> </span>
|
|
345
|
+
<span class="cline-any cline-neutral"> </span>
|
|
346
|
+
<span class="cline-any cline-neutral"> </span>
|
|
347
|
+
<span class="cline-any cline-neutral"> </span>
|
|
348
|
+
<span class="cline-any cline-neutral"> </span>
|
|
349
|
+
<span class="cline-any cline-neutral"> </span>
|
|
350
|
+
<span class="cline-any cline-neutral"> </span>
|
|
351
|
+
<span class="cline-any cline-neutral"> </span>
|
|
352
|
+
<span class="cline-any cline-neutral"> </span>
|
|
353
|
+
<span class="cline-any cline-neutral"> </span>
|
|
354
|
+
<span class="cline-any cline-neutral"> </span>
|
|
355
|
+
<span class="cline-any cline-neutral"> </span>
|
|
356
|
+
<span class="cline-any cline-neutral"> </span>
|
|
357
|
+
<span class="cline-any cline-neutral"> </span>
|
|
358
|
+
<span class="cline-any cline-neutral"> </span>
|
|
359
|
+
<span class="cline-any cline-neutral"> </span>
|
|
360
|
+
<span class="cline-any cline-neutral"> </span>
|
|
361
|
+
<span class="cline-any cline-neutral"> </span>
|
|
362
|
+
<span class="cline-any cline-neutral"> </span>
|
|
363
|
+
<span class="cline-any cline-neutral"> </span>
|
|
364
|
+
<span class="cline-any cline-neutral"> </span>
|
|
365
|
+
<span class="cline-any cline-neutral"> </span>
|
|
366
|
+
<span class="cline-any cline-neutral"> </span>
|
|
367
|
+
<span class="cline-any cline-neutral"> </span>
|
|
368
|
+
<span class="cline-any cline-neutral"> </span>
|
|
369
|
+
<span class="cline-any cline-neutral"> </span>
|
|
370
|
+
<span class="cline-any cline-neutral"> </span>
|
|
371
|
+
<span class="cline-any cline-neutral"> </span>
|
|
372
|
+
<span class="cline-any cline-neutral"> </span>
|
|
373
|
+
<span class="cline-any cline-neutral"> </span>
|
|
374
|
+
<span class="cline-any cline-neutral"> </span>
|
|
375
|
+
<span class="cline-any cline-neutral"> </span>
|
|
376
|
+
<span class="cline-any cline-neutral"> </span>
|
|
377
|
+
<span class="cline-any cline-neutral"> </span>
|
|
378
|
+
<span class="cline-any cline-neutral"> </span>
|
|
379
|
+
<span class="cline-any cline-neutral"> </span>
|
|
380
|
+
<span class="cline-any cline-neutral"> </span>
|
|
381
|
+
<span class="cline-any cline-neutral"> </span>
|
|
382
|
+
<span class="cline-any cline-neutral"> </span>
|
|
383
|
+
<span class="cline-any cline-yes">5x</span>
|
|
384
|
+
<span class="cline-any cline-neutral"> </span>
|
|
385
|
+
<span class="cline-any cline-neutral"> </span>
|
|
386
|
+
<span class="cline-any cline-neutral"> </span>
|
|
387
|
+
<span class="cline-any cline-neutral"> </span>
|
|
388
|
+
<span class="cline-any cline-neutral"> </span>
|
|
389
|
+
<span class="cline-any cline-neutral"> </span>
|
|
390
|
+
<span class="cline-any cline-neutral"> </span>
|
|
391
|
+
<span class="cline-any cline-neutral"> </span>
|
|
392
|
+
<span class="cline-any cline-neutral"> </span>
|
|
393
|
+
<span class="cline-any cline-neutral"> </span>
|
|
394
|
+
<span class="cline-any cline-neutral"> </span>
|
|
395
|
+
<span class="cline-any cline-neutral"> </span>
|
|
396
|
+
<span class="cline-any cline-yes">5x</span>
|
|
397
|
+
<span class="cline-any cline-neutral"> </span>
|
|
398
|
+
<span class="cline-any cline-neutral"> </span>
|
|
399
|
+
<span class="cline-any cline-neutral"> </span>
|
|
400
|
+
<span class="cline-any cline-neutral"> </span>
|
|
401
|
+
<span class="cline-any cline-neutral"> </span>
|
|
402
|
+
<span class="cline-any cline-neutral"> </span>
|
|
403
|
+
<span class="cline-any cline-neutral"> </span>
|
|
404
|
+
<span class="cline-any cline-neutral"> </span>
|
|
405
|
+
<span class="cline-any cline-neutral"> </span>
|
|
406
|
+
<span class="cline-any cline-neutral"> </span>
|
|
407
|
+
<span class="cline-any cline-neutral"> </span>
|
|
408
|
+
<span class="cline-any cline-neutral"> </span>
|
|
409
|
+
<span class="cline-any cline-yes">5x</span>
|
|
410
|
+
<span class="cline-any cline-neutral"> </span>
|
|
411
|
+
<span class="cline-any cline-neutral"> </span>
|
|
412
|
+
<span class="cline-any cline-neutral"> </span>
|
|
413
|
+
<span class="cline-any cline-neutral"> </span>
|
|
414
|
+
<span class="cline-any cline-neutral"> </span>
|
|
415
|
+
<span class="cline-any cline-neutral"> </span>
|
|
416
|
+
<span class="cline-any cline-neutral"> </span>
|
|
417
|
+
<span class="cline-any cline-neutral"> </span>
|
|
418
|
+
<span class="cline-any cline-neutral"> </span>
|
|
419
|
+
<span class="cline-any cline-neutral"> </span>
|
|
420
|
+
<span class="cline-any cline-neutral"> </span>
|
|
421
|
+
<span class="cline-any cline-neutral"> </span>
|
|
422
|
+
<span class="cline-any cline-neutral"> </span>
|
|
423
|
+
<span class="cline-any cline-neutral"> </span>
|
|
424
|
+
<span class="cline-any cline-yes">9x</span>
|
|
425
|
+
<span class="cline-any cline-yes">12x</span>
|
|
426
|
+
<span class="cline-any cline-yes">12x</span>
|
|
427
|
+
<span class="cline-any cline-yes">12x</span>
|
|
428
|
+
<span class="cline-any cline-yes">1x</span>
|
|
429
|
+
<span class="cline-any cline-neutral"> </span>
|
|
430
|
+
<span class="cline-any cline-neutral"> </span>
|
|
431
|
+
<span class="cline-any cline-yes">11x</span>
|
|
432
|
+
<span class="cline-any cline-yes">3x</span>
|
|
433
|
+
<span class="cline-any cline-neutral"> </span>
|
|
434
|
+
<span class="cline-any cline-neutral"> </span>
|
|
435
|
+
<span class="cline-any cline-yes">3x</span>
|
|
436
|
+
<span class="cline-any cline-yes">2x</span>
|
|
437
|
+
<span class="cline-any cline-neutral"> </span>
|
|
438
|
+
<span class="cline-any cline-neutral"> </span>
|
|
439
|
+
<span class="cline-any cline-neutral"> </span>
|
|
440
|
+
<span class="cline-any cline-neutral"> </span>
|
|
441
|
+
<span class="cline-any cline-neutral"> </span>
|
|
442
|
+
<span class="cline-any cline-yes">6x</span>
|
|
443
|
+
<span class="cline-any cline-yes">2x</span>
|
|
444
|
+
<span class="cline-any cline-yes">1x</span>
|
|
445
|
+
<span class="cline-any cline-neutral"> </span>
|
|
446
|
+
<span class="cline-any cline-neutral"> </span>
|
|
447
|
+
<span class="cline-any cline-neutral"> </span>
|
|
448
|
+
<span class="cline-any cline-yes">5x</span>
|
|
449
|
+
<span class="cline-any cline-neutral"> </span>
|
|
450
|
+
<span class="cline-any cline-neutral"> </span></td><td class="text"><pre class="prettyprint lang-js">import { z } from 'zod';
|
|
451
|
+
import { RuntimeNameSchema } from '../sdk/runtimes/types.js';
|
|
452
|
+
import { SeveritySchema } from '../types/index.js';
|
|
453
|
+
import type { Finding } from '../types/index.js';
|
|
454
|
+
import type { RuntimeName } from '../sdk/runtimes/types.js';
|
|
455
|
+
|
|
456
|
+
/** Default model for eval skill execution and judging. */
|
|
457
|
+
export const DEFAULT_EVAL_MODEL = 'claude-sonnet-4-6';
|
|
458
|
+
|
|
459
|
+
/** Default runtime for eval skill execution. */
|
|
460
|
+
export const DEFAULT_EVAL_RUNTIME: RuntimeName = 'claude';
|
|
461
|
+
|
|
462
|
+
/**
|
|
463
|
+
* A "should find" assertion in BDD style.
|
|
464
|
+
*/
|
|
465
|
+
export const ShouldFindSchema = z.object({
|
|
466
|
+
/** Natural language description of the expected finding for the LLM judge */
|
|
467
|
+
finding: z.string(),
|
|
468
|
+
/** Expected severity level. When provided, evals require an exact normalized match. */
|
|
469
|
+
severity: SeveritySchema.optional(),
|
|
470
|
+
/** If true (default), eval fails when this is not found */
|
|
471
|
+
required: z.boolean().default(true),
|
|
472
|
+
});
|
|
473
|
+
export type ShouldFind = z.infer<typeof ShouldFindSchema>;
|
|
474
|
+
|
|
475
|
+
/**
|
|
476
|
+
* Maintainer-only provenance for standalone evals. Runners ignore this field.
|
|
477
|
+
*/
|
|
478
|
+
export const EvalScenarioNotesSchema = z.object({
|
|
479
|
+
/** Original PR, issue, benchmark, or other source URL. */
|
|
480
|
+
source: z.string().optional(),
|
|
481
|
+
/** Which source side was captured when scaffolded from a PR. */
|
|
482
|
+
side: z.string().optional(),
|
|
483
|
+
/** Optional copied source notes, such as a PR body. */
|
|
484
|
+
body: z.string().optional(),
|
|
485
|
+
}).passthrough();
|
|
486
|
+
|
|
487
|
+
/**
|
|
488
|
+
* A single eval scenario within a YAML eval file or standalone JSON file.
|
|
489
|
+
*/
|
|
490
|
+
export const EvalScenarioSchema = z.object({
|
|
491
|
+
/** Scenario name (e.g., "null-property-access") */
|
|
492
|
+
name: z.string(),
|
|
493
|
+
/** What this eval tests (BDD "given" / description) */
|
|
494
|
+
given: z.string(),
|
|
495
|
+
/** Fixture files to use, relative to evals/ directory */
|
|
496
|
+
files: z.array(z.string()).min(1),
|
|
497
|
+
/** Model override for this specific scenario */
|
|
498
|
+
model: z.string().optional(),
|
|
499
|
+
/** Runtime override for this specific scenario */
|
|
500
|
+
runtime: RuntimeNameSchema.optional(),
|
|
501
|
+
/** What Warden should find (BDD "then") */
|
|
502
|
+
should_find: z.array(ShouldFindSchema).min(1),
|
|
503
|
+
/** What Warden should NOT report (precision assertions) */
|
|
504
|
+
should_not_find: z.array(z.string()).default([]),
|
|
505
|
+
/** Optional maintainer-only provenance, ignored by eval execution */
|
|
506
|
+
notes: EvalScenarioNotesSchema.optional(),
|
|
507
|
+
});
|
|
508
|
+
export type EvalScenario = z.infer<typeof EvalScenarioSchema>;
|
|
509
|
+
|
|
510
|
+
/**
|
|
511
|
+
* A standalone eval scenario file. The scenario name defaults to the file
|
|
512
|
+
* basename so adding cases does not require repeating it in JSON.
|
|
513
|
+
*/
|
|
514
|
+
export const EvalScenarioFileSchema = EvalScenarioSchema.extend({
|
|
515
|
+
name: z.string().optional(),
|
|
516
|
+
});
|
|
517
|
+
export type EvalScenarioFile = z.infer<typeof EvalScenarioFileSchema>;
|
|
518
|
+
|
|
519
|
+
/**
|
|
520
|
+
* Root schema for a YAML eval file. Each file defines a category of evals
|
|
521
|
+
* sharing a common skill.
|
|
522
|
+
*
|
|
523
|
+
* Example YAML:
|
|
524
|
+
* skill: skills/bug-detection.md
|
|
525
|
+
* evals:
|
|
526
|
+
* - name: null-property-access
|
|
527
|
+
* given: code that accesses .find() result without null checking
|
|
528
|
+
* files: [fixtures/null-property-access/handler.ts]
|
|
529
|
+
* should_find:
|
|
530
|
+
* - finding: null access on user.name
|
|
531
|
+
* severity: high
|
|
532
|
+
*/
|
|
533
|
+
export const EvalFileSchema = z.object({
|
|
534
|
+
/** Skill to run, relative to evals/ directory */
|
|
535
|
+
skill: z.string(),
|
|
536
|
+
/** Default runtime for all evals in this file */
|
|
537
|
+
runtime: RuntimeNameSchema.default(DEFAULT_EVAL_RUNTIME),
|
|
538
|
+
/** Default model for all evals in this file */
|
|
539
|
+
model: z.string().default(DEFAULT_EVAL_MODEL),
|
|
540
|
+
/** List of eval scenarios */
|
|
541
|
+
evals: z.array(EvalScenarioSchema).min(1),
|
|
542
|
+
});
|
|
543
|
+
export type EvalFile = z.infer<typeof EvalFileSchema>;
|
|
544
|
+
|
|
545
|
+
/**
|
|
546
|
+
* Resolved eval metadata ready for execution. Combines the file-level
|
|
547
|
+
* defaults with scenario-level overrides.
|
|
548
|
+
*/
|
|
549
|
+
export interface EvalMeta {
|
|
550
|
+
/** Scenario name (e.g., "null-property-access") */
|
|
551
|
+
name: string;
|
|
552
|
+
/** Category name from the YAML filename (e.g., "eval-bug-detection") */
|
|
553
|
+
category: string;
|
|
554
|
+
/** Resolved skill name from the skill frontmatter or file path */
|
|
555
|
+
skillName: string;
|
|
556
|
+
/** What this eval tests (BDD "given") */
|
|
557
|
+
given: string;
|
|
558
|
+
/** Resolved absolute path to the skill file */
|
|
559
|
+
skillPath: string;
|
|
560
|
+
/** Resolved absolute paths to fixture files */
|
|
561
|
+
filePaths: string[];
|
|
562
|
+
/** Model to use for skill execution */
|
|
563
|
+
model: string;
|
|
564
|
+
/** Runtime to use for skill execution */
|
|
565
|
+
runtime: RuntimeName;
|
|
566
|
+
/** What Warden should find */
|
|
567
|
+
should_find: ShouldFind[];
|
|
568
|
+
/** What Warden should NOT report */
|
|
569
|
+
should_not_find: string[];
|
|
570
|
+
}
|
|
571
|
+
|
|
572
|
+
/**
|
|
573
|
+
* Judge verdict for a single expectation.
|
|
574
|
+
*/
|
|
575
|
+
export const ExpectationVerdictSchema = z.object({
|
|
576
|
+
/** Whether this expectation was met */
|
|
577
|
+
met: z.boolean(),
|
|
578
|
+
/** Which finding matched (by index), or null if none */
|
|
579
|
+
matchedFindingIndex: z.number().int().nonnegative().nullable(),
|
|
580
|
+
/** Brief reasoning from the judge */
|
|
581
|
+
reasoning: z.string(),
|
|
582
|
+
});
|
|
583
|
+
export type ExpectationVerdict = z.infer<typeof ExpectationVerdictSchema>;
|
|
584
|
+
|
|
585
|
+
/**
|
|
586
|
+
* Judge verdict for a single anti-expectation.
|
|
587
|
+
*/
|
|
588
|
+
export const AntiExpectationVerdictSchema = z.object({
|
|
589
|
+
/** Whether a finding violated this anti-expectation (true = violation found) */
|
|
590
|
+
violated: z.boolean(),
|
|
591
|
+
/** Which finding violated (by index), or null if none */
|
|
592
|
+
violatingFindingIndex: z.number().int().nonnegative().nullable(),
|
|
593
|
+
/** Brief reasoning from the judge */
|
|
594
|
+
reasoning: z.string(),
|
|
595
|
+
});
|
|
596
|
+
export type AntiExpectationVerdict = z.infer<typeof AntiExpectationVerdictSchema>;
|
|
597
|
+
|
|
598
|
+
/**
|
|
599
|
+
* Complete judge response for an eval.
|
|
600
|
+
*/
|
|
601
|
+
export const JudgeResponseSchema = z.object({
|
|
602
|
+
expectations: z.array(ExpectationVerdictSchema),
|
|
603
|
+
antiExpectations: z.array(AntiExpectationVerdictSchema),
|
|
604
|
+
});
|
|
605
|
+
export type JudgeResponse = z.infer<typeof JudgeResponseSchema>;
|
|
606
|
+
|
|
607
|
+
/**
|
|
608
|
+
* Determine if an eval passed based on judge response and eval metadata.
|
|
609
|
+
*/
|
|
610
|
+
export function evalPassed(
|
|
611
|
+
meta: EvalMeta,
|
|
612
|
+
judgeResponse: JudgeResponse,
|
|
613
|
+
findings?: Finding[]
|
|
614
|
+
): boolean {
|
|
615
|
+
// Check required should_find assertions are met
|
|
616
|
+
for (let i = 0; i < meta.should_find.length; i++) {
|
|
617
|
+
const assertion = meta.should_find[i];
|
|
618
|
+
const verdict = judgeResponse.expectations[i];
|
|
619
|
+
if (assertion?.required && !verdict?.met) {
|
|
620
|
+
return false;
|
|
621
|
+
}
|
|
622
|
+
|
|
623
|
+
if (assertion?.required && assertion.severity && verdict?.met && findings) {
|
|
624
|
+
const matchedFinding = verdict.matchedFindingIndex === null
|
|
625
|
+
? undefined
|
|
626
|
+
: findings[verdict.matchedFindingIndex];
|
|
627
|
+
if (matchedFinding?.severity !== assertion.severity) {
|
|
628
|
+
return false;
|
|
629
|
+
}
|
|
630
|
+
}
|
|
631
|
+
}
|
|
632
|
+
|
|
633
|
+
// Check no should_not_find assertions are violated
|
|
634
|
+
for (const verdict of judgeResponse.antiExpectations) {
|
|
635
|
+
if (verdict.violated) {
|
|
636
|
+
return false;
|
|
637
|
+
}
|
|
638
|
+
}
|
|
639
|
+
|
|
640
|
+
return true;
|
|
641
|
+
}
|
|
642
|
+
</pre></td></tr></table></pre>
|
|
643
|
+
|
|
644
|
+
<div class='push'></div><!-- for sticky footer -->
|
|
645
|
+
</div><!-- /wrapper -->
|
|
646
|
+
<div class='footer quiet pad2 space-top1 center small'>
|
|
647
|
+
Code coverage generated by
|
|
648
|
+
<a href="https://istanbul.js.org/" target="_blank" rel="noopener noreferrer">istanbul</a>
|
|
649
|
+
at 2026-05-17T04:35:52.381Z
|
|
650
|
+
</div>
|
|
651
|
+
<script src="../../prettify.js"></script>
|
|
652
|
+
<script>
|
|
653
|
+
window.onload = function () {
|
|
654
|
+
prettyPrint();
|
|
655
|
+
};
|
|
656
|
+
</script>
|
|
657
|
+
<script src="../../sorter.js"></script>
|
|
658
|
+
<script src="../../block-navigation.js"></script>
|
|
659
|
+
</body>
|
|
660
|
+
</html>
|
|
661
|
+
|