@percepta/kaizen 0.5.1 → 0.7.0
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/README.md +54 -126
- package/agent/claude-command.md +23 -0
- package/agent/evals.md +41 -0
- package/agent/overview.md +53 -0
- package/agent/variant-builder.md +22 -0
- package/agent/views.md +51 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/BUILD_ID +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/build-manifest.json +22 -22
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/prerender-manifest.json +3 -3
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/routes-manifest.json +42 -10
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/chunks/27.js +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/chunks/516.js +8 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/chunks/913.js +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/middleware-build-manifest.js +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/404.html +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/500.html +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/[system]/benchmarks.html +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/[system]/benchmarks.js.nft.json +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/[system]/data/[[...path]].html +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/[system]/data/[[...path]].js.nft.json +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/[system]/eval.html +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/[system]/eval.js.nft.json +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/[system]/experiments/[[...path]].html +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/[system]/experiments/[[...path]].js.nft.json +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/[system]/ideas.html +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/[system]/ideas.js.nft.json +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/langfuse-action.js +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/langfuse-action.js.nft.json +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/langfuse-dataset-item.js +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/langfuse-dataset-item.js.nft.json +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/langfuse-dataset-mutation.js +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/langfuse-dataset-mutation.js.nft.json +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/langfuse-dataset.js +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/langfuse-dataset.js.nft.json +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/langfuse-datasets.js +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/langfuse-datasets.js.nft.json +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/langfuse-trace.js +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/langfuse-trace.js.nft.json +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/langfuse-traces.js +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/langfuse-traces.js.nft.json +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/linear-ideas.js +2 -2
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/linear-ideas.js.nft.json +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/run-events.js +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/run-events.js.nft.json +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/run-failures.js +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/run-failures.js.nft.json +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/run-traces.js +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/run-traces.js.nft.json +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/runs.js +2 -2
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/runs.js.nft.json +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/systems.js +2 -2
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/systems.js.nft.json +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/trace-renderer.js +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/api/trace-renderer.js.nft.json +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/index.html +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/index.js.nft.json +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages-manifest.json +7 -2
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/static/SCF0o7YxElB9rzWaOohsA/_buildManifest.js +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/static/chunks/253-85c76c34f33c9604.js +8 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/static/chunks/pages/[system]/{benchmarks-ea3ad9fe4e28dd88.js → benchmarks-30a17b7659010b8c.js} +1 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/static/chunks/pages/[system]/data/[[...path]]-e5f4083fe9ffe429.js +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/static/chunks/pages/[system]/eval-160237a604b47416.js +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/static/chunks/pages/[system]/experiments/[[...path]]-91e47a4893093600.js +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/static/chunks/pages/[system]/ideas-96e58e4624952e26.js +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/static/chunks/pages/index-d3306bb6f5d7d235.js +1 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/static/css/cd3873236eb77caa.css +1 -0
- package/dashboard/.next/standalone/packages/kaizen/package.json +6 -3
- package/dashboard/.next/standalone/packages/kaizen/shared/workspace-paths.js +84 -0
- package/dist/commands/create-view.js +58 -0
- package/dist/commands/create-view.js.map +1 -0
- package/dist/commands/guide.js +66 -0
- package/dist/commands/guide.js.map +1 -0
- package/dist/commands/ideas.js +4 -8
- package/dist/commands/ideas.js.map +1 -1
- package/dist/commands/init-system.js +22 -20
- package/dist/commands/init-system.js.map +1 -1
- package/dist/commands/init.js +28 -64
- package/dist/commands/init.js.map +1 -1
- package/dist/commands/log.js +5 -11
- package/dist/commands/log.js.map +1 -1
- package/dist/commands/rebuild.js +7 -9
- package/dist/commands/rebuild.js.map +1 -1
- package/dist/commands/run.js +5 -9
- package/dist/commands/run.js.map +1 -1
- package/dist/commands/studio.js +3 -3
- package/dist/commands/studio.js.map +1 -1
- package/dist/index.js +17 -21
- package/dist/index.js.map +1 -1
- package/dist/lib/cli.js +20 -0
- package/dist/lib/cli.js.map +1 -0
- package/dist/lib/events.js.map +1 -1
- package/dist/lib/fs-utils.js +3 -27
- package/dist/lib/fs-utils.js.map +1 -1
- package/dist/lib/leaderboard.js +1 -1
- package/dist/lib/leaderboard.js.map +1 -1
- package/dist/lib/paths.js +3 -3
- package/dist/lib/paths.js.map +1 -1
- package/dist/lib/promotion.js.map +1 -1
- package/dist/lib/run-dir.js +1 -1
- package/dist/lib/run-dir.js.map +1 -1
- package/dist/lib/runner.js +6 -5
- package/dist/lib/runner.js.map +1 -1
- package/dist/lib/system.js +4 -2
- package/dist/lib/system.js.map +1 -1
- package/dist/package.js +6 -3
- package/dist/shared/view-types.d.ts +67 -0
- package/dist/shared/view-types.d.ts.map +1 -0
- package/dist/shared/workspace-paths.js +84 -0
- package/dist/shared/workspace-paths.js.map +1 -0
- package/dist/types.d.ts +3 -10
- package/dist/types.d.ts.map +1 -1
- package/package.json +6 -3
- package/shared/view-types.d.ts +69 -0
- package/shared/view-types.js +1 -0
- package/shared/workspace-paths.d.ts +19 -0
- package/shared/workspace-paths.js +84 -0
- package/templates/system/eval.py +13 -6
- package/templates/system/eval.ts +11 -5
- package/templates/system/rubric.md +1 -1
- package/templates/system/system.md +6 -5
- package/templates/view/dataset-item.tsx +63 -0
- package/templates/view/trace.tsx +10 -0
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/chunks/424.js +0 -3
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/[system]/data.html +0 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/[system]/data.js.nft.json +0 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/[system]/experiments.html +0 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/server/pages/[system]/experiments.js.nft.json +0 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/static/chunks/374-421036d63d323cc9.js +0 -3
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/static/chunks/pages/[system]/data-57686b9546f2794a.js +0 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/static/chunks/pages/[system]/eval-d9b5f1b8db0f0f90.js +0 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/static/chunks/pages/[system]/experiments-4d2122d6ada9a04a.js +0 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/static/chunks/pages/[system]/ideas-6c1ff7f9e0da750b.js +0 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/static/chunks/pages/index-1556edd8356dd19f.js +0 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/static/css/e75cf1946c214544.css +0 -1
- package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/static/q7hDbHI4NR8DdLDV3kWYh/_buildManifest.js +0 -1
- package/dist/lib/env.js +0 -2
- package/dist/shared/env.js +0 -4
- package/templates/workspace/.claude/agents/variant-builder.md +0 -51
- package/templates/workspace/.claude/commands/kaizen.md +0 -65
- /package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/static/{q7hDbHI4NR8DdLDV3kWYh → SCF0o7YxElB9rzWaOohsA}/_ssgManifest.js +0 -0
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":1,"files":["../../webpack-runtime.js","../../chunks/825.js","../../chunks/916.js","../../chunks/215.js","../../chunks/785.js","../../chunks/946.js","../../chunks/424.js","../../../../../shared/linear-ideas.js","../../../../../package.json","../../../../../../../package.json"]}
|
|
@@ -1,3 +0,0 @@
|
|
|
1
|
-
(self.webpackChunk_N_E=self.webpackChunk_N_E||[]).push([[374],{374:(e,t,a)=>{"use strict";a.d(t,{W:()=>eG});var s=a(144),r=a(4540),l=a(6584),n=a(6048),i=a.n(n),o=a(684),d=a.n(o);let c={code:"Owned by system source code. Agent changes edit the customer repo.",langfuse:"Owned by Langfuse. Agent changes use the Langfuse API.",linear:"Owned by Linear. Agent changes use the Linear API.",local:"Owned by local Kaizen state on this machine."};function u({source:e}){return(0,s.jsx)("span",{className:`${d().sourceChip} ${d()[`sourceChip_${e}`]}`,title:c[e],"aria-label":c[e],children:(0,s.jsx)(m,{source:e})})}function m({source:e}){return"langfuse"===e?(0,s.jsx)("img",{src:"/source-icons/langfuse.svg",alt:"",className:d().sourceLogo}):"linear"===e?(0,s.jsx)("img",{src:"/source-icons/linear.svg",alt:"",className:d().sourceLogo}):"local"===e?(0,s.jsx)(p,{}):(0,s.jsx)(_,{})}function _(){return(0,s.jsxs)("svg",{className:d().sourceSvgIcon,viewBox:"0 0 24 24",fill:"none","aria-hidden":"true",children:[(0,s.jsx)("path",{d:"M8.5 7L3.5 12L8.5 17",stroke:"currentColor",strokeWidth:"2",strokeLinecap:"square",strokeLinejoin:"miter"}),(0,s.jsx)("path",{d:"M15.5 7L20.5 12L15.5 17",stroke:"currentColor",strokeWidth:"2",strokeLinecap:"square",strokeLinejoin:"miter"})]})}function p(){return(0,s.jsxs)("svg",{className:d().sourceSvgIcon,viewBox:"0 0 24 24",fill:"none","aria-hidden":"true",children:[(0,s.jsx)("path",{d:"M5 4H17L20 7V20H5V4Z",stroke:"currentColor",strokeWidth:"2",strokeLinejoin:"miter"}),(0,s.jsx)("path",{d:"M8 4V10H16V4",stroke:"currentColor",strokeWidth:"2"}),(0,s.jsx)("path",{d:"M8 16H16",stroke:"currentColor",strokeWidth:"2"})]})}var h=a(7810),f=a.n(h);function x(){let{systems:e,activeSystemId:t,setActiveSystemId:a}=(0,l.V)(),r=e.find(e=>e.id===t);return(0,s.jsxs)("label",{className:f().wrapper,children:[(0,s.jsx)("span",{className:f().label,children:"System"}),(0,s.jsxs)("select",{className:f().select,value:t??"",onChange:e=>a(e.target.value||null),children:[!t&&(0,s.jsx)("option",{value:"",disabled:!0,children:"Choose"}),t&&!r&&(0,s.jsx)("option",{value:t,children:"Loading system..."}),e.map(e=>(0,s.jsx)("option",{value:e.id,children:e.name},e.id))]}),t&&(0,s.jsx)(u,{source:"code"})]})}let j=[{id:"data",label:"Data"},{id:"benchmarks",label:"Benchmarks"},{id:"ideas",label:"Ideas"},{id:"experiments",label:"Experiments"}];function v({activeSystem:e,activeSurface:t,children:a}){return(0,s.jsxs)("div",{className:d().page,children:[(0,s.jsxs)("header",{className:d().topBar,children:[(0,s.jsx)(i(),{href:"/",className:d().logoLink,"aria-label":"Kaizen home",children:(0,s.jsx)("img",{src:"/logo-cream.svg",alt:"",className:d().topLogo})}),(0,s.jsx)("div",{className:d().systemSlot,children:(0,s.jsx)(x,{})}),(0,s.jsx)("nav",{className:d().surfaceNav,"aria-label":"Kaizen surfaces",children:j.map(a=>e?(0,s.jsx)(i(),{href:`/${e.id}/${a.id}`,className:`${d().surfaceLink} ${t===a.id?d().surfaceLinkActive:""}`,children:a.label},a.id):(0,s.jsx)("span",{className:`${d().surfaceLink} ${d().surfaceLinkDisabled}`,children:a.label},a.id))}),(0,s.jsx)("div",{})]}),(0,s.jsx)("main",{className:d().content,children:a})]})}function g(e){let[t,a]=(0,r.useState)([]),[s,l]=(0,r.useState)(!1);return(0,r.useEffect)(()=>{let t=null,s=null,r=null,n=(t=[])=>{a(e?t.filter(t=>t.system===e):t)},i=()=>{fetch("/api/runs").then(e=>e.json()).then(e=>n(e.runs??[])).catch(()=>{})},o=()=>{(t=new EventSource("/api/runs")).onmessage=e=>{n(JSON.parse(e.data).runs??[]),l(!0),s&&(clearInterval(s),s=null)},t.onerror=()=>{l(!1),t?.close(),t=null,s||(s=setInterval(i,5e3),i()),r||(r=setTimeout(()=>{r=null,o()},1e4))}};return o(),()=>{t?.close(),s&&clearInterval(s),r&&clearTimeout(r)}},[e]),{runs:t,connected:s}}function N(e){if(!e)return"";let t=new Date(e);return Number.isNaN(t.getTime())?e:new Intl.DateTimeFormat(void 0,{year:"numeric",month:"short",day:"numeric",hour:"numeric",minute:"2-digit"}).format(t)}var b=a(2390);b.Ik({runId:b.Yj(),runName:b.Yj(),status:b.k5(["running","complete","crashed","aborted"]),worktreeBranch:b.Yj(),parentId:b.Yj().nullable(),system:b.Yj(),progress:b.Ik({completed:b.ai(),total:b.ai(),lastHeartbeat:b.Yj()}).optional(),metrics:b.g1(b.Yj(),b.ai()),totalItems:b.ai().nullable(),langfuseRunId:b.Yj().optional(),linearIssue:b.Ik({id:b.Yj(),url:b.Yj()}).optional(),evalConfig:b.Ik({dataset:b.Yj(),evalVersion:b.ai().optional(),datasetItemCount:b.ai(),judge:b.Ik({rubric:b.Yj(),rubricHash:b.Yj(),model:b.Yj(),temperature:b.ai()}).optional()}).optional(),startedAt:b.Yj(),updatedAt:b.Yj()});let y={actionAccuracy:{label:"Action Accuracy",format:"percent"},parameterExtraction:{label:"Param Extraction",format:"percent"},clarificationAppropriateness:{label:"Clarification",format:"percent"},multiActionDetection:{label:"Multi-Action",format:"percent"},f1:{label:"F1",format:"percent"},f2:{label:"F2",format:"percent"},precision:{label:"Precision",format:"percent"},recall:{label:"Recall",format:"percent"},cost_per_item:{label:"Cost / Item",format:"raw"},avg_latency_s:{label:"Avg Latency (s)",format:"raw"},true_positives:{label:"TP",format:"integer"},false_positives:{label:"FP",format:"integer"},false_negatives:{label:"FN",format:"integer"},total_clis_scored:{label:"CLIs Scored",format:"integer"},items_with_metrics:{label:"Items w/ Metrics",format:"integer"},timeouts:{label:"Timeouts",format:"integer"},judge_quality:{label:"Judge Quality",format:"percent"},judge_quality_chartNotes:{label:"Quality: Chart Notes",format:"percent"},judge_quality_followUp:{label:"Quality: Follow-Up",format:"percent"}};function S(e){return y[e]?.label??e.replace(/[_-]+/g," ").replace(/([A-Z])/g," $1").replace(/^./,e=>e.toUpperCase()).trim()}function R(e,t){let a=y[e];return a?.format==="integer"?Math.round(t).toLocaleString():a?.format==="raw"?t.toFixed(2):a?.format==="percent"?(100*t).toFixed(1)+"%":t>1?Math.round(t).toLocaleString():(100*t).toFixed(1)+"%"}function k(e,t){if(!t)return null;if(t in e.metrics)return e.metrics[t];if("f2"===t&&"precision"in e.metrics&&"recall"in e.metrics){var a,s;return(a=e.metrics.precision)+(s=e.metrics.recall)===0?0:5*a*s/(4*a+s)}return null}function I(e,t){var a,s;let r=e.evalConfig?.dataset??"unknown dataset",l="number"==typeof e.evalConfig?.evalVersion?String(e.evalConfig.evalVersion):"unknown",n=(a=e,s=t,a.evalConfig?.judge?.rubric??s??"unknown eval"),i=`Eval v${l} \xb7 ${S(n)}`;return{key:`${r}::${l}::${n}`,dataset:r,datasetLabel:L(r),evalVersion:l,evalMetric:n,evalLabel:i}}function L(e){return e.replace(/--/g," ").replace(/[-_]+/g," ").replace(/\bgt\b/gi,"Ground Truth").replace(/\bv(\d+)\b/gi,"v$1").replace(/\b\w/g,e=>e.toUpperCase()).trim()}function w({title:e,children:t,source:a,syncLabel:r,onRefresh:l,refreshing:n}){return(0,s.jsxs)("section",{className:d().surfaceBanner,children:[(0,s.jsxs)("div",{className:d().surfaceBannerText,children:[(0,s.jsxs)("div",{className:d().surfaceBannerTitle,children:[(0,s.jsx)("span",{children:e}),a&&(0,s.jsx)(u,{source:a})]}),(0,s.jsx)("div",{className:d().surfaceBannerCopy,children:t})]}),(null!=r||null!=l)&&(0,s.jsx)(T,{label:r,onRefresh:l,refreshing:n})]})}function T({label:e,onRefresh:t,refreshing:a}){return(0,s.jsxs)("div",{className:d().syncStatus,children:[null!=e&&(0,s.jsx)("span",{className:d().syncStatusLabel,children:e}),t&&(0,s.jsx)("button",{className:d().iconButton,onClick:t,title:"Refresh","aria-label":"Refresh","aria-busy":a?"true":"false",children:(0,s.jsx)("span",{"aria-hidden":"true",children:"↻"})})]})}function C({system:e}){var t;let{runs:a}=g(e.id),l=e.primaryMetric??null,n=(0,r.useMemo)(()=>(function(e,t){let a=new Map;for(let i of e){var s,r,l,n;let e=I(i,t),o=a.get(e.key);o?(o.count++,s=o.itemCount,r=i.totalItems,o.itemCount=null===s?r:null===r?s:Math.max(s,r),l=o.latestAt,n=i.updatedAt,o.latestAt=l?Date.parse(n)>Date.parse(l)?n:l:n):a.set(e.key,{...e,count:1,itemCount:i.totalItems,latestAt:i.updatedAt})}return[...a.values()].sort((e,t)=>(t.latestAt??"").localeCompare(e.latestAt??""))})(a,l),[l,a]);return(0,s.jsxs)("div",{className:d().surface,children:[(0,s.jsx)(w,{title:"Benchmarks",source:"code",syncLabel:"Live updates",children:"A benchmark is the fixed dataset plus scoring setup for this system. The current benchmark comes from the system file, and the rows below show which benchmarks have actually been used by local experiments."}),(0,s.jsxs)("section",{className:d().detailPanel,children:[(0,s.jsxs)("div",{className:d().detailHeader,children:[(0,s.jsxs)("div",{children:[(0,s.jsx)("h2",{className:d().detailTitle,children:"Current benchmark"}),(0,s.jsx)("div",{className:d().detailMeta,children:"This is what the next Kaizen run will use unless a coding agent changes the system file."})]}),(0,s.jsx)(u,{source:"code"})]}),(0,s.jsxs)("div",{className:d().evalDefinitionGrid,children:[(0,s.jsx)(P,{label:"Dataset",value:e.evalDataset?L(e.evalDataset):"Not configured",detail:e.evalDataset}),(0,s.jsx)(P,{label:"Eval version",value:null===e.evalVersion?"Not configured":`v${e.evalVersion}`}),(0,s.jsx)(P,{label:"Primary metric",value:e.primaryMetric?S(e.primaryMetric):"Not configured",detail:e.primaryMetric}),(0,s.jsx)(P,{label:"Style",value:"ground_truth"===(t=e.evalType)?"Ground truth":"llm_judge"===t?"LLM judge":"human"===t?"Human review":"policy"===t?"Policy":"Not configured"})]})]}),(0,s.jsxs)("section",{className:d().evalHistorySection,children:[(0,s.jsxs)("div",{className:d().evalSectionHeader,children:[(0,s.jsxs)("div",{children:[(0,s.jsx)("h2",{className:d().detailTitle,children:"Benchmarks used in experiments"}),(0,s.jsx)("div",{className:d().detailMeta,children:"Each row is a separate comparison group."})]}),(0,s.jsx)(u,{source:"local"})]}),0===n.length?(0,s.jsx)("div",{className:d().emptyPanel,children:"No benchmarks yet. Run an experiment to create one."}):(0,s.jsxs)("div",{className:d().evalSummaryTable,children:[(0,s.jsxs)("div",{className:d().evalSummaryHeader,children:[(0,s.jsx)("span",{children:"Dataset"}),(0,s.jsx)("span",{children:"Scoring"}),(0,s.jsx)("span",{children:"Runs"}),(0,s.jsx)("span",{children:"Items"}),(0,s.jsx)("span",{children:"Latest"})]}),n.map(e=>(0,s.jsxs)("div",{className:d().evalSummaryRow,children:[(0,s.jsxs)("div",{children:[(0,s.jsx)("span",{className:d().evalSummaryPrimary,children:e.datasetLabel}),(0,s.jsx)("span",{className:d().evalSummaryMeta,children:e.dataset})]}),(0,s.jsxs)("div",{children:[(0,s.jsx)("span",{className:d().evalSummaryPrimary,children:e.evalLabel}),(0,s.jsx)("span",{className:d().evalSummaryMeta,children:S(e.evalMetric)})]}),(0,s.jsx)("span",{children:e.count}),(0,s.jsx)("span",{children:e.itemCount??"?"}),(0,s.jsx)("span",{children:e.latestAt?N(e.latestAt):"-"})]},e.key))]})]})]})}function P({label:e,value:t,detail:a}){return(0,s.jsxs)("div",{className:d().evalDefinitionCell,children:[(0,s.jsx)("span",{className:d().metaLabel,children:e}),(0,s.jsx)("span",{className:d().metaValue,children:t}),a&&(0,s.jsx)("code",{className:d().evalDefinitionCode,children:a})]})}function M({status:e,currentValues:t}){let a=[e.requiredEnvVars?.length?{title:"Required",values:e.requiredEnvVars}:null,e.missingEnvVars?.length?{title:"Missing",values:e.missingEnvVars}:null,e.optionalEnvVars?.length?{title:"Optional",values:e.optionalEnvVars}:null,e.expectedEnvFile?{title:"Put values here",values:[e.expectedEnvFile]}:null,t?.length?{title:"Current query",values:t}:null].filter(E);return(0,s.jsxs)("div",{className:d().setupBanner,children:[(0,s.jsxs)("div",{className:d().setupHeader,children:[(0,s.jsxs)("div",{children:[(0,s.jsx)("div",{className:d().setupTitle,children:e.message}),(0,s.jsx)("div",{className:d().setupCopy,children:e.remediation})]}),(0,s.jsx)("span",{className:d().setupState,children:e.state})]}),a.length>0&&(0,s.jsx)("div",{className:d().setupGrid,children:a.map(e=>(0,s.jsx)(A,{title:e.title,values:e.values},`${e.title}:${e.values.join(",")}`))}),e.lookup&&e.lookup.length>0&&(0,s.jsxs)("div",{className:d().setupLookup,children:[(0,s.jsx)("span",{className:d().setupLookupLabel,children:"Kaizen checks"}),e.lookup.join(" -> ")]}),e.detail&&(0,s.jsx)("div",{className:d().setupDetail,children:e.detail})]})}function A({title:e,values:t}){return(0,s.jsxs)("div",{className:d().setupBlock,children:[(0,s.jsx)("span",{className:d().setupBlockTitle,children:e}),t.map(e=>(0,s.jsx)("code",{className:d().setupCode,children:e},e))]})}function E(e){return null!==e}function B(e=new Date){return new Intl.DateTimeFormat(void 0,{hour:"numeric",minute:"2-digit"}).format(e)}function $(e,t){var a;let s=(a=t)%6e4==0?`${a/6e4}m`:a%1e3==0?`${a/1e3}s`:`${a}ms`;return e?`Updated ${e} \xb7 Every ${s}`:`Every ${s}`}function D({system:e}){let[t,a]=(0,r.useState)("datasets"),[l,n]=(0,r.useState)([]),[i,o]=(0,r.useState)({name:null,systemId:null}),[c,u]=(0,r.useState)([]),[m,_]=(0,r.useState)(null),[p,h]=(0,r.useState)(null),[f,x]=(0,r.useState)(!1),[j,v]=(0,r.useState)(null),[g,N]=(0,r.useState)(!1),[b,y]=(0,r.useState)(!1),[S,R]=(0,r.useState)(!1),[k,I]=(0,r.useState)(null),[L,T]=(0,r.useState)(null),[C,P]=(0,r.useState)(null),[A,E]=(0,r.useState)(null),[D,O]=(0,r.useState)(!1),[z,G]=(0,r.useState)(null),K=e.evalDataset,Y=(0,r.useRef)(!1),Z=(0,r.useRef)(e.id),X=(0,r.useRef)(K),ee=(0,r.useRef)(async()=>{}),et=(0,r.useRef)(i);Z.current=e.id,X.current=K,et.current=i;let es=i.name,el=i.systemId,ei=(0,r.useCallback)(()=>{a("datasets")},[]),eo=(0,r.useCallback)(()=>{a("traces")},[]),ed=(0,r.useCallback)(()=>{a("add")},[]),ec=(0,r.useCallback)(()=>{ee.current()},[]),eu=(0,r.useCallback)(e=>{o({name:e,systemId:e?Z.current:null}),u([]),_(null),h(null),P(null),E(null)},[]),em=(0,r.useCallback)(async()=>{let t=e.id;x(!0);try{var a,s,r,l;let e=await fetch(`/api/langfuse-datasets?systemId=${encodeURIComponent(t)}`),i=(a=await e.json(),ea(a)?{data:Array.isArray(a.data)?a.data.map(J).filter(er):void 0,connection:Q(a.connection),error:en(a.error)}:{});if(Z.current===t&&T(i.connection??null),!e.ok)throw Error(i.error??`HTTP ${e.status}`);if(Z.current!==t)return null;let d=i.data??[],c=et.current,m=(s=c.systemId===t?c.name:null,r=d,l=X.current,s&&r.some(e=>e.name===s)?s:l&&r.some(e=>e.name===l)?l:r[0]?.name??null);return n(d),v(null),G(B()),o({name:m,systemId:m?t:null}),0===d.length&&(u([]),_(null),h(null)),m}catch(e){if(Z.current!==t)return null;return Y.current||(n([]),o({name:null,systemId:null}),u([]),_(null),h(null)),v(e instanceof Error?e.message:String(e)),null}finally{Z.current===t&&(Y.current=!0,y(!0),x(!1))}},[e.id]),e_=(0,r.useCallback)(async(t=es)=>{if(!t)return null;let a=e.id;N(!0);try{var s;let e=await fetch(`/api/langfuse-dataset?dataset=${encodeURIComponent(t)}&systemId=${encodeURIComponent(a)}`),r=(s=await e.json(),ea(s)?{data:Array.isArray(s.data)?s.data.map(W).filter(er):void 0,connection:Q(s.connection),error:en(s.error)}:{});if(Z.current===a&&T(r.connection??null),!e.ok)throw Error(r.error??`HTTP ${e.status}`);if(Z.current!==a)return null;let l=r.data??[],n=l.filter(q);u(l),I(null),_(e=>e&&n.some(t=>t.id===e)?e:n[0]?.id??null);let i=n.map(U).find(Boolean);return h(e=>e&&n.some(t=>U(t)===e)?e:i??null),G(B()),l}catch(e){if(Z.current!==a)return null;return I(e instanceof Error?e.message:String(e)),null}finally{Z.current===a&&(R(!0),N(!1))}},[e.id,es]),ep=(0,r.useCallback)(async(e,t,a)=>{let s=et.current.name;if(!s)return P("Choose a dataset first."),!1;O(!0),P(null),E(null);try{var r;let l=await fetch("/api/langfuse-dataset-item",{method:"POST",headers:{"Content-Type":"application/json"},body:JSON.stringify({...e,datasetName:s,systemId:Z.current})}),n=(r=await l.json(),ea(r)?{data:W(r.data)??void 0,error:en(r.error)}:{});if(!l.ok)throw Error(n.error??`HTTP ${l.status}`);let i=await e_(s),o=n.data?.id??null;return a&&o&&i?.some(e=>e.id===o&&q(e))&&_(o),E(t),!0}catch(e){return P(e instanceof Error?e.message:String(e)),!1}finally{O(!1)}},[e_]),eh=(0,r.useCallback)(async e=>{await ep({action:"add-from-trace",source:e.source,expectedOutput:e.expectedOutput,notes:e.notes},"Dataset item added.",!0)&&a("datasets")},[ep]),ef=(0,r.useCallback)(async e=>{await ep({action:"update",itemId:e.itemId,expectedOutput:e.expectedOutput,notes:e.notes},"Dataset item saved.",!0)},[ep]),ex=(0,r.useCallback)(async e=>{await ep({action:"archive",itemId:e},"Dataset item archived.",!1)},[ep]);(0,r.useEffect)(()=>{Y.current=!1,n([]),y(!1),R(!1),o({name:null,systemId:null}),u([]),_(null),h(null),v(null),I(null),T(null),G(null),em();let e=setInterval(()=>void em(),6e4),t=()=>void em();return window.addEventListener("focus",t),()=>{clearInterval(e),window.removeEventListener("focus",t)}},[K,em,e.id]),(0,r.useEffect)(()=>{if(R(!1),!es||el!==e.id)return;e_(es);let t=setInterval(()=>void e_(es),15e3),a=()=>void e_(es);return window.addEventListener("focus",a),()=>{clearInterval(t),window.removeEventListener("focus",a)}},[e_,e.id,es,el]);let ej=(0,r.useMemo)(()=>c.filter(q),[c]),ev=(0,r.useMemo)(()=>ej.find(e=>e.id===m)??ej[0]??null,[ej,m]),eg=(0,r.useMemo)(()=>{let e=new Map;for(let t of ej){let a=U(t);a&&!e.has(a)&&e.set(a,t)}return[...e.entries()].map(([e,t])=>({traceId:e,item:t}))},[ej]),eN=p??eg[0]?.traceId??U(ev);return ee.current=(0,r.useCallback)(async()=>{let t=await em();t&&Z.current===e.id&&await e_(t)},[em,e_,e.id]),(0,s.jsxs)("div",{className:d().dataLayout,children:[(0,s.jsxs)("aside",{className:d().dataNav,children:[(0,s.jsx)("button",{className:`${d().dataNavItem} ${"datasets"===t?d().dataNavItemActive:""}`,onClick:ei,children:"Datasets"}),(0,s.jsx)("button",{className:`${d().dataNavItem} ${"traces"===t?d().dataNavItemActive:""}`,onClick:eo,children:"Traces"}),(0,s.jsx)("button",{className:`${d().dataNavItem} ${"add"===t?d().dataNavItemActive:""}`,onClick:ed,children:"Add item"})]}),(0,s.jsxs)("section",{className:d().dataMain,children:[(0,s.jsx)(w,{title:"Langfuse data",source:"langfuse",syncLabel:$(z,15e3),onRefresh:ec,refreshing:g||f,children:"This is the evaluation data for the selected system. Use Datasets to inspect golden examples, Traces to review the source calls behind those examples, and Add item to prepare new examples. The view updates automatically while open; Refresh pulls the latest now."}),L&&"connected"!==L.state&&(0,s.jsx)(M,{status:L}),"datasets"===t?(0,s.jsx)(H,{system:e,datasets:l,selectedDatasetName:es,configuredDatasetName:K,items:ej,selectedItem:ev,loading:f&&!b||g&&!S,datasetsError:j,error:k,mutationError:C,mutationMessage:A,mutating:D,onSelectDataset:eu,onSelect:_,onUpdateItem:ef,onArchiveItem:ex}):"traces"===t?(0,s.jsx)(F,{systemId:e.id,linkedTraces:eg,selectedTraceId:eN,onSelectTrace:h,error:j??k}):(0,s.jsx)(V,{system:e,datasetName:es,error:C,message:A,mutating:D,onAddItem:eh})]})]})}function H({system:e,datasets:t,selectedDatasetName:a,configuredDatasetName:l,items:n,selectedItem:i,loading:o,datasetsError:c,error:m,mutationError:_,mutationMessage:p,mutating:h,onSelectDataset:f,onSelect:x,onUpdateItem:j,onArchiveItem:v}){let g=(0,r.useCallback)(e=>{f(e.target.value?e.target.value:null)},[f]);return(0,s.jsxs)("div",{className:d().surface,children:[(0,s.jsx)("div",{className:d().surfaceToolbar,children:(0,s.jsxs)("div",{children:[(0,s.jsxs)("label",{className:d().inlineSelector,children:[(0,s.jsx)("span",{className:d().inlineSelectorLabel,children:"Dataset"}),(0,s.jsxs)("select",{className:d().select,value:a??"",onChange:g,disabled:0===t.length,children:[(0,s.jsx)("option",{value:"",disabled:!0,children:0===t.length?"No datasets available":"Select dataset"}),t.map(e=>(0,s.jsx)("option",{value:e.name,children:e.name},e.name))]})]}),l&&(0,s.jsxs)("div",{className:d().provenanceLine,children:[(0,s.jsx)("span",{children:"Default dataset"}),(0,s.jsx)("code",{children:l}),(0,s.jsx)(u,{source:"code"})]})]})}),c&&(0,s.jsx)("div",{className:d().errorPanel,children:c}),_&&(0,s.jsx)("div",{className:d().errorPanel,children:_}),p&&(0,s.jsx)("div",{className:d().successPanel,children:p}),o?(0,s.jsx)("div",{className:d().loadingPanel,children:"Loading Langfuse data..."}):a?m?(0,s.jsx)("div",{className:d().errorPanel,children:m}):(0,s.jsxs)("div",{className:d().splitSurface,children:[(0,s.jsx)(O,{items:n,selectedId:i?.id??null,onSelect:x}),(0,s.jsx)(G,{item:i,systemId:e.id,datasetName:a,mutating:h,onUpdate:j,onArchive:v})]}):(0,s.jsxs)("div",{className:d().emptyPanel,children:[(0,s.jsx)("div",{className:d().emptyPanelTitle,children:0===t.length?"No Langfuse datasets found":"No Langfuse dataset selected"}),(0,s.jsx)("div",{className:d().emptyPanelCopy,children:0===t.length?"Kaizen did not find any datasets for this system.":"Choose a dataset to inspect its examples."})]})]})}function F({systemId:e,linkedTraces:t,selectedTraceId:a,onSelectTrace:r,error:l}){return(0,s.jsxs)("div",{className:d().surface,children:[(0,s.jsx)("div",{className:d().surfaceToolbar,children:(0,s.jsxs)("div",{children:[(0,s.jsx)("div",{className:d().surfaceTitle,children:"Traces"}),(0,s.jsxs)("div",{className:d().surfaceSubtitle,children:[t.length," linked source trace",1===t.length?"":"s"]})]})}),l?(0,s.jsx)("div",{className:d().errorPanel,children:l}):(0,s.jsxs)("div",{className:d().splitSurface,children:[(0,s.jsx)(z,{traces:t,selectedTraceId:a,onSelect:r}),(0,s.jsx)(K,{traceId:a,systemId:e})]})]})}function V({system:e,datasetName:t,error:a,message:l,mutating:n,onAddItem:i}){let[o,c]=(0,r.useState)(""),[u,m]=(0,r.useState)(""),[_,p]=(0,r.useState)(""),h=(0,r.useCallback)(e=>{c(e.target.value)},[]),f=(0,r.useCallback)(e=>{m(e.target.value)},[]),x=(0,r.useCallback)(e=>{p(e.target.value)},[]),j=(0,r.useCallback)(e=>{e.preventDefault(),i({source:o,expectedOutput:Z(u),notes:_})},[u,_,i,o]),v=!!(t&&o.trim()&&!n);return(0,s.jsxs)("form",{className:d().detailPanel,onSubmit:j,children:[(0,s.jsxs)("div",{className:d().detailHeader,children:[(0,s.jsxs)("div",{children:[(0,s.jsx)("h2",{className:d().detailTitle,children:"Add dataset item"}),(0,s.jsx)("div",{className:d().detailMeta,children:t??e.evalDataset??"No dataset selected"})]}),(0,s.jsx)("span",{className:d().statusPill,children:"Langfuse"})]}),a&&(0,s.jsx)("div",{className:d().errorPanel,children:a}),l&&(0,s.jsx)("div",{className:d().successPanel,children:l}),(0,s.jsxs)("div",{className:d().formRow,children:[(0,s.jsx)("label",{className:d().formLabel,htmlFor:"dataset-source",children:"Source trace"}),(0,s.jsx)("input",{id:"dataset-source",className:d().textInput,disabled:!t||n,placeholder:"Langfuse trace URL or trace ID",value:o,onChange:h}),(0,s.jsx)("div",{className:d().detailMeta,children:"Orbit URL lookup will be wired in a follow-up. For now, paste the Langfuse trace URL or trace ID for the call you want in the golden dataset."})]}),(0,s.jsxs)("div",{className:d().formRow,children:[(0,s.jsx)("label",{className:d().formLabel,htmlFor:"dataset-expected-output",children:"Expected output"}),(0,s.jsx)("textarea",{id:"dataset-expected-output",className:d().textArea,disabled:!t||n,placeholder:"Leave blank to use the trace output, or paste text/JSON ground truth.",value:u,onChange:f})]}),(0,s.jsxs)("div",{className:d().formRow,children:[(0,s.jsx)("label",{className:d().formLabel,htmlFor:"dataset-notes",children:"Notes"}),(0,s.jsx)("textarea",{id:"dataset-notes",className:d().textArea,disabled:!t||n,placeholder:"Why this belongs in the golden dataset.",value:_,onChange:x})]}),(0,s.jsx)("button",{className:d().secondaryButton,disabled:!v,children:n?"Adding...":"Add item"})]})}function O({items:e,selectedId:t,onSelect:a}){let l=(0,r.useCallback)(e=>{let t=ee(e.target,"itemId");t&&a(t)},[a]);return 0===e.length?(0,s.jsx)("div",{className:d().listPanelEmpty,children:"No dataset items"}):(0,s.jsx)("aside",{className:d().listPanel,onClick:l,children:e.map((e,a)=>{let r=U(e);return(0,s.jsxs)("button",{"data-item-id":e.id,className:`${d().listRow} ${t===e.id?d().listRowSelected:""}`,children:[(0,s.jsx)("span",{className:d().listRowTitle,children:X(e,a)}),(0,s.jsx)("span",{className:d().listRowMeta,children:r?`trace ${r.slice(0,8)}`:e.status??"item"})]},e.id)})})}function z({traces:e,selectedTraceId:t,onSelect:a}){let l=(0,r.useCallback)(e=>{let t=ee(e.target,"traceId");t&&a(t)},[a]);return 0===e.length?(0,s.jsx)("div",{className:d().listPanelEmpty,children:"No linked traces"}):(0,s.jsx)("aside",{className:d().listPanel,onClick:l,children:e.map(({traceId:e,item:a},r)=>(0,s.jsxs)("button",{"data-trace-id":e,className:`${d().listRow} ${t===e?d().listRowSelected:""}`,children:[(0,s.jsx)("span",{className:d().listRowTitle,children:X(a,r)}),(0,s.jsx)("span",{className:d().listRowMeta,children:e})]},e))})}function G({item:e,systemId:t,datasetName:a,mutating:l,onUpdate:n,onArchive:i}){let o=e?U(e):null,c=e?.id??null,[u,m]=(0,r.useState)(""),[_,p]=(0,r.useState)("");(0,r.useEffect)(()=>{var t;m(et(e?.expectedOutput)),p((t=e?.metadata)?ea(t.kaizen)&&"string"==typeof t.kaizen.notes?t.kaizen.notes:"string"==typeof t.notes?t.notes:"":"")},[c]);let h=(0,r.useCallback)(e=>{m(e.target.value)},[]),f=(0,r.useCallback)(e=>{p(e.target.value)},[]),x=(0,r.useCallback)(()=>{e&&n({itemId:e.id,expectedOutput:Z(u),notes:_})},[u,e,_,n]),j=(0,r.useCallback)(()=>{e&&i(e.id)},[e,i]);if(!e)return(0,s.jsx)("section",{className:d().detailPanel,children:"No item selected"});let v=!l,g=!!(a&&!l);return(0,s.jsxs)("section",{className:d().detailPanel,children:[(0,s.jsxs)("div",{className:d().detailHeader,children:[(0,s.jsxs)("div",{children:[(0,s.jsx)("h2",{className:d().detailTitle,children:X(e,0)}),(0,s.jsx)("div",{className:d().detailMeta,children:e.id})]}),e.status&&(0,s.jsx)("span",{className:d().statusPill,children:e.status})]}),(0,s.jsxs)("div",{className:d().formRow,children:[(0,s.jsx)("label",{className:d().formLabel,htmlFor:"dataset-item-notes",children:"Notes"}),(0,s.jsx)("textarea",{id:"dataset-item-notes",className:d().textArea,disabled:l,placeholder:"Add context for future experiments.",value:_,onChange:f})]}),(0,s.jsxs)("div",{className:d().formRow,children:[(0,s.jsx)("label",{className:d().formLabel,htmlFor:"dataset-item-expected",children:"Expected output"}),(0,s.jsx)("textarea",{id:"dataset-item-expected",className:d().textAreaLarge,disabled:l,value:u,onChange:h})]}),(0,s.jsxs)("div",{className:d().actionFooter,children:[(0,s.jsx)("div",{className:d().detailMeta,children:"Archived examples are hidden from the active dataset view but remain in Langfuse history."}),(0,s.jsxs)("div",{className:d().actionRow,children:[(0,s.jsx)("button",{className:d().dangerButton,disabled:!v,onClick:j,type:"button",children:"Archive item"}),(0,s.jsx)("button",{className:d().primaryButton,disabled:!g,onClick:x,type:"button",children:l?"Saving...":"Save changes"})]})]}),(0,s.jsxs)("div",{className:d().detailGrid,children:[(0,s.jsx)(Y,{title:"Input",value:e.input}),(0,s.jsx)(Y,{title:"Current expected output",value:e.expectedOutput})]}),(0,s.jsx)(Y,{title:"Metadata",value:e.metadata}),(0,s.jsxs)("div",{className:d().traceSection,children:[(0,s.jsx)("div",{className:d().sectionTitle,children:"Source Trace"}),(0,s.jsx)(K,{traceId:o,systemId:t,compact:!0})]})]})}function K({traceId:e,systemId:t,compact:a}){let[l,n]=(0,r.useState)(null),[i,o]=(0,r.useState)(null);(0,r.useEffect)(()=>{if(n(null),o(null),!e)return;let a=!1;return fetch(`/api/langfuse-trace?traceId=${encodeURIComponent(e)}&systemId=${encodeURIComponent(t)}`).then(async e=>{var t;let s=ea(t=await e.json())?{id:en(t.id),name:en(t.name),tags:Array.isArray(t.tags)?t.tags.filter(es):void 0,timestamp:en(t.timestamp),metadata:t.metadata,input:t.input,output:t.output,error:en(t.error)}:{};if(!e.ok)throw Error(s.error??`HTTP ${e.status}`);a||n(s)}).catch(e=>{a||o(e instanceof Error?e.message:String(e))}),()=>{a=!0}},[t,e]);let c=e?i?(0,s.jsx)("div",{className:d().errorPanel,children:i}):l?(0,s.jsxs)(s.Fragment,{children:[(0,s.jsxs)("div",{className:d().traceHeader,children:[(0,s.jsx)("span",{className:d().traceName,children:l.name??l.id}),l.timestamp&&(0,s.jsx)("span",{className:d().detailMeta,children:l.timestamp})]}),l.tags&&l.tags.length>0&&(0,s.jsx)("div",{className:d().tagRow,children:l.tags.map(e=>(0,s.jsx)("span",{className:d().tagPill,children:e},e))}),(0,s.jsx)(Y,{title:"Trace metadata",value:l.metadata}),(0,s.jsxs)("div",{className:d().detailGrid,children:[(0,s.jsx)(Y,{title:"Trace input",value:l.input}),(0,s.jsx)(Y,{title:"Trace output",value:l.output})]})]}):(0,s.jsx)("div",{className:d().loadingPanel,children:"Loading trace..."}):(0,s.jsx)("div",{className:d().mutedText,children:"No source trace linked."});return a?c:(0,s.jsx)("section",{className:d().detailPanel,children:c})}function Y({title:e,value:t}){return(0,s.jsxs)("div",{className:d().jsonBlock,children:[(0,s.jsx)("div",{className:d().jsonTitle,children:e}),(0,s.jsx)("pre",{className:d().jsonPre,children:et(t)})]})}function U(e){if(!e)return null;if("string"==typeof e.sourceTraceId&&e.sourceTraceId)return e.sourceTraceId;let t=e.metadata,a=t?.sourceTraceId??t?.traceId;return"string"==typeof a&&a?a:null}function q(e){return e.status?.toUpperCase()!=="ARCHIVED"}function Z(e){let t=e.trim();if(t&&"(empty)"!==t)try{return JSON.parse(t)}catch{return t}}function Q(e){if(ea(e)&&"string"==typeof e.state&&"string"==typeof e.message&&"string"==typeof e.remediation)return{state:e.state,message:e.message,remediation:e.remediation,requiredEnvVars:el(e.requiredEnvVars),optionalEnvVars:el(e.optionalEnvVars),expectedEnvFile:en(e.expectedEnvFile),missingEnvVars:el(e.missingEnvVars),lookup:el(e.lookup),detail:en(e.detail)}}function J(e){return ea(e)&&"string"==typeof e.name?{id:en(e.id),name:e.name,description:"string"==typeof e.description||null===e.description?e.description:void 0,metadata:ea(e.metadata)||null===e.metadata?e.metadata:void 0,createdAt:en(e.createdAt),updatedAt:en(e.updatedAt)}:null}function W(e){return ea(e)&&"string"==typeof e.id?{id:e.id,status:en(e.status),sourceTraceId:"string"==typeof e.sourceTraceId||null===e.sourceTraceId?e.sourceTraceId:void 0,input:e.input,expectedOutput:e.expectedOutput,metadata:ea(e.metadata)||null===e.metadata?e.metadata:void 0,createdAt:en(e.createdAt),updatedAt:en(e.updatedAt)}:null}function X(e,t){let a=e.metadata;for(let t of[a?.callSummary,a?.summary,a?.name,a?.analysisId,e.id])if("string"==typeof t&&t.trim())return t;return`Item ${t+1}`}function ee(e,t){if(!(e instanceof Element))return null;let a=e.closest("button");return a instanceof HTMLButtonElement?a.dataset[t]??null:null}function et(e){return null==e?"(empty)":"string"==typeof e?e:JSON.stringify(e,null,2)}function ea(e){return"object"==typeof e&&null!==e&&!Array.isArray(e)}function es(e){return"string"==typeof e}function er(e){return null!==e}function el(e){return Array.isArray(e)?e.filter(es):[]}function en(e){return"string"==typeof e?e:void 0}function ei(e){return ec(e)?{ideas:Array.isArray(e.ideas)?e.ideas.map(eo).filter(em):void 0,config:function(e){if(ec(e))return{teamKey:ep(e.teamKey),projectName:ep(e.projectName),projectId:ep(e.projectId),projectUrl:ep(e.projectUrl),projectRef:ep(e.projectRef),projectRefKind:ep(e.projectRefKind),label:e_(e.label)??"Kaizen"}}(e.config),connection:function(e){if(ec(e)&&function(e){switch(e){case"connected":case"missing_project":case"missing_api_key":case"project_not_found":case"auth_failed":case"query_failed":case"network_error":return!0;default:return!1}}(e.state)&&"string"==typeof e.message&&"string"==typeof e.remediation)return{state:e.state,message:e.message,remediation:e.remediation,requiredEnvVars:ed(e.requiredEnvVars),optionalEnvVars:ed(e.optionalEnvVars),expectedEnvFile:e_(e.expectedEnvFile),missingEnvVars:ed(e.missingEnvVars),lookup:ed(e.lookup),detail:e_(e.detail)}}(e.connection),error:e_(e.error)}:{}}function eo(e){var t,a,s,r,l;return ec(e)&&"string"==typeof e.id&&"string"==typeof e.identifier&&"string"==typeof e.title&&"string"==typeof e.url&&"string"==typeof e.createdAt&&"string"==typeof e.updatedAt?{id:e.id,identifier:e.identifier,title:e.title,description:"string"==typeof e.description||null===e.description?e.description:null,url:e.url,priority:"number"==typeof e.priority?e.priority:0,createdAt:e.createdAt,updatedAt:e.updatedAt,state:ec(t=e.state)&&"string"==typeof t.name&&"string"==typeof t.type?{name:t.name,type:t.type}:null,assignee:ec(a=e.assignee)&&"string"==typeof a.name?{name:a.name}:null,project:ec(s=e.project)&&"string"==typeof s.id&&"string"==typeof s.name?{id:s.id,name:s.name,url:"string"==typeof s.url?s.url:null}:null,team:ec(r=e.team)&&"string"==typeof r.key&&"string"==typeof r.name?{key:r.key,name:r.name}:null,labels:ec(l=e.labels)&&Array.isArray(l.nodes)?{nodes:l.nodes.map(e=>ec(e)&&"string"==typeof e.name?{name:e.name}:null).filter(em)}:{nodes:[]}}:null}function ed(e){return Array.isArray(e)?e.filter(eu):[]}function ec(e){return"object"==typeof e&&null!==e&&!Array.isArray(e)}function eu(e){return"string"==typeof e}function em(e){return null!==e}function e_(e){return"string"==typeof e?e:void 0}function ep(e){return"string"==typeof e?e:null}Object.keys(y);var eh=a(9212),ef=a.n(eh);function ex({issue:e,issueTitle:t,showTitle:a=!1}){if(!e)return null;let r=t?`${e.id} \xb7 ${t}`:e.id;return(0,s.jsxs)("a",{className:ef().linearIssueLink,href:e.url,target:"_blank",rel:"noreferrer",title:t?`${e.id}: ${t}`:`Open ${e.id} in Linear`,onClick:ej,children:[(0,s.jsx)("img",{src:"/source-icons/linear.svg",alt:""}),(0,s.jsx)("span",{children:a?r:e.id})]})}function ej(e){e.stopPropagation()}let ev={running:{color:"#f5a623",bg:"rgba(245,166,35,0.1)",border:"rgba(245,166,35,0.3)"},complete:{color:"#00d4aa",bg:"rgba(0,212,170,0.1)",border:"rgba(0,212,170,0.3)"},crashed:{color:"#e8553d",bg:"rgba(232,85,61,0.1)",border:"rgba(232,85,61,0.3)"},aborted:{color:"rgba(250,250,250,0.4)",bg:"rgba(255,255,255,0.04)",border:"rgba(255,255,255,0.08)"}},eg=ev.running;function eN({metrics:e}){let t=Object.entries(e);return 0===t.length?null:(0,s.jsxs)("table",{className:ef().metricsTable,children:[(0,s.jsx)("thead",{children:(0,s.jsxs)("tr",{children:[(0,s.jsx)("th",{className:ef().metricsTableHeader,children:"Metric"}),(0,s.jsx)("th",{className:ef().metricsTableHeader,style:{textAlign:"right"},children:"Value"}),(0,s.jsx)("th",{className:ef().metricsTableHeader,style:{width:"40%"}})]})}),(0,s.jsx)("tbody",{children:t.map(([e,t])=>(0,s.jsxs)("tr",{className:ef().metricsTableRow,children:[(0,s.jsx)("td",{className:ef().metricsTableLabel,children:S(e)}),(0,s.jsx)("td",{className:ef().metricsTableValue,children:R(e,t)}),(0,s.jsx)("td",{className:ef().metricsTableBarCell,children:t>=0&&t<=1&&(0,s.jsx)("div",{className:ef().metricsTableBarTrack,children:(0,s.jsx)("div",{className:ef().metricsTableBar,style:{width:`${100*t}%`}})})})]},e))})]})}function eb({run:e,primaryMetric:t,linearIssueTitle:a}){let r=ev[e.status]??eg,l=k(e,t),n=function(e){let t=e.progress?.total||e.totalItems||e.evalConfig?.datasetItemCount||null;if(!t||t<=0)return null;let a="complete"===e.status?t:"number"==typeof e.progress?.completed?Math.min(e.progress.completed,t):null;return null===a?null:{completed:a,total:t,percent:Math.min(100,Math.max(0,a/t*100))}}(e);return(0,s.jsxs)("div",{children:[(0,s.jsxs)("div",{className:ef().detailPanelHeader,children:[(0,s.jsxs)("div",{className:ef().detailTitleStack,children:[(0,s.jsx)("h2",{className:ef().detailPanelTitle,children:e.runName}),e.linearIssue&&(0,s.jsx)(ex,{issue:e.linearIssue,issueTitle:a,showTitle:!0})]}),null!=l&&(0,s.jsx)("span",{className:ef().bestAccuracyInline,children:R(t??"score",l)}),(0,s.jsx)("span",{className:ef().statusBadge,style:{color:r.color,backgroundColor:r.bg,border:`1px solid ${r.border}`},children:e.status})]}),n&&(0,s.jsxs)("div",{className:ef().runProgressBlock,children:[(0,s.jsxs)("div",{className:ef().runProgressHeader,children:[(0,s.jsx)("span",{children:"Dataset items"}),(0,s.jsxs)("span",{children:[n.completed,"/",n.total]})]}),(0,s.jsx)("div",{className:ef().runProgressTrack,children:(0,s.jsx)("div",{className:ef().runProgressFill,style:{width:`${n.percent}%`}})})]}),(0,s.jsxs)("div",{className:ef().runMetaGrid,children:[(0,s.jsx)(ey,{label:"Dataset",value:e.evalConfig?.dataset}),(0,s.jsx)(ey,{label:"Eval version",value:"number"==typeof e.evalConfig?.evalVersion?`v${e.evalConfig.evalVersion}`:null}),(0,s.jsx)(ey,{label:"Started",value:N(e.startedAt)}),(0,s.jsx)(ey,{label:"Updated",value:N(e.updatedAt)}),e.worktreeBranch&&(0,s.jsx)(ey,{label:"Branch",value:e.worktreeBranch}),e.parentId&&(0,s.jsx)(ey,{label:"Parent",value:e.parentId})]}),(0,s.jsxs)("div",{className:ef().runSection,children:[(0,s.jsx)("span",{className:ef().runLabel,children:"Metrics"}),(0,s.jsx)(eN,{metrics:e.metrics}),0===Object.keys(e.metrics).length&&(0,s.jsx)("div",{className:ef().detailEmpty,children:"No metrics yet."})]})]})}function ey({label:e,value:t}){return t?(0,s.jsxs)("div",{className:ef().runMetaCell,children:[(0,s.jsx)("span",{className:ef().runMetaLabel,children:e}),(0,s.jsx)("span",{className:ef().runMetaValue,children:t})]}):null}function eS({run:e,primaryMetric:t,linearIssueTitle:a,onClose:r}){return(0,s.jsxs)(s.Fragment,{children:[(0,s.jsx)("div",{className:ef().detailPanelBackdrop,onClick:r}),(0,s.jsxs)("div",{className:ef().detailPanel,children:[(0,s.jsx)("button",{className:ef().detailPanelClose,onClick:r,children:"\xd7"}),(0,s.jsx)(eb,{run:e,primaryMetric:t,linearIssueTitle:a})]})]})}var eR=a(6971),ek=a(8463),eI=a(7784),eL=a(4456);function ew(e){return e<60?`${e}s`:e<3600?`${Math.floor(e/60)}m`:`${Math.floor(e/3600)}h ${Math.floor(e%3600/60)}m`}function eT({node:e,isSelected:t,isBestGlobal:a,isHighlighted:l,primaryMetric:n,linearIssueTitles:i,onClick:o}){let d=e.data.status,c=ev[d.status]??eg,u=k(d,n),m=null!=u?`${(100*u).toFixed(1)}%`:null,_="running"===d.status,p=function(e,t){let[a,s]=(0,r.useState)(Date.now());if((0,r.useEffect)(()=>{if(!t)return;let e=setInterval(()=>s(Date.now()),1e3);return()=>clearInterval(e)},[t]),!e)return null;let l=new Date(e).getTime();return isNaN(l)?null:ew(Math.max(0,Math.floor((a-l)/1e3)))}(d.startedAt,_),h=_?null:function(e,t){if(!e||!t)return null;let a=new Date(e).getTime(),s=new Date(t).getTime();if(isNaN(a)||isNaN(s))return null;let r=Math.max(0,Math.floor((s-a)/1e3));return 0===r?null:ew(r)}(d.startedAt,d.updatedAt),f=d.linearIssue?i?.[d.linearIssue.id]??null:null;return(0,s.jsx)("foreignObject",{x:e.x-100,y:e.y-63,width:200,height:126,children:(0,s.jsxs)("button",{"data-run-id":d.runId,className:[ef().treeNodeCard,t?ef().treeNodeSelected:"",a?ef().treeNodeBest:"",l?ef().treeNodeHighlight:""].filter(Boolean).join(" "),onClick:o,title:f?`${d.runName}
|
|
2
|
-
${f}`:d.runName,children:[(0,s.jsx)("div",{className:ef().treeNodeName,children:d.runName}),(0,s.jsxs)("div",{className:ef().treeNodeBadgeRow,children:[(0,s.jsx)("span",{className:ef().treeNodeBadge,style:{color:c.color,backgroundColor:c.bg,border:`1px solid ${c.border}`},children:d.status}),_&&p&&(0,s.jsx)("span",{className:ef().treeNodeElapsed,children:p}),h&&(0,s.jsx)("span",{className:ef().treeNodeElapsed,children:h})]}),(0,s.jsx)("div",{className:ef().treeNodeScoreRow,children:m?(0,s.jsxs)(s.Fragment,{children:[(0,s.jsx)("span",{className:ef().treeNodeAccuracy,children:m}),null!=d.totalItems&&(0,s.jsxs)("span",{className:ef().treeNodeItems,children:["n=",d.totalItems]})]}):"aborted"===d.status||"crashed"===d.status?(0,s.jsx)("span",{className:ef().treeNodeItems,children:"—"}):(0,s.jsxs)("span",{className:ef().treeNodeLoading,children:[(0,s.jsx)("span",{className:ef().treeNodePulse}),(0,s.jsx)("span",{className:ef().treeNodeLoadingText,children:"running"===d.status&&d.progress?`${d.progress.completed}/${d.progress.total}`:"running"===d.status?"Running eval...":"Pending"})]})}),d.linearIssue&&(0,s.jsx)("div",{className:ef().treeNodeIssueRow,children:(0,s.jsx)(ex,{issue:d.linearIssue,issueTitle:f,showTitle:!0})})]})})}function eC({link:e}){return(0,s.jsx)(eL.A,{data:e,stroke:"rgba(255, 255, 255, 0.25)",strokeWidth:1.5,fill:"none"})}function eP(e){return{status:e.status,children:e.children.map(eP)}}function eM(e){return e.depth>0&&null!==e.data.status}function eA({roots:e,selectedId:t,bestGlobalId:a,highlightId:l,primaryMetric:n,linearIssueTitles:i,onSelect:o}){let d=(0,r.useMemo)(()=>({status:null,children:e.map(eP)}),[e]),c=(0,r.useMemo)(()=>(0,ek.Ay)(d,e=>e.children.length>0?e.children:null),[d]);return 0===e.length?null:(0,s.jsx)("div",{className:ef().treeForest,children:(0,s.jsx)(eI.A,{root:c,nodeSize:[216,158],separation:(e,t)=>e.parent===t.parent?1:1.1,children:e=>{let r=e.descendants(),d=e.links(),c=r.filter(eM),u=d.filter(e=>e.source.depth>0);if(0===c.length)return null;let m=1/0,_=-1/0,p=1/0,h=-1/0;for(let e of c)m=Math.min(m,e.x),_=Math.max(_,e.x),p=Math.min(p,e.y),h=Math.max(h,e.y);let f=_-m+200+80,x=h-p+126+80,j=-m+100+40,v=-p+63+40;return(0,s.jsx)("svg",{width:f,height:x,className:ef().treeSvg,children:(0,s.jsxs)(eR.A,{top:v,left:j,children:[u.map((e,t)=>(0,s.jsx)(eC,{link:e},`link-${t}`)),c.map(e=>(0,s.jsx)(eT,{node:e,isSelected:t===e.data.status.runId,isBestGlobal:a===e.data.status.runId,isHighlighted:l===e.data.status.runId,primaryMetric:n,linearIssueTitles:i,onClick:()=>o(e.data.status.runId)},e.data.status.runId))]})})}})})}var eE=a(9965),eB=a.n(eE);function e$({runs:e,primaryMetric:t,bestGlobalId:a,selectedId:l,linearIssueTitles:n,onSelect:i}){let[o,d]=(0,r.useState)(t??""),[c,u]=(0,r.useState)(!0),m=(0,r.useRef)(new Map),_=(0,r.useRef)(new Map),p=(0,r.useMemo)(()=>{let a=new Set;for(let t of e)for(let e of Object.keys(t.metrics))a.add(e);return[...a].sort((e,a)=>e===t?-1:a===t?1:e.localeCompare(a))},[e,t]);r.useEffect(()=>{t&&(d(t),u(!0))},[t]);let h=(0,r.useMemo)(()=>{let t=[...e];return t.sort((e,t)=>{let a=o?e.metrics[o]??-1/0:0,s=o?t.metrics[o]??-1/0:0,r=c?s-a:a-s;return 0!==r?r:(t.totalItems??0)-(e.totalItems??0)}),t},[e,o,c]),f=(0,r.useCallback)(e=>{for(let[e,t]of m.current)_.current.set(e,t.getBoundingClientRect().top);e===o?u(e=>!e):(d(e),u(!0))},[o]);return((0,r.useLayoutEffect)(()=>{if(0!==_.current.size){for(let[e,t]of m.current){let a=_.current.get(e);if(void 0===a)continue;let s=a-t.getBoundingClientRect().top;1>Math.abs(s)||(t.style.transform=`translateY(${s}px)`,t.style.transition="none",requestAnimationFrame(()=>{requestAnimationFrame(()=>{t.style.transition="transform 0.5s cubic-bezier(0.25, 1, 0.5, 1)",t.style.transform=""})}))}_.current.clear()}},[h]),0===e.length)?(0,s.jsx)("p",{className:eB().noData,children:"No experiments yet."}):(0,s.jsx)("div",{className:eB().tableWrapper,children:(0,s.jsxs)("table",{className:eB().table,children:[(0,s.jsx)("thead",{children:(0,s.jsxs)("tr",{children:[(0,s.jsx)("th",{className:eB().thRank,children:"#"}),(0,s.jsx)("th",{className:eB().thName,children:"Run"}),(0,s.jsx)("th",{className:eB().thStatus,children:"Status"}),p.map(e=>(0,s.jsxs)("th",{className:`${eB().thMetric} ${e===t?eB().thPrimary:""} ${e===o?eB().thSorted:""}`,onClick:()=>f(e),children:[(0,s.jsx)("span",{className:eB().thLabel,children:S(e)}),e===o&&(0,s.jsx)("span",{className:eB().sortArrow,children:c?"▼":"▲"})]},e)),(0,s.jsx)("th",{className:eB().thItems,children:"n"}),(0,s.jsx)("th",{className:eB().thStarted,children:"Started"})]})}),(0,s.jsx)("tbody",{children:h.map((e,r)=>{let o=ev[e.status]??eg,d=e.runId===a,c=e.runId===l;return(0,s.jsxs)("tr",{ref:t=>{t?m.current.set(e.runId,t):m.current.delete(e.runId)},className:`${eB().row} ${d?eB().rowBest:""} ${c?eB().rowSelected:""}`,onClick:()=>i(e.runId),children:[(0,s.jsx)("td",{className:eB().tdRank,children:r+1}),(0,s.jsxs)("td",{className:eB().tdName,children:[(0,s.jsx)("span",{className:eB().expName,children:e.runName}),(0,s.jsx)(ex,{issue:e.linearIssue,issueTitle:e.linearIssue?n?.[e.linearIssue.id]:null}),d&&(0,s.jsx)("span",{className:eB().bestBadge,children:"BEST"})]}),(0,s.jsx)("td",{className:eB().tdStatus,children:(0,s.jsx)("span",{className:eB().statusDot,style:{backgroundColor:o.color},title:e.status})}),p.map(a=>{let r=e.metrics[a],l=a===t;return(0,s.jsx)("td",{className:`${eB().tdMetric} ${l?eB().tdPrimary:""}`,children:null!=r?(0,s.jsx)("span",{className:eB().metricValue,children:R(a,r)}):(0,s.jsx)("span",{className:eB().metricEmpty,children:"—"})},a)}),(0,s.jsx)("td",{className:eB().tdItems,children:e.totalItems??"?"}),(0,s.jsx)("td",{className:eB().tdStarted,children:N(e.startedAt)})]},e.runId)})})]})})}function eD({runs:e,primaryMetric:t,bestRunId:a,selectedRunId:r,linearIssueTitles:l,onSelectRun:n}){return(0,s.jsx)(e$,{runs:e,primaryMetric:t,bestGlobalId:a,selectedId:r,linearIssueTitles:l,onSelect:n})}function eH({system:e}){let{runs:t}=g(e.id),[a,l]=(0,r.useState)(null),[n]=(0,r.useState)(null),[i,o]=(0,r.useState)(null),[c,u]=(0,r.useState)(280),[m,_]=(0,r.useState)({});(0,r.useEffect)(()=>{l(null),o(null),_({})},[e.id]),(0,r.useEffect)(()=>{let t=!1;return fetch(`/api/linear-ideas?systemId=${encodeURIComponent(e.id)}`).then(e=>e.json()).then(e=>{t||_(function(e){let t={};for(let a of e)t[a.identifier]=a.title;return t}(ei(e).ideas??[]))}).catch(()=>{t||_({})}),()=>{t=!0}},[e.id]);let p=(0,r.useCallback)(()=>{l(null)},[]),h=(0,r.useCallback)(e=>{let t=function(e){if(!(e instanceof Element))return null;let t=e.closest("button");return t instanceof HTMLButtonElement?t.dataset.groupKey??null:null}(e.target);t&&o(t)},[]),f=(0,r.useCallback)(e=>{e.preventDefault();let t=e.clientX,a=e=>{u(Math.min(560,Math.max(220,c+e.clientX-t)))};document.body.style.cursor="col-resize",document.body.style.userSelect="none",document.addEventListener("pointermove",a),document.addEventListener("pointerup",()=>{document.removeEventListener("pointermove",a),document.body.style.cursor="",document.body.style.userSelect=""},{once:!0})},[c]),x=e.primaryMetric??null,j=(0,r.useMemo)(()=>{let e=new Map;for(let a of t){let t=I(a,x),s=e.get(t.key);s?s.count++:e.set(t.key,{...t,count:1})}return[...e.values()].sort((e,t)=>{let a=e.dataset.localeCompare(t.dataset);return 0===a?e.evalLabel.localeCompare(t.evalLabel):a})},[x,t]),v=i&&j.some(e=>e.key===i)?i:j[0]?.key??null,N=(0,r.useMemo)(()=>v?t.filter(e=>I(e,x).key===v):t,[t,v,x]);(0,r.useEffect)(()=>{a&&!N.some(e=>e.runId===a)&&l(null)},[N,a]);let b=(0,r.useMemo)(()=>(function(e,t){let a=new Map;for(let t of e)a.set(t.runId,{status:t,children:[]});let s=[];for(let e of a.values()){let t=e.status.parentId;t&&a.has(t)?a.get(t).children.push(e):s.push(e)}let r=e=>{for(let a of(e.sort((e,a)=>(k(a.status,t)??-1)-(k(e.status,t)??-1)),e))r(a.children)};return r(s),s})(N,x),[N,x]),y=(0,r.useMemo)(()=>{let e=null,t=-1;for(let a of N){let s=k(a,x);null!=s&&s>t&&(t=s,e=a.runId)}return e},[N,x]),S=a?N.find(e=>e.runId===a):null;return(0,s.jsxs)("div",{className:d().experimentsLayout,children:[(0,s.jsx)("aside",{className:`${d().dataNav} ${d().experimentNav}`,style:{width:c},onClick:h,children:0===j.length?(0,s.jsx)("div",{className:d().dataNavEmpty,children:"No experiments yet"}):j.map(e=>(0,s.jsxs)("button",{"data-group-key":e.key,title:`${e.datasetLabel}
|
|
3
|
-
${e.evalLabel} \xb7 ${e.count} experiments`,className:`${d().dataNavItem} ${d().dataNavGroupItem} ${d().experimentNavItem} ${v===e.key?d().dataNavItemActive:""}`,children:[(0,s.jsx)("span",{className:d().dataNavItemText,children:e.datasetLabel}),(0,s.jsxs)("span",{className:d().dataNavItemMeta,children:[e.evalLabel," \xb7 ",e.count]})]},e.key))}),(0,s.jsx)("div",{className:d().sidebarResizeHandle,role:"separator","aria-label":"Resize benchmarks","aria-orientation":"vertical",onPointerDown:f}),(0,s.jsxs)("section",{className:`${d().dataMain} ${d().experimentsMain}`,children:[(0,s.jsx)(w,{title:"Local experiments",source:"local",syncLabel:"Live updates",children:"Choose a benchmark on the left. Kaizen only compares experiments inside the selected benchmark, because scores from different datasets or scoring setups are not scientifically comparable."}),(0,s.jsx)(eA,{roots:b,selectedId:a,bestGlobalId:y,highlightId:n,primaryMetric:x,linearIssueTitles:m,onSelect:l}),S&&(0,s.jsx)(eS,{run:S,primaryMetric:x,linearIssueTitle:S.linearIssue?m[S.linearIssue.id]:null,onClose:p}),(0,s.jsx)(eD,{runs:N,primaryMetric:x,bestRunId:y,selectedRunId:a,linearIssueTitles:m,onSelectRun:l})]})]})}function eF({system:e}){let[t,a]=(0,r.useState)([]),[l,n]=(0,r.useState)(null),[i,o]=(0,r.useState)(void 0),[c,m]=(0,r.useState)(null),[_,p]=(0,r.useState)(null),[h,f]=(0,r.useState)(!1),[x,j]=(0,r.useState)(!1),[v,g]=(0,r.useState)(null),N=(0,r.useRef)(e.id);N.current=e.id;let b=(0,r.useCallback)(async()=>{let t=e.id;f(!0);try{let e=await fetch(`/api/linear-ideas?systemId=${encodeURIComponent(t)}`),s=ei(await e.json());if(!e.ok)throw Error(s.error??`HTTP ${e.status}`);if(N.current!==t)return;a(s.ideas??[]),o(s.config),m(s.connection??null),n(e=>e&&(s.ideas??[]).some(t=>t.id===e)?e:s.ideas?.[0]?.id??null),p(s.error??null),g(B())}catch(e){if(N.current!==t)return;m(null),p(e instanceof Error?e.message:String(e))}finally{N.current===t&&(j(!0),f(!1))}},[e.id]),y=(0,r.useCallback)(()=>{b()},[b]);(0,r.useEffect)(()=>{j(!1),a([]),n(null),o(void 0),m(null),p(null),g(null),b();let e=setInterval(()=>void b(),2e4),t=()=>void b();return window.addEventListener("focus",t),()=>{clearInterval(e),window.removeEventListener("focus",t)}},[b,e.id]);let S=(0,r.useMemo)(()=>t.find(e=>e.id===l)??t[0]??null,[t,l]),R=i?.teamKey??null,k=i?.projectRef??null,I=i?.label??"Kaizen",L=(0,r.useMemo)(()=>{var e;return[(e={teamKey:R,projectRef:k,label:I}).teamKey?`LINEAR_TEAM_KEY=${e.teamKey}`:"any team",e.projectRef?`linear_project=${e.projectRef}`:"linear_project not set",`label=${e.label}`]},[I,k,R]);return(0,s.jsxs)("div",{className:d().surface,children:[(0,s.jsxs)(w,{title:"Linear ideas",source:"linear",syncLabel:$(v,2e4),onRefresh:y,refreshing:h,children:["This shows Linear issues that Kaizen can try as experiments. For this system, Kaizen looks",i?.projectName||i?.projectRef?(0,s.jsxs)(s.Fragment,{children:[" in project ",(0,s.jsxs)("span",{className:d().inlineSourceValue,children:[(0,s.jsx)("code",{children:i.projectName??i.projectRef}),(0,s.jsx)(u,{source:"code"})]})]}):" in the Linear project configured in this system's code"," ","for issues labeled ",(0,s.jsx)("code",{children:"Kaizen"}),". Label a Linear issue"," ",(0,s.jsx)("code",{children:"Kaizen"})," when you want Kaizen to try it as an experiment."]}),c&&"connected"!==c.state&&(0,s.jsx)(M,{status:c,currentValues:L}),_&&(0,s.jsx)("div",{className:d().errorPanel,children:_}),h&&!x?(0,s.jsx)("div",{className:d().loadingPanel,children:"Loading Linear ideas..."}):(0,s.jsxs)("div",{className:d().splitSurface,children:[(0,s.jsx)(eV,{ideas:t,selectedId:S?.id??null,onSelect:n}),(0,s.jsx)(eO,{idea:S})]})]})}function eV({ideas:e,selectedId:t,onSelect:a}){let l=(0,r.useCallback)(e=>{let t=function(e){if(!(e instanceof Element))return null;let t=e.closest("button");return t instanceof HTMLButtonElement?t.dataset.ideaId??null:null}(e.target);t&&a(t)},[a]);return 0===e.length?(0,s.jsx)("div",{className:d().listPanelEmpty,children:"No Linear ideas"}):(0,s.jsx)("aside",{className:d().listPanel,onClick:l,children:e.map(e=>(0,s.jsxs)("button",{"data-idea-id":e.id,className:`${d().listRow} ${t===e.id?d().listRowSelected:""}`,children:[(0,s.jsx)("span",{className:d().listRowTitle,children:e.title}),(0,s.jsxs)("span",{className:d().listRowMeta,children:[e.identifier," \xb7 ",e.state?.name??"No status"]})]},e.id))})}function eO({idea:e}){var t;let a;return e?(0,s.jsxs)("section",{className:d().detailPanel,children:[(0,s.jsxs)("div",{className:d().detailHeader,children:[(0,s.jsxs)("div",{children:[(0,s.jsx)("h2",{className:d().detailTitle,children:e.title}),(0,s.jsx)("a",{className:d().detailLink,href:e.url,target:"_blank",rel:"noreferrer",children:e.identifier})]}),e.state&&(0,s.jsx)("span",{className:d().statusPill,children:e.state.name})]}),(0,s.jsxs)("div",{className:d().ideaMetaGrid,children:[(0,s.jsx)(ez,{label:"Team",value:e.team?.key??e.team?.name}),(0,s.jsx)(ez,{label:"Project",value:e.project?.name}),(0,s.jsx)(ez,{label:"Assignee",value:e.assignee?.name}),(0,s.jsx)(ez,{label:"Updated",value:Number.isNaN((a=new Date(t=e.updatedAt)).getTime())?t:a.toLocaleString()})]}),e.labels.nodes.length>0&&(0,s.jsx)("div",{className:d().tagRow,children:e.labels.nodes.map(e=>(0,s.jsx)("span",{className:d().tagPill,children:e.name},e.name))}),(0,s.jsx)("div",{className:d().descriptionBlock,children:e.description||"No description."})]}):(0,s.jsx)("section",{className:d().detailPanel,children:"No idea selected"})}function ez({label:e,value:t}){return(0,s.jsxs)("div",{className:d().metaCell,children:[(0,s.jsx)("span",{className:d().metaLabel,children:e}),(0,s.jsx)("span",{className:d().metaValue,children:t??"Unassigned"})]})}function eG({surface:e}){let{systems:t,activeSystemId:a}=(0,l.V)(),n=(0,r.useMemo)(()=>t.find(e=>e.id===a)??null,[a,t]);return(0,s.jsx)(v,{activeSystem:n,activeSurface:e,children:n?"data"===e?(0,s.jsx)(D,{system:n}):"benchmarks"===e?(0,s.jsx)(C,{system:n}):"ideas"===e?(0,s.jsx)(eF,{system:n}):(0,s.jsx)(eH,{system:n}):(0,s.jsx)(eK,{systemCount:t.length})})}function eK({systemCount:e}){return(0,s.jsxs)("div",{className:d().emptyState,children:[(0,s.jsx)("img",{src:"/logo-cream.svg",alt:"Kaizen",className:d().emptyLogo}),(0,s.jsx)("h1",{className:d().emptyTitle,children:"Kaizen"}),(0,s.jsx)("p",{className:d().emptyCopy,children:"Select a system to inspect its data, benchmarks, ideas, and experiments."}),(0,s.jsxs)("span",{className:d().emptyMeta,children:[e," system",1===e?"":"s"]})]})}},684:e=>{e.exports={page:"Studio_page__X6enu",topBar:"Studio_topBar__OlLt6",logoLink:"Studio_logoLink__Hv20q",topLogo:"Studio_topLogo__YEKp4",systemSlot:"Studio_systemSlot__5m3MM",surfaceNav:"Studio_surfaceNav__RKZjS",surfaceLink:"Studio_surfaceLink__ivMXp",surfaceLinkActive:"Studio_surfaceLinkActive__Der5b",surfaceLinkDisabled:"Studio_surfaceLinkDisabled__gfVYR",statusPill:"Studio_statusPill__m2ERk",content:"Studio_content__LELHT",surface:"Studio_surface__lxZ_I",surfaceBanner:"Studio_surfaceBanner__UNvfB",surfaceBannerText:"Studio_surfaceBannerText__6jNss",surfaceBannerTitle:"Studio_surfaceBannerTitle__vXJox",surfaceBannerCopy:"Studio_surfaceBannerCopy__Abpvv",sourceChip:"Studio_sourceChip___xMTP",sourceChip_code:"Studio_sourceChip_code__DPzDj",sourceChip_langfuse:"Studio_sourceChip_langfuse__h6LSy",sourceChip_linear:"Studio_sourceChip_linear__9NNQ5",sourceChip_local:"Studio_sourceChip_local__3Ws2K",sourceLogo:"Studio_sourceLogo__iWmOz",sourceSvgIcon:"Studio_sourceSvgIcon__dpcyH",provenanceLine:"Studio_provenanceLine__a5uDd",inlineSourceValue:"Studio_inlineSourceValue__WMOb9",syncStatus:"Studio_syncStatus__exlkz",syncStatusLabel:"Studio_syncStatusLabel__C5cmS",iconButton:"Studio_iconButton__aVxFx",dataLayout:"Studio_dataLayout__RTqFZ",experimentsLayout:"Studio_experimentsLayout__dJ9_W",dataNav:"Studio_dataNav__Hhqcz",experimentNav:"Studio_experimentNav__pUm9S",sidebarResizeHandle:"Studio_sidebarResizeHandle__Bfcoi",experimentsMain:"Studio_experimentsMain__8IYj4",dataNavItem:"Studio_dataNavItem__rzaJQ",dataNavGroupItem:"Studio_dataNavGroupItem__0bfKJ",dataNavItemText:"Studio_dataNavItemText__nSadp",dataNavItemMeta:"Studio_dataNavItemMeta__PEhqv",experimentNavItem:"Studio_experimentNavItem__VrdaI",dataNavEmpty:"Studio_dataNavEmpty__sffpq",dataNavItemActive:"Studio_dataNavItemActive__TANbd",dataMain:"Studio_dataMain__rSYxU",surfaceToolbar:"Studio_surfaceToolbar__ayM6U",surfaceTitle:"Studio_surfaceTitle__7SjZP",surfaceSubtitle:"Studio_surfaceSubtitle__i6YB6",inlineSelector:"Studio_inlineSelector__er_h_",inlineSelectorLabel:"Studio_inlineSelectorLabel__T4g6q",filterBar:"Studio_filterBar__I6G_F",select:"Studio_select__rbQ0i",primaryButton:"Studio_primaryButton__t0Otf",secondaryButton:"Studio_secondaryButton__gEhyV",dangerButton:"Studio_dangerButton__9Bclj",splitSurface:"Studio_splitSurface__v2lGr",listPanel:"Studio_listPanel__KQbqx",listPanelEmpty:"Studio_listPanelEmpty__Ofh2D",detailPanel:"Studio_detailPanel__upjwv",errorPanel:"Studio_errorPanel__OFovk",loadingPanel:"Studio_loadingPanel__VwGMY",emptyPanel:"Studio_emptyPanel__nqsJ1",listRow:"Studio_listRow__NObDS",listRowSelected:"Studio_listRowSelected__v7gF6",listRowTitle:"Studio_listRowTitle__Dg_xF",listRowMeta:"Studio_listRowMeta__fmtZx",detailHeader:"Studio_detailHeader__an_N2",detailTitle:"Studio_detailTitle__H8HXN",detailMeta:"Studio_detailMeta__F__Kq",mutedText:"Studio_mutedText__gxlbZ",detailLink:"Studio_detailLink__QdlV6",detailGrid:"Studio_detailGrid__eKjPQ",jsonBlock:"Studio_jsonBlock__V4yWY",jsonTitle:"Studio_jsonTitle__4rJZO",sectionTitle:"Studio_sectionTitle__AgJZy",jsonPre:"Studio_jsonPre__NT0BZ",traceSection:"Studio_traceSection__7pMTX",traceHeader:"Studio_traceHeader__Ulrn7",traceName:"Studio_traceName__YsChW",tagRow:"Studio_tagRow__Prgun",tagPill:"Studio_tagPill__TRnnW",successPanel:"Studio_successPanel__03cvQ",evalDefinitionGrid:"Studio_evalDefinitionGrid__YMID0",evalDefinitionCell:"Studio_evalDefinitionCell__wS4ZE",evalDefinitionCode:"Studio_evalDefinitionCode___J0aO",evalHistorySection:"Studio_evalHistorySection__e_91j",evalSectionHeader:"Studio_evalSectionHeader___IiLV",evalSummaryTable:"Studio_evalSummaryTable__gZ5vh",evalSummaryHeader:"Studio_evalSummaryHeader__ku_ov",evalSummaryRow:"Studio_evalSummaryRow__KtPSR",evalSummaryPrimary:"Studio_evalSummaryPrimary__VAArE",evalSummaryMeta:"Studio_evalSummaryMeta__RgcSF",setupBanner:"Studio_setupBanner__rn3R6",setupTitle:"Studio_setupTitle__rrPH_",setupCopy:"Studio_setupCopy__InFUm",setupHeader:"Studio_setupHeader__0z51Y",setupState:"Studio_setupState__u13N3",setupGrid:"Studio_setupGrid__43Ac0",setupBlock:"Studio_setupBlock__pXdLU",setupBlockTitle:"Studio_setupBlockTitle__HS_1S",setupCode:"Studio_setupCode__3AADb",setupLookup:"Studio_setupLookup__S7xvC",setupDetail:"Studio_setupDetail__BYnnc",setupLookupLabel:"Studio_setupLookupLabel__OOHGt",ideaMetaGrid:"Studio_ideaMetaGrid__QA5ZP",metaCell:"Studio_metaCell__lTOia",metaLabel:"Studio_metaLabel__orMyw",metaValue:"Studio_metaValue__wPOIw",descriptionBlock:"Studio_descriptionBlock__RgZHz",formRow:"Studio_formRow__CkHSQ",formLabel:"Studio_formLabel__pObCQ",textInput:"Studio_textInput__9WjCF",textArea:"Studio_textArea__yi_Lm",textAreaLarge:"Studio_textAreaLarge__20NQc",actionRow:"Studio_actionRow__0pQf0",actionFooter:"Studio_actionFooter__wFiYP",emptyState:"Studio_emptyState__dwICN",emptyLogo:"Studio_emptyLogo__ehso4",emptyTitle:"Studio_emptyTitle__ByiEP",emptyCopy:"Studio_emptyCopy__NP7hT",emptyMeta:"Studio_emptyMeta__Bsrtp",emptyPanelCopy:"Studio_emptyPanelCopy__EExQw",emptyPanelTitle:"Studio_emptyPanelTitle__g31eb"}},7810:e=>{e.exports={wrapper:"SystemSelector_wrapper__ERtiP",label:"SystemSelector_label__4i0LZ",select:"SystemSelector_select__NXXNb"}},9212:e=>{e.exports={layout:"Runs_layout__Ih4Cn",heroLogo:"Runs_heroLogo__BAUu8",headerLogo:"Runs_headerLogo__TgYiK",container:"Runs_container__0aqIq",headerRow:"Runs_headerRow__abEUi",title:"Runs_title__oroey",connectedBadge:"Runs_connectedBadge__d1Ss1",disconnectedBadge:"Runs_disconnectedBadge__v3kiS",systemTabs:"Runs_systemTabs__Gtuvz",systemTab:"Runs_systemTab__XYpgy",systemTabActive:"Runs_systemTabActive__10lyk",systemTabName:"Runs_systemTabName__pER4w",systemTabStatus:"Runs_systemTabStatus__YbTNa",systemStatus_in_progress:"Runs_systemStatus_in_progress__i_6lJ",systemStatus_completed:"Runs_systemStatus_completed__pjTng",systemStatus_not_started:"Runs_systemStatus_not_started__FqOI6",subtitle:"Runs_subtitle__ZfT5k",targetBadge:"Runs_targetBadge__XoNhG",grid:"Runs_grid__7DXl9",card:"Runs_card__AxGEL",cardFlash:"Runs_cardFlash__pXo7_",cardGlow:"Runs_cardGlow__mf3t7",badgeFlash:"Runs_badgeFlash__Q0C8G",badgePop:"Runs_badgePop__PdKoP",textFlash:"Runs_textFlash__O_0mj",textHighlight:"Runs_textHighlight__9znCz",rowSlideIn:"Runs_rowSlideIn__rGZ9I",slideInFade:"Runs_slideInFade__SsIak",diagramFlash:"Runs_diagramFlash__4ZTFP",diagramPulse:"Runs_diagramPulse__yN0dx",cardHeader:"Runs_cardHeader__7EpGx",cardTitle:"Runs_cardTitle__l82f2",bestAccuracyInline:"Runs_bestAccuracyInline__0XYna",linearIssueLink:"Runs_linearIssueLink__aVa6L",statusBadge:"Runs_statusBadge__T4ifS",detailTitleStack:"Runs_detailTitleStack__a_E4h",runProgressBlock:"Runs_runProgressBlock__Hg4fw",runProgressHeader:"Runs_runProgressHeader__e69Sy",runProgressTrack:"Runs_runProgressTrack__c_Wjv",runProgressFill:"Runs_runProgressFill__JOLgM",runMetaGrid:"Runs_runMetaGrid__kiPS9",runMetaCell:"Runs_runMetaCell__lEV1l",runMetaLabel:"Runs_runMetaLabel__DxMGL",runMetaValue:"Runs_runMetaValue__gJoib",detailEmpty:"Runs_detailEmpty__wy74V",metricsTable:"Runs_metricsTable__C_A_z",iterRowClickable:"Runs_iterRowClickable__g1tjc",iterRowExpanded:"Runs_iterRowExpanded__Nbl8_",bestRow:"Runs_bestRow__A_kku",iterDetailRow:"Runs_iterDetailRow__y10KK",cardFooter:"Runs_cardFooter__o_f_i",footerUpdated:"Runs_footerUpdated__21L7o",chartContainer:"Runs_chartContainer__bTh0u",chartTitle:"Runs_chartTitle__OkYqw",chart:"Runs_chart__XhGIf",chartRow:"Runs_chartRow__DbGz_",chartLabel:"Runs_chartLabel__3FlRl",chartBarContainer:"Runs_chartBarContainer__zzYv4",chartBarTrack:"Runs_chartBarTrack__5ogbA",chartBar:"Runs_chartBar__64D55",chartRowClickable:"Runs_chartRowClickable__RxDfH",chartValue:"Runs_chartValue__lmV61",metricsTableHeader:"Runs_metricsTableHeader__8d74Z",metricsTableRow:"Runs_metricsTableRow__FEGrk",metricsTableLabel:"Runs_metricsTableLabel__K8De_",metricsTableValue:"Runs_metricsTableValue__kaegJ",metricsTableBarCell:"Runs_metricsTableBarCell__8h1wf",metricsTableBarTrack:"Runs_metricsTableBarTrack__b_CnH",metricsTableBar:"Runs_metricsTableBar__HScEe",chartItems:"Runs_chartItems__Quwyx",noData:"Runs_noData__fi5_n",runSection:"Runs_runSection__Ggi_B",runLabel:"Runs_runLabel__uGajV",treeForest:"Runs_treeForest__6GBAM",treeSvg:"Runs_treeSvg___AATo",treeNodeCard:"Runs_treeNodeCard__Y5XEX",treeNodeSelected:"Runs_treeNodeSelected__QaPg2",treeNodeBest:"Runs_treeNodeBest__U8FaL",treeNodeHighlight:"Runs_treeNodeHighlight__A7a9x",nodeHighlight:"Runs_nodeHighlight__Zuhi5",treeNodeName:"Runs_treeNodeName__XtYDH",treeNodeBadgeRow:"Runs_treeNodeBadgeRow__ZwAvw",treeNodeBadge:"Runs_treeNodeBadge__jLVIw",treeNodeElapsed:"Runs_treeNodeElapsed__vsyq2",treeNodeScoreRow:"Runs_treeNodeScoreRow__BN81e",treeNodeIssueRow:"Runs_treeNodeIssueRow__7fXg_",treeNodeAccuracy:"Runs_treeNodeAccuracy__ehSVZ",treeNodeItems:"Runs_treeNodeItems__4SRsq",treeNodeLoading:"Runs_treeNodeLoading__P7_CS",treeNodePulse:"Runs_treeNodePulse__nC6rt",pulse:"Runs_pulse__LMgDF",treeNodeLoadingText:"Runs_treeNodeLoadingText__C1_bJ",detailPanelBackdrop:"Runs_detailPanelBackdrop__UMPlF",fadeIn:"Runs_fadeIn__1nqka",detailPanel:"Runs_detailPanel__BII9F",slideInFromRight:"Runs_slideInFromRight__uHSvO",detailPanelClose:"Runs_detailPanelClose__vgQbo",detailPanelHeader:"Runs_detailPanelHeader__XMEDA",detailPanelTitle:"Runs_detailPanelTitle__vAuOY"}},9965:e=>{e.exports={tableWrapper:"Leaderboard_tableWrapper__6uunR",table:"Leaderboard_table__8gTqK",noData:"Leaderboard_noData__aKd3L",thRank:"Leaderboard_thRank__e16FK",thName:"Leaderboard_thName__i_zQZ",thStatus:"Leaderboard_thStatus__F4ZyW",thMetric:"Leaderboard_thMetric__ZMVo_",thPrimary:"Leaderboard_thPrimary__SzN76",thSorted:"Leaderboard_thSorted__13fmP",thLabel:"Leaderboard_thLabel__2cqYD",sortArrow:"Leaderboard_sortArrow__3IJre",thItems:"Leaderboard_thItems__3MS6a",thStarted:"Leaderboard_thStarted__9zNIp",row:"Leaderboard_row__A9H8t",rowBest:"Leaderboard_rowBest__JCqYu",rowSelected:"Leaderboard_rowSelected__GFyz5",tdRank:"Leaderboard_tdRank__kZKT_",tdName:"Leaderboard_tdName__2ymmx",expName:"Leaderboard_expName__6rkrZ",bestBadge:"Leaderboard_bestBadge__33jMs",tdStatus:"Leaderboard_tdStatus__zRhgm",statusDot:"Leaderboard_statusDot__hEPYp",tdMetric:"Leaderboard_tdMetric__Ll0O_",tdPrimary:"Leaderboard_tdPrimary__SZr_5",metricValue:"Leaderboard_metricValue__IhFZf",metricEmpty:"Leaderboard_metricEmpty__l1hwN",tdItems:"Leaderboard_tdItems__4_5Q6",tdStarted:"Leaderboard_tdStarted__p97S3"}}}]);
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
(self.webpackChunk_N_E=self.webpackChunk_N_E||[]).push([[680],{318:(_,s,u)=>{(window.__NEXT_P=window.__NEXT_P||[]).push(["/[system]/data",function(){return u(5897)}])},5897:(_,s,u)=>{"use strict";u.r(s),u.d(s,{default:()=>t});var e=u(144),n=u(374);function t(){return(0,e.jsx)(n.W,{surface:"data"})}}},_=>{_.O(0,[431,374,636,593,792],()=>_(_.s=318)),_N_E=_.O()}]);
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
(self.webpackChunk_N_E=self.webpackChunk_N_E||[]).push([[250],{9403:(e,s,_)=>{"use strict";_.r(s),_.d(s,{default:()=>r});var n=_(144),u=_(374);function r(){return(0,n.jsx)(u.W,{surface:"benchmarks"})}},9578:(e,s,_)=>{(window.__NEXT_P=window.__NEXT_P||[]).push(["/[system]/eval",function(){return _(9403)}])}},e=>{e.O(0,[431,374,636,593,792],()=>e(e.s=9578)),_N_E=e.O()}]);
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
(self.webpackChunk_N_E=self.webpackChunk_N_E||[]).push([[24],{5103:(e,s,n)=>{"use strict";n.r(s),n.d(s,{default:()=>r});var _=n(144),u=n(374);function r(){return(0,_.jsx)(u.W,{surface:"experiments"})}},8758:(e,s,n)=>{(window.__NEXT_P=window.__NEXT_P||[]).push(["/[system]/experiments",function(){return n(5103)}])}},e=>{e.O(0,[431,374,636,593,792],()=>e(e.s=8758)),_N_E=e.O()}]);
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
(self.webpackChunk_N_E=self.webpackChunk_N_E||[]).push([[912],{4983:(s,e,_)=>{"use strict";_.r(e),_.d(e,{default:()=>r});var u=_(144),n=_(374);function r(){return(0,u.jsx)(n.W,{surface:"ideas"})}},8694:(s,e,_)=>{(window.__NEXT_P=window.__NEXT_P||[]).push(["/[system]/ideas",function(){return _(4983)}])}},s=>{s.O(0,[431,374,636,593,792],()=>s(s.s=8694)),_N_E=s.O()}]);
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
(self.webpackChunk_N_E=self.webpackChunk_N_E||[]).push([[332],{1881:(e,_,n)=>{"use strict";n.r(_),n.d(_,{default:()=>r});var u=n(144),s=n(374);function r(){return(0,u.jsx)(s.W,{surface:"experiments"})}},4842:(e,_,n)=>{(window.__NEXT_P=window.__NEXT_P||[]).push(["/",function(){return n(1881)}])}},e=>{e.O(0,[431,374,636,593,792],()=>e(e.s=4842)),_N_E=e.O()}]);
|
package/dashboard/.next/standalone/packages/kaizen/dashboard/.next/static/css/e75cf1946c214544.css
DELETED
|
@@ -1 +0,0 @@
|
|
|
1
|
-
.Studio_page__X6enu{min-height:100%;background:#0a0a0a;color:#fafafa;font-family:Geist,ui-sans-serif,system-ui,-apple-system,sans-serif}.Studio_topBar__OlLt6{display:grid;grid-template-columns:max-content max-content max-content minmax(0,1fr);align-items:center;gap:.9rem;padding:1rem 1.5rem}.Studio_logoLink__Hv20q{display:inline-flex;align-items:center;justify-content:center;width:26px;height:26px;text-decoration:none}.Studio_topLogo__YEKp4{display:block;width:22px;height:22px;object-fit:contain}.Studio_systemSlot__5m3MM{min-width:0}.Studio_surfaceNav__RKZjS{display:flex;align-items:center;gap:.25rem;min-width:0}.Studio_surfaceLink__ivMXp{display:inline-flex;align-items:center;justify-content:center;height:32px;padding:0 .8rem;border-radius:0;color:var(--text-tertiary);text-decoration:none;font-size:.85rem;font-weight:500;white-space:nowrap;transition:background .15s,color .15s}.Studio_surfaceLink__ivMXp:hover{color:var(--text-secondary);background:rgba(255,255,255,.05)}.Studio_surfaceLinkActive__Der5b{color:#fafafa;background:rgba(255,255,255,.08)}.Studio_surfaceLinkDisabled__gfVYR{opacity:.45;pointer-events:none}.Studio_statusPill__m2ERk{display:inline-flex;align-items:center;min-height:22px;border-radius:0;padding:0 .45rem;color:var(--text-secondary);background:rgba(255,255,255,.06);border:1px solid rgba(255,255,255,.08);font-size:.68rem;font-weight:600;white-space:nowrap}.Studio_content__LELHT{padding:0 1.5rem 2rem}.Studio_surface__lxZ_I{min-width:0}.Studio_surfaceBanner__UNvfB{display:flex;align-items:flex-start;justify-content:space-between;gap:1rem;margin-bottom:1rem;border:1px solid rgba(255,255,255,.08);border-radius:0;background:rgba(255,255,255,.035);padding:.8rem .9rem}.Studio_surfaceBannerText__6jNss{min-width:0}.Studio_surfaceBannerTitle__vXJox{display:flex;align-items:center;gap:.45rem;color:#fafafa;font-size:.9rem;font-weight:650}.Studio_surfaceBannerCopy__Abpvv{margin-top:.2rem;color:var(--text-secondary);font-size:.78rem;line-height:1.45}.Studio_surfaceBannerCopy__Abpvv code{color:#fafafa;background:rgba(0,0,0,.24);border-radius:0;padding:.05rem .25rem}.Studio_sourceChip___xMTP{display:inline-flex;align-items:center;justify-content:center;width:22px;height:22px;border:1px solid rgba(255,255,255,.1);border-radius:0;background:rgba(255,255,255,.04);color:#fafafa;line-height:1;padding:0;vertical-align:middle}.Studio_sourceChip_code__DPzDj{color:#fafafa}.Studio_sourceChip_langfuse__h6LSy,.Studio_sourceChip_linear__9NNQ5{background:rgba(255,255,255,.06)}.Studio_sourceChip_local__3Ws2K{color:#fafafa}.Studio_sourceLogo__iWmOz{object-fit:contain}.Studio_sourceLogo__iWmOz,.Studio_sourceSvgIcon__dpcyH{display:block;width:15px;height:15px}.Studio_provenanceLine__a5uDd{display:flex;align-items:center;flex-wrap:wrap;gap:.4rem;margin-top:.45rem;color:var(--text-tertiary);font-size:.74rem}.Studio_provenanceLine__a5uDd code{color:#fafafa;background:rgba(0,0,0,.24);border-radius:0;padding:.08rem .28rem}.Studio_inlineSourceValue__WMOb9{display:inline-flex;align-items:center;gap:.35rem;vertical-align:middle;white-space:nowrap}.Studio_syncStatus__exlkz{display:flex;align-items:center;justify-content:flex-end;gap:.4rem;flex-shrink:0}.Studio_syncStatusLabel__C5cmS{color:var(--text-tertiary);font-size:.74rem;font-weight:500;white-space:nowrap}.Studio_iconButton__aVxFx{display:inline-flex;align-items:center;justify-content:center;width:32px;height:32px;border:1px solid rgba(255,255,255,.09);border-radius:0;background:rgba(255,255,255,.04);color:#fafafa;cursor:pointer;font-family:inherit;font-size:1rem;line-height:1;padding:0;white-space:nowrap}.Studio_iconButton__aVxFx:hover:not(:disabled){background:rgba(255,255,255,.07);border-color:rgba(255,255,255,.15)}.Studio_iconButton__aVxFx:disabled{color:var(--text-tertiary);cursor:default}.Studio_dataLayout__RTqFZ{display:grid;grid-template-columns:164px minmax(0,1fr);gap:1rem;align-items:start}.Studio_experimentsLayout__dJ9_W{display:flex;align-items:flex-start;min-width:0}.Studio_dataNav__Hhqcz{display:flex;flex-direction:column;gap:.25rem;align-self:stretch;min-height:calc(100vh - 8rem);padding:0 1rem 0 0;border-right:1px solid rgba(255,255,255,.08)}.Studio_experimentNav__pUm9S{min-width:220px;max-width:min(520px,45vw);overflow:auto;flex:0 0 auto;border-right:none}.Studio_sidebarResizeHandle__Bfcoi{position:relative;align-self:stretch;flex:0 0 16px;min-height:calc(100vh - 8rem);cursor:col-resize}.Studio_sidebarResizeHandle__Bfcoi:before{content:"";position:absolute;top:0;bottom:0;left:7px;width:1px;background:rgba(255,255,255,.08);transition:background .12s ease}.Studio_sidebarResizeHandle__Bfcoi:hover:before{background:rgba(255,255,255,.22)}.Studio_experimentsMain__8IYj4{margin-left:1rem}.Studio_dataNavItem__rzaJQ{display:inline-flex;align-items:center;justify-content:flex-start;width:fit-content;min-height:32px;border:1px solid transparent;border-radius:0;background:transparent;color:var(--text-tertiary);cursor:pointer;font-family:inherit;font-size:.82rem;padding:0 .65rem;text-align:left}.Studio_dataNavGroupItem__0bfKJ{flex-direction:column;align-items:flex-start;gap:.12rem;width:100%;min-height:48px;padding:.4rem .65rem}.Studio_dataNavItemMeta__PEhqv,.Studio_dataNavItemText__nSadp{display:block;width:100%;min-width:0;overflow:hidden;text-overflow:ellipsis;white-space:nowrap}.Studio_experimentNavItem__VrdaI .Studio_dataNavItemMeta__PEhqv,.Studio_experimentNavItem__VrdaI .Studio_dataNavItemText__nSadp{overflow:visible;text-overflow:clip;white-space:normal}.Studio_dataNavItemText__nSadp{color:inherit;line-height:1.2}.Studio_dataNavItemMeta__PEhqv{color:var(--text-tertiary);font-size:.7rem;line-height:1.2}.Studio_dataNavEmpty__sffpq{color:var(--text-tertiary);font-size:.78rem;padding:.5rem .65rem}.Studio_dataNavItem__rzaJQ:hover{color:var(--text-secondary);background:rgba(255,255,255,.03)}.Studio_dataNavItemActive__TANbd{color:#fafafa;background:rgba(255,255,255,.08);border-color:rgba(255,255,255,.14)}.Studio_dataMain__rSYxU{min-width:0;flex:1 1 auto}.Studio_surfaceToolbar__ayM6U{display:flex;align-items:center;justify-content:space-between;gap:1rem;margin-bottom:1rem}.Studio_surfaceTitle__7SjZP{color:#fafafa;font-size:.95rem;font-weight:600}.Studio_surfaceSubtitle__i6YB6{margin-top:.15rem;color:var(--text-tertiary);font-size:.75rem}.Studio_inlineSelector__er_h_{display:inline-flex;align-items:center;gap:.45rem;min-width:0;width:max-content}.Studio_inlineSelectorLabel__T4g6q{color:var(--text-tertiary);font-size:.72rem;font-weight:600;letter-spacing:.04em;text-transform:uppercase;white-space:nowrap}.Studio_filterBar__I6G_F{display:flex;justify-content:flex-end;margin-bottom:1rem}.Studio_dangerButton__9Bclj,.Studio_primaryButton__t0Otf,.Studio_secondaryButton__gEhyV,.Studio_select__rbQ0i{height:32px;border-radius:0;border:1px solid rgba(255,255,255,.09);background:rgba(255,255,255,.04);color:#fafafa;font-family:inherit;font-size:.8rem}.Studio_select__rbQ0i{max-width:340px;padding:0 .5rem}.Studio_primaryButton__t0Otf{padding:0 .85rem;cursor:pointer;color:#0a0a0a;background:#fafafa;border-color:#fafafa;font-weight:600}.Studio_dangerButton__9Bclj,.Studio_secondaryButton__gEhyV{padding:0 .75rem;cursor:pointer}.Studio_dangerButton__9Bclj{color:#ffb3a6;background:rgba(232,85,61,.08);border-color:rgba(232,85,61,.22)}.Studio_primaryButton__t0Otf:hover:not(:disabled){background:var(--text-secondary);border-color:var(--text-secondary)}.Studio_secondaryButton__gEhyV:hover:not(:disabled){background:rgba(255,255,255,.07);border-color:rgba(255,255,255,.15)}.Studio_dangerButton__9Bclj:hover:not(:disabled){background:rgba(232,85,61,.13);border-color:rgba(232,85,61,.34)}.Studio_dangerButton__9Bclj:disabled,.Studio_primaryButton__t0Otf:disabled,.Studio_secondaryButton__gEhyV:disabled{color:var(--text-tertiary);cursor:default}.Studio_primaryButton__t0Otf:disabled{background:rgba(255,255,255,.08);border-color:rgba(255,255,255,.09)}.Studio_splitSurface__v2lGr{display:grid;grid-template-columns:minmax(240px,340px) minmax(0,1fr);gap:1rem;min-height:560px}.Studio_detailPanel__upjwv,.Studio_emptyPanel__nqsJ1,.Studio_errorPanel__OFovk,.Studio_listPanelEmpty__Ofh2D,.Studio_listPanel__KQbqx,.Studio_loadingPanel__VwGMY{border:1px solid rgba(255,255,255,.08);border-radius:0;background:rgba(255,255,255,.03)}.Studio_listPanel__KQbqx{overflow:auto;max-height:calc(100vh - 280px)}.Studio_emptyPanel__nqsJ1,.Studio_listPanelEmpty__Ofh2D,.Studio_loadingPanel__VwGMY{display:flex;align-items:center;justify-content:center;min-height:260px;color:var(--text-tertiary);font-size:.86rem;text-align:center;padding:2rem}.Studio_listRow__NObDS{display:flex;flex-direction:column;align-items:stretch;gap:.2rem;width:100%;padding:.75rem;border:0;border-bottom:1px solid rgba(255,255,255,.05);background:transparent;color:inherit;font-family:inherit;text-align:left;cursor:pointer}.Studio_listRow__NObDS:hover{background:rgba(255,255,255,.04)}.Studio_listRowSelected__v7gF6{background:rgba(91,141,239,.1);box-shadow:inset 2px 0 0 #5b8def}.Studio_listRowTitle__Dg_xF{color:#fafafa;font-size:.82rem;font-weight:500}.Studio_listRowMeta__fmtZx,.Studio_listRowTitle__Dg_xF{min-width:0;overflow:hidden;text-overflow:ellipsis;white-space:nowrap}.Studio_listRowMeta__fmtZx{color:var(--text-tertiary);font-size:.72rem}.Studio_detailPanel__upjwv{min-width:0;padding:1rem;overflow:auto;max-height:calc(100vh - 280px)}.Studio_detailHeader__an_N2{display:flex;align-items:flex-start;justify-content:space-between;gap:1rem;margin-bottom:1rem}.Studio_detailTitle__H8HXN{margin:0;color:#fafafa;font-size:1rem;font-weight:600;line-height:1.35;overflow-wrap:anywhere}.Studio_detailMeta__F__Kq,.Studio_mutedText__gxlbZ{margin-top:.2rem;color:var(--text-tertiary);font-size:.74rem;overflow-wrap:anywhere}.Studio_detailLink__QdlV6{color:#5b8def;font-size:.78rem;text-decoration:none}.Studio_detailLink__QdlV6:hover{text-decoration:underline}.Studio_detailGrid__eKjPQ{display:grid;grid-template-columns:repeat(2,minmax(0,1fr));gap:.75rem}.Studio_jsonBlock__V4yWY{min-width:0;margin-bottom:.75rem}.Studio_jsonTitle__4rJZO,.Studio_sectionTitle__AgJZy{margin-bottom:.35rem;color:var(--text-secondary);font-size:.72rem;font-weight:600;text-transform:uppercase;letter-spacing:.04em}.Studio_jsonPre__NT0BZ{max-height:320px;overflow:auto;margin:0;padding:.75rem;border-radius:0;background:rgba(0,0,0,.28);color:var(--text-secondary);font-size:.75rem;line-height:1.45;white-space:pre-wrap;overflow-wrap:anywhere}.Studio_traceSection__7pMTX{margin-top:1rem}.Studio_traceHeader__Ulrn7{display:flex;align-items:baseline;gap:.75rem;margin-bottom:.6rem}.Studio_traceName__YsChW{color:#fafafa;font-size:.9rem;font-weight:600;overflow-wrap:anywhere}.Studio_tagRow__Prgun{display:flex;flex-wrap:wrap;gap:.35rem;margin:.75rem 0}.Studio_tagPill__TRnnW{display:inline-flex;align-items:center;min-height:22px;padding:0 .45rem;border-radius:0;background:rgba(91,141,239,.12);color:#9ebcff;border:1px solid rgba(91,141,239,.22);font-size:.7rem;font-weight:500}.Studio_errorPanel__OFovk{margin-bottom:1rem;padding:.75rem;color:#ffb3a6;background:rgba(232,85,61,.1);border-color:rgba(232,85,61,.3);font-size:.82rem}.Studio_successPanel__03cvQ{margin-bottom:1rem;padding:.75rem;color:#b8f5df;background:rgba(0,212,170,.08);border:1px solid rgba(0,212,170,.22);border-radius:0;font-size:.82rem}.Studio_evalDefinitionGrid__YMID0{display:grid;grid-template-columns:repeat(4,minmax(0,1fr));gap:.75rem}.Studio_evalDefinitionCell__wS4ZE{min-width:0}.Studio_evalDefinitionCode___J0aO{display:block;width:fit-content;max-width:100%;margin-top:.2rem;background:rgba(0,0,0,.24);color:var(--text-tertiary);font-size:.72rem;padding:.1rem .28rem;overflow-wrap:anywhere}.Studio_evalHistorySection__e_91j{margin-top:1rem}.Studio_evalSectionHeader___IiLV{display:flex;align-items:flex-start;justify-content:space-between;gap:1rem;margin-bottom:.75rem}.Studio_evalSummaryTable__gZ5vh{border:1px solid rgba(255,255,255,.08);background:rgba(255,255,255,.03)}.Studio_evalSummaryHeader__ku_ov,.Studio_evalSummaryRow__KtPSR{display:grid;grid-template-columns:minmax(180px,1.35fr) minmax(180px,1.35fr) 80px 80px 160px;gap:.75rem;align-items:center}.Studio_evalSummaryHeader__ku_ov{padding:.55rem .75rem;border-bottom:1px solid rgba(255,255,255,.08);color:var(--text-tertiary);font-size:.7rem;font-weight:600;letter-spacing:.04em;text-transform:uppercase}.Studio_evalSummaryRow__KtPSR{padding:.65rem .75rem;border-bottom:1px solid rgba(255,255,255,.05);color:var(--text-secondary);font-size:.78rem}.Studio_evalSummaryRow__KtPSR:last-child{border-bottom:0}.Studio_evalSummaryMeta__RgcSF,.Studio_evalSummaryPrimary__VAArE{display:block;min-width:0;overflow:hidden;text-overflow:ellipsis;white-space:nowrap}.Studio_evalSummaryPrimary__VAArE{color:#fafafa;font-weight:500}.Studio_evalSummaryMeta__RgcSF{margin-top:.12rem;color:var(--text-tertiary);font-size:.7rem}.Studio_setupBanner__rn3R6{margin-bottom:1rem;border:1px solid rgba(245,166,35,.34);border-radius:0;background:rgba(245,166,35,.08);padding:.85rem}.Studio_setupTitle__rrPH_{color:#fafafa;font-size:.88rem;font-weight:600}.Studio_setupCopy__InFUm{margin-top:.2rem;color:var(--text-secondary);font-size:.8rem;line-height:1.5}.Studio_setupHeader__0z51Y{display:flex;align-items:flex-start;justify-content:space-between;gap:1rem}.Studio_setupState__u13N3{display:inline-flex;align-items:center;min-height:22px;border-radius:0;padding:0 .45rem;color:#f5c77b;background:rgba(245,166,35,.1);border:1px solid rgba(245,166,35,.3);font-size:.68rem;font-weight:600;white-space:nowrap}.Studio_setupGrid__43Ac0{display:grid;grid-template-columns:repeat(3,minmax(0,1fr));gap:.75rem;margin-top:.8rem}.Studio_setupBlock__pXdLU{min-width:0}.Studio_setupBlockTitle__HS_1S{display:block;margin-bottom:.35rem;color:var(--text-tertiary);font-size:.68rem;font-weight:600;text-transform:uppercase;letter-spacing:.04em}.Studio_setupCode__3AADb{display:block;width:fit-content;max-width:100%;margin-top:.25rem;border-radius:0;background:rgba(0,0,0,.26);color:#fafafa;padding:.18rem .38rem;font-size:.72rem;overflow-wrap:anywhere}.Studio_setupDetail__BYnnc,.Studio_setupLookup__S7xvC{margin-top:.75rem;color:var(--text-tertiary);font-size:.75rem;line-height:1.5;overflow-wrap:anywhere}.Studio_setupLookupLabel__OOHGt{color:var(--text-secondary);font-weight:600;margin-right:.35rem}.Studio_ideaMetaGrid__QA5ZP{display:grid;grid-template-columns:repeat(4,minmax(0,1fr));gap:.75rem;margin-bottom:.75rem}.Studio_metaCell__lTOia{min-width:0}.Studio_metaLabel__orMyw{display:block;color:var(--text-tertiary);font-size:.7rem}.Studio_metaValue__wPOIw{display:block;margin-top:.1rem;color:var(--text-secondary);font-size:.78rem;overflow:hidden;text-overflow:ellipsis;white-space:nowrap}.Studio_descriptionBlock__RgZHz{white-space:pre-wrap;overflow-wrap:anywhere;color:var(--text-secondary);font-size:.84rem;line-height:1.55}.Studio_formRow__CkHSQ{display:flex;flex-direction:column;gap:.35rem;margin-bottom:.75rem}.Studio_formLabel__pObCQ{color:var(--text-tertiary);font-size:.72rem;font-weight:600;text-transform:uppercase;letter-spacing:.04em}.Studio_textInput__9WjCF{min-height:34px;padding:0 .65rem}.Studio_textAreaLarge__20NQc,.Studio_textArea__yi_Lm,.Studio_textInput__9WjCF{border:1px solid rgba(255,255,255,.1);border-radius:0;background:rgba(0,0,0,.26);color:#fafafa;font-family:inherit;font-size:.84rem}.Studio_textAreaLarge__20NQc,.Studio_textArea__yi_Lm{min-height:96px;resize:vertical;line-height:1.45;padding:.6rem .65rem}.Studio_textAreaLarge__20NQc{min-height:180px;font-family:Geist Mono,ui-monospace,monospace;font-size:.78rem}.Studio_textAreaLarge__20NQc::placeholder,.Studio_textArea__yi_Lm::placeholder,.Studio_textInput__9WjCF::placeholder{color:var(--text-tertiary)}.Studio_actionRow__0pQf0{display:flex;flex-wrap:wrap;justify-content:flex-end;gap:.5rem}.Studio_actionFooter__wFiYP{display:flex;align-items:center;justify-content:space-between;gap:1rem;margin:1rem 0;padding-top:.75rem;border-top:1px solid rgba(255,255,255,.08)}.Studio_emptyState__dwICN{display:flex;flex-direction:column;align-items:center;justify-content:center;min-height:560px;gap:.9rem;text-align:center}.Studio_emptyLogo__ehso4{width:220px;height:auto;opacity:.85}.Studio_emptyTitle__ByiEP{margin:0;color:#fafafa;font-size:1.45rem;font-weight:600}.Studio_emptyCopy__NP7hT,.Studio_emptyMeta__Bsrtp,.Studio_emptyPanelCopy__EExQw{color:var(--text-tertiary);font-size:.86rem;line-height:1.5}.Studio_emptyPanel__nqsJ1{flex-direction:column;gap:.35rem}.Studio_emptyPanelTitle__g31eb{color:#fafafa;font-size:.95rem;font-weight:600}@media (max-width:900px){.Studio_topBar__OlLt6{grid-template-columns:minmax(0,1fr);align-items:stretch}.Studio_surfaceNav__RKZjS{overflow-x:auto}.Studio_actionFooter__wFiYP,.Studio_detailHeader__an_N2,.Studio_setupHeader__0z51Y,.Studio_surfaceBanner__UNvfB,.Studio_surfaceToolbar__ayM6U,.Studio_traceHeader__Ulrn7{flex-direction:column;align-items:stretch}.Studio_syncStatus__exlkz{justify-content:flex-start}.Studio_dataLayout__RTqFZ,.Studio_detailGrid__eKjPQ,.Studio_evalDefinitionGrid__YMID0,.Studio_evalSummaryHeader__ku_ov,.Studio_evalSummaryRow__KtPSR,.Studio_ideaMetaGrid__QA5ZP,.Studio_setupGrid__43Ac0,.Studio_splitSurface__v2lGr{grid-template-columns:1fr}.Studio_detailPanel__upjwv,.Studio_listPanel__KQbqx{max-height:none}}.SystemSelector_wrapper__ERtiP{display:inline-flex;align-items:center;gap:.45rem;min-width:0;width:max-content}.SystemSelector_label__4i0LZ{color:var(--text-tertiary,#666);font-size:.72rem;font-weight:600;letter-spacing:.04em;text-transform:uppercase;white-space:nowrap}.SystemSelector_select__NXXNb{width:auto;max-width:280px;height:32px;border:1px solid rgba(255,255,255,.1);border-radius:0;background:rgba(255,255,255,.04);color:var(--text-primary,#fafafa);font-family:inherit;font-size:.84rem;padding:0 1.85rem 0 .65rem;cursor:pointer}.SystemSelector_select__NXXNb:hover{background:rgba(255,255,255,.07);border-color:rgba(255,255,255,.16)}.Runs_layout__Ih4Cn{display:flex;flex:1 1;min-height:100%;background:#0a0a0a}.Runs_heroLogo__BAUu8{width:240px;height:auto;opacity:.85}.Runs_headerLogo__TgYiK{width:18px;height:18px;object-fit:contain;border-radius:0;flex-shrink:0}.Runs_container__0aqIq{flex:1 1;min-width:0;margin:0;padding:2rem;font-family:Geist,ui-sans-serif,system-ui,-apple-system,sans-serif;background:#0a0a0a;color:#fafafa;min-height:100%;box-sizing:border-box;overflow-x:hidden}.Runs_headerRow__abEUi{display:flex;align-items:center;justify-content:space-between;gap:.75rem;padding:.6rem 0;border-bottom:1px solid rgba(255,255,255,.06);margin-bottom:.5rem}.Runs_title__oroey{font-size:1rem;font-weight:500;margin:0;color:#fff;white-space:nowrap;flex-shrink:0}.Runs_connectedBadge__d1Ss1{border:1px solid rgba(0,212,170,.4);color:#00d4aa;background:rgba(0,212,170,.1)}.Runs_connectedBadge__d1Ss1,.Runs_disconnectedBadge__v3kiS{font-size:.65rem;font-weight:500;text-transform:uppercase;letter-spacing:1px;padding:.2rem .6rem}.Runs_disconnectedBadge__v3kiS{border:1px solid rgba(245,166,35,.4);color:#f5a623;background:rgba(245,166,35,.1)}.Runs_systemTabs__Gtuvz{display:flex;gap:.25rem;margin-bottom:.5rem;border-bottom:1px solid rgba(255,255,255,.06);padding-bottom:0;overflow-x:auto}.Runs_systemTab__XYpgy{display:flex;align-items:center;gap:.5rem;padding:.6rem 1rem;background:none;border:none;border-bottom:2px solid transparent;color:var(--text-tertiary);font-family:inherit;font-size:.85rem;font-weight:400;cursor:pointer;transition:color .15s,border-color .15s;white-space:nowrap}.Runs_systemTab__XYpgy:hover{color:var(--text-secondary)}.Runs_systemTabActive__10lyk{color:#fff;border-bottom-color:#00d4aa}.Runs_systemTabName__pER4w{font-weight:400}.Runs_systemTabStatus__YbTNa{font-size:.6rem;font-weight:500;text-transform:uppercase;letter-spacing:.5px;padding:.1rem .4rem}.Runs_systemStatus_in_progress__i_6lJ{color:#00d4aa;background:rgba(0,212,170,.1);border:1px solid rgba(0,212,170,.3)}.Runs_systemStatus_completed__pjTng{color:#5b8def;background:rgba(91,141,239,.1);border:1px solid rgba(91,141,239,.3)}.Runs_systemStatus_not_started__FqOI6{color:var(--text-tertiary);background:rgba(255,255,255,.04);border:1px solid rgba(255,255,255,.08)}.Runs_subtitle__ZfT5k{color:var(--text-tertiary);font-size:.75rem;font-weight:400;line-height:1.4;margin-top:.15rem}.Runs_targetBadge__XoNhG{font-size:.65rem;color:var(--text-tertiary);background:rgba(255,255,255,.05);border:1px solid rgba(255,255,255,.08);border-radius:0;padding:.1rem .4rem;white-space:nowrap;flex-shrink:0}.Runs_grid__7DXl9{display:grid;grid-template-columns:repeat(3,1fr);gap:1.25rem;margin-bottom:3rem}.Runs_card__AxGEL{border:1px solid rgba(255,255,255,.08);padding:1.5rem;background:rgba(255,255,255,.03);transition:border-color .4s,box-shadow .4s;min-width:0;overflow:hidden}.Runs_card__AxGEL:hover{border-color:rgba(255,255,255,.15)}@keyframes Runs_cardGlow__mf3t7{0%{border-color:rgba(0,212,170,.8);box-shadow:0 0 20px rgba(0,212,170,.15),inset 0 0 20px rgba(0,212,170,.03)}to{border-color:rgba(255,255,255,.08);box-shadow:none}}@keyframes Runs_badgePop__PdKoP{0%{transform:scale(1.3);filter:brightness(1.8)}40%{transform:scale(.95)}to{transform:scale(1);filter:brightness(1)}}@keyframes Runs_textHighlight__9znCz{0%{color:#00d4aa}to{color:var(--text-secondary)}}@keyframes Runs_slideInFade__SsIak{0%{opacity:0;transform:translateY(-8px);background:rgba(0,212,170,.12)}40%{opacity:1;transform:translateY(0)}to{background:transparent}}@keyframes Runs_diagramPulse__yN0dx{0%{border-color:rgba(0,212,170,.5);background:rgba(0,212,170,.06)}to{border-color:rgba(255,255,255,.06);background:rgba(255,255,255,.04)}}.Runs_cardFlash__pXo7_{animation:Runs_cardGlow__mf3t7 1.2s ease-out forwards}.Runs_badgeFlash__Q0C8G{animation:Runs_badgePop__PdKoP .5s cubic-bezier(.34,1.56,.64,1) forwards}.Runs_textFlash__O_0mj{animation:Runs_textHighlight__9znCz 1.2s ease-out forwards}.Runs_rowSlideIn__rGZ9I{animation:Runs_slideInFade__SsIak .8s ease-out forwards}.Runs_diagramFlash__4ZTFP{animation:Runs_diagramPulse__yN0dx 1.2s ease-out forwards}.Runs_cardHeader__7EpGx{display:flex;align-items:center;gap:.5rem;margin-bottom:.5rem}.Runs_cardTitle__l82f2{font-size:1.1rem;font-weight:300;letter-spacing:-.2px;margin:0;color:#fff}.Runs_bestAccuracyInline__0XYna{margin-left:auto;font-size:.9rem;font-weight:500;font-feature-settings:"tnum";font-variant-numeric:tabular-nums;color:#00d4aa}.Runs_linearIssueLink__aVa6L{display:inline-flex;align-items:center;gap:.3rem;max-width:100%;min-height:24px;padding:0 .35rem;border:1px solid rgba(255,255,255,.1);background:rgba(255,255,255,.04);color:var(--text-secondary);font-size:.72rem;font-weight:600;text-decoration:none;white-space:nowrap}.Runs_linearIssueLink__aVa6L span{min-width:0;overflow:hidden;text-overflow:ellipsis}.Runs_linearIssueLink__aVa6L:hover{background:rgba(255,255,255,.07);border-color:rgba(255,255,255,.16);color:#fafafa}.Runs_linearIssueLink__aVa6L img{display:block;width:14px;height:14px;object-fit:contain}.Runs_statusBadge__T4ifS{display:inline-block;padding:.2rem .6rem;font-size:.6rem;font-weight:500;text-transform:uppercase;letter-spacing:1px;flex-shrink:0}.Runs_detailTitleStack__a_E4h{display:flex;flex-direction:column;min-width:0;gap:.35rem}.Runs_runProgressBlock__Hg4fw{margin-bottom:1rem;padding:.75rem;border:1px solid rgba(255,255,255,.08);background:rgba(255,255,255,.03)}.Runs_runProgressHeader__e69Sy{display:flex;align-items:center;justify-content:space-between;gap:1rem;margin-bottom:.45rem;color:var(--text-secondary);font-size:.76rem;font-feature-settings:"tnum";font-variant-numeric:tabular-nums}.Runs_runProgressTrack__c_Wjv{height:8px;background:rgba(255,255,255,.08);overflow:hidden}.Runs_runProgressFill__JOLgM{height:100%;background:#00d4aa}.Runs_runMetaGrid__kiPS9{display:grid;grid-template-columns:repeat(2,minmax(0,1fr));gap:.65rem;margin-bottom:1rem}.Runs_runMetaCell__lEV1l{min-width:0;padding:.6rem .65rem;border:1px solid rgba(255,255,255,.07);background:rgba(255,255,255,.025)}.Runs_runMetaLabel__DxMGL{display:block;color:var(--text-tertiary);font-size:.65rem;font-weight:600;letter-spacing:.04em;text-transform:uppercase}.Runs_runMetaValue__gJoib{display:block;min-width:0;margin-top:.18rem;color:var(--text-secondary);font-size:.78rem;overflow:hidden;text-overflow:ellipsis;white-space:nowrap}.Runs_detailEmpty__wy74V{padding:.75rem;border:1px solid rgba(255,255,255,.07);background:rgba(255,255,255,.025);color:var(--text-tertiary);font-size:.8rem}.Runs_metricsTable__C_A_z{font-size:.8rem;margin-bottom:1rem;table-layout:fixed}.Runs_metricsTable__C_A_z td,.Runs_metricsTable__C_A_z th{padding:.5rem;text-align:center;border-bottom:1px solid rgba(255,255,255,.06)}.Runs_metricsTable__C_A_z th{font-weight:500;font-size:.7rem;text-transform:uppercase;letter-spacing:.5px;color:var(--text-tertiary);background:transparent;cursor:help;-webkit-text-decoration:underline dotted rgba(250,250,250,.2);text-decoration:underline dotted rgba(250,250,250,.2);text-underline-offset:3px}.Runs_metricsTable__C_A_z td{color:var(--text-primary);font-feature-settings:"tnum";font-variant-numeric:tabular-nums}.Runs_iterRowClickable__g1tjc{cursor:pointer;transition:background .15s}.Runs_iterRowClickable__g1tjc:hover{background:rgba(255,255,255,.04)}.Runs_iterRowExpanded__Nbl8_{background:rgba(255,255,255,.03)}.Runs_iterRowExpanded__Nbl8_ td{border-bottom-color:transparent}.Runs_bestRow__A_kku{background:rgba(0,212,170,.06)}.Runs_bestRow__A_kku.Runs_iterRowClickable__g1tjc:hover{background:rgba(0,212,170,.1)}.Runs_bestRow__A_kku td{color:#00d4aa;font-weight:500}.Runs_iterDetailRow__y10KK td{padding:0;border-bottom:1px solid rgba(255,255,255,.06)}.Runs_cardFooter__o_f_i{display:flex;flex-wrap:wrap;justify-content:space-between;gap:.5rem .9rem;font-size:.7rem;color:var(--text-tertiary);letter-spacing:.2px;padding-top:.35rem}.Runs_footerUpdated__21L7o{margin-left:auto}.Runs_chartContainer__bTh0u{border:1px solid rgba(255,255,255,.08);padding:1.5rem;background:rgba(255,255,255,.03)}.Runs_chartTitle__OkYqw{font-size:1.25rem;font-weight:300;letter-spacing:-.3px;margin:0 0 1.25rem;color:#fff}.Runs_chart__XhGIf{display:flex;flex-direction:column;gap:.35rem}.Runs_chartRow__DbGz_{display:flex;align-items:center;gap:1rem;height:32px}.Runs_chartLabel__3FlRl{width:180px;font-size:.75rem;font-weight:400;text-align:right;flex-shrink:0;color:var(--text-secondary);overflow:hidden;text-overflow:ellipsis;white-space:nowrap}.Runs_chartBarContainer__zzYv4{flex:1 1;display:flex;align-items:center;gap:.5rem}.Runs_chartBarTrack__5ogbA{flex:1 1;height:24px;position:relative;background:transparent}.Runs_chartBar__64D55{height:100%;background:#00d4aa;transition:width .3s ease}.Runs_chartRowClickable__RxDfH{cursor:pointer;transition:background .15s}.Runs_chartRowClickable__RxDfH:hover{background:rgba(255,255,255,.04)}.Runs_chartValue__lmV61{font-size:.85rem;font-weight:500;white-space:nowrap;color:var(--text-primary);font-feature-settings:"tnum";font-variant-numeric:tabular-nums;width:52px;text-align:right;flex-shrink:0}.Runs_metricsTable__C_A_z{width:100%;border-collapse:collapse;font-size:.85rem}.Runs_metricsTableHeader__8d74Z{font-size:.7rem;font-weight:500;text-transform:uppercase;letter-spacing:.05em;color:var(--text-tertiary);padding:.4rem .75rem;border-bottom:1px solid rgba(255,255,255,.08);text-align:left}.Runs_metricsTableRow__FEGrk{border-bottom:1px solid rgba(255,255,255,.04)}.Runs_metricsTableRow__FEGrk:hover{background:rgba(255,255,255,.03)}.Runs_metricsTableLabel__K8De_{padding:.5rem .75rem;color:var(--text-secondary);font-weight:400;white-space:nowrap}.Runs_metricsTableValue__kaegJ{padding:.5rem .75rem;text-align:right;font-weight:600;font-feature-settings:"tnum";font-variant-numeric:tabular-nums;color:var(--text-primary);white-space:nowrap}.Runs_metricsTableBarCell__8h1wf{padding:.5rem .75rem}.Runs_metricsTableBarTrack__b_CnH{height:6px;border-radius:0;background:rgba(255,255,255,.06);overflow:hidden}.Runs_metricsTableBar__HScEe{height:100%;border-radius:0;background:#00d4aa;transition:width .3s ease}.Runs_chartItems__Quwyx{font-size:.65rem;color:var(--text-tertiary);width:32px;text-align:right;flex-shrink:0;font-feature-settings:"tnum";font-variant-numeric:tabular-nums}.Runs_noData__fi5_n{color:var(--text-tertiary);text-align:center;padding:3rem;font-size:.9rem}.Runs_runSection__Ggi_B{margin-bottom:.75rem}.Runs_runSection__Ggi_B:last-child{margin-bottom:0}.Runs_runLabel__uGajV{display:block;font-size:.65rem;font-weight:500;text-transform:uppercase;letter-spacing:.5px;color:var(--text-tertiary);margin-bottom:.3rem}.Runs_treeForest__6GBAM{overflow-x:auto;overflow-y:hidden;padding-bottom:2rem;margin-bottom:3rem}.Runs_treeSvg___AATo{display:block;flex-shrink:0}.Runs_treeNodeCard__Y5XEX{width:100%;height:100%;box-sizing:border-box;padding:.5rem .6rem;background:#0d0d0d;border:1px solid rgba(255,255,255,.08);cursor:pointer;text-align:left;font-family:Geist,ui-sans-serif,system-ui,-apple-system,sans-serif;color:#fafafa;transition:border-color .2s,box-shadow .2s,background .2s;display:flex;flex-direction:column;overflow:hidden}.Runs_treeNodeCard__Y5XEX:hover{border-color:rgba(255,255,255,.2);background:#131313}.Runs_treeNodeSelected__QaPg2{border-color:rgba(0,212,170,.5);box-shadow:0 0 12px rgba(0,212,170,.1);background:#0b1210}.Runs_treeNodeBest__U8FaL{box-shadow:0 0 0 1px rgba(0,212,170,.3),0 0 16px rgba(0,212,170,.08)}@keyframes Runs_nodeHighlight__Zuhi5{0%{border-color:rgba(91,141,239,1);box-shadow:0 0 24px rgba(91,141,239,.4),inset 0 0 12px rgba(91,141,239,.05)}70%{border-color:rgba(91,141,239,.6);box-shadow:0 0 16px rgba(91,141,239,.2)}to{border-color:rgba(255,255,255,.08);box-shadow:none}}.Runs_treeNodeHighlight__A7a9x{animation:Runs_nodeHighlight__Zuhi5 5s ease-out forwards}.Runs_treeNodeName__XtYDH{font-size:.65rem;font-weight:400;color:#fff;overflow:hidden;display:-webkit-box;-webkit-line-clamp:2;-webkit-box-orient:vertical;line-height:1.3;margin-bottom:.15rem}.Runs_treeNodeBadgeRow__ZwAvw{display:flex;align-items:center;gap:.3rem;margin-bottom:.15rem}.Runs_treeNodeBadge__jLVIw{display:inline-block;padding:.1rem .3rem;font-size:.5rem;font-weight:500;text-transform:uppercase;letter-spacing:.5px;flex-shrink:0}.Runs_treeNodeElapsed__vsyq2{font-size:.5rem;color:var(--text-tertiary);font-feature-settings:"tnum";font-variant-numeric:tabular-nums}.Runs_treeNodeScoreRow__BN81e{margin-bottom:.25rem;min-height:1.2rem;display:flex;align-items:center}.Runs_treeNodeIssueRow__7fXg_{min-width:0;margin-top:auto}.Runs_treeNodeIssueRow__7fXg_ .Runs_linearIssueLink__aVa6L{width:100%;min-height:22px;font-size:.58rem;padding:0 .28rem}.Runs_treeNodeIssueRow__7fXg_ .Runs_linearIssueLink__aVa6L img{width:12px;height:12px}.Runs_treeNodeAccuracy__ehSVZ{font-size:1rem;font-weight:600;font-feature-settings:"tnum";font-variant-numeric:tabular-nums;color:#00d4aa;letter-spacing:-.3px}.Runs_treeNodeItems__4SRsq{font-size:.6rem;color:var(--text-tertiary);font-feature-settings:"tnum";font-variant-numeric:tabular-nums;margin-left:.3rem}.Runs_treeNodeLoading__P7_CS{display:flex;align-items:center;gap:.4rem}@keyframes Runs_pulse__LMgDF{0%,to{opacity:.3}50%{opacity:1}}.Runs_treeNodePulse__nC6rt{width:6px;height:6px;border-radius:0;background:#f5a623;animation:Runs_pulse__LMgDF 1.5s ease-in-out infinite}.Runs_treeNodeLoadingText__C1_bJ{font-size:.7rem;color:var(--text-tertiary);font-weight:400}@keyframes Runs_slideInFromRight__uHSvO{0%{transform:translateX(100%)}to{transform:translateX(0)}}@keyframes Runs_fadeIn__1nqka{0%{opacity:0}to{opacity:1}}.Runs_detailPanelBackdrop__UMPlF{position:fixed;inset:0;background:rgba(0,0,0,.5);z-index:99;animation:Runs_fadeIn__1nqka .15s ease-out}.Runs_detailPanel__BII9F{position:fixed;top:0;right:0;width:50vw;max-width:90vw;height:100vh;background:#0f0f0f;border-left:1px solid rgba(255,255,255,.08);overflow-y:auto;z-index:100;padding:2rem;animation:Runs_slideInFromRight__uHSvO .2s ease-out}.Runs_detailPanelClose__vgQbo{position:absolute;top:1rem;right:1rem;background:none;border:1px solid rgba(255,255,255,.08);color:var(--text-secondary);font-size:1.25rem;width:32px;height:32px;display:flex;align-items:center;justify-content:center;cursor:pointer;transition:color .15s,border-color .15s;font-family:inherit}.Runs_detailPanelClose__vgQbo:hover{color:#fff;border-color:rgba(255,255,255,.2)}.Runs_detailPanelHeader__XMEDA{display:flex;align-items:center;gap:.5rem;margin-bottom:1.25rem;padding-right:2.5rem}.Runs_detailPanelTitle__vAuOY{font-size:1.3rem;font-weight:300;letter-spacing:-.3px;margin:0;color:#fff}.Leaderboard_tableWrapper__6uunR{overflow-x:auto;margin:1.5rem 0;border-radius:0;border:1px solid rgba(255,255,255,.08)}.Leaderboard_table__8gTqK{width:100%;border-collapse:collapse;font-size:.82rem;font-family:Geist,ui-sans-serif,system-ui,sans-serif}.Leaderboard_noData__aKd3L{color:rgba(255,255,255,.3);font-style:italic;text-align:center;padding:2rem 0}.Leaderboard_table__8gTqK thead{background:rgba(255,255,255,.03);border-bottom:1px solid rgba(255,255,255,.1);position:sticky;top:0;z-index:1}.Leaderboard_table__8gTqK th{padding:.6rem .75rem;text-align:right;color:rgba(255,255,255,.45);font-weight:500;font-size:.72rem;text-transform:uppercase;letter-spacing:.5px;white-space:nowrap;-webkit-user-select:none;user-select:none}.Leaderboard_thRank__e16FK{width:2rem;text-align:center!important}.Leaderboard_thName__i_zQZ{text-align:left!important;min-width:180px}.Leaderboard_thStatus__F4ZyW{width:2.5rem;text-align:center!important}.Leaderboard_thMetric__ZMVo_{cursor:pointer;transition:color .15s ease;min-width:5rem}.Leaderboard_thMetric__ZMVo_:hover{color:rgba(255,255,255,.8)}.Leaderboard_thPrimary__SzN76{color:rgba(0,212,170,.7)}.Leaderboard_thPrimary__SzN76:hover{color:rgba(0,212,170,1)}.Leaderboard_thSorted__13fmP{color:rgba(255,255,255,.9)!important}.Leaderboard_thPrimary__SzN76.Leaderboard_thSorted__13fmP{color:rgba(0,212,170,1)!important}.Leaderboard_thLabel__2cqYD{margin-right:.3rem}.Leaderboard_sortArrow__3IJre{font-size:.55rem;vertical-align:middle;opacity:.8}.Leaderboard_thItems__3MS6a{width:3rem;text-align:right}.Leaderboard_thStarted__9zNIp{width:11rem;text-align:left!important}.Leaderboard_row__A9H8t{border-bottom:1px solid rgba(255,255,255,.04);cursor:pointer;transition:background-color .15s ease;will-change:transform}.Leaderboard_row__A9H8t:hover{background:rgba(255,255,255,.04)}.Leaderboard_rowBest__JCqYu{background:rgba(0,212,170,.04)}.Leaderboard_rowBest__JCqYu:hover{background:rgba(0,212,170,.08)}.Leaderboard_rowSelected__GFyz5{background:rgba(0,120,255,.08)!important;box-shadow:inset 3px 0 0 rgba(0,120,255,.6)}.Leaderboard_table__8gTqK td{padding:.55rem .75rem;text-align:right;color:rgba(255,255,255,.75);font-feature-settings:"tnum";font-variant-numeric:tabular-nums}.Leaderboard_tdRank__kZKT_{text-align:center!important;color:rgba(255,255,255,.3);font-size:.72rem}.Leaderboard_tdName__2ymmx{text-align:left!important;display:flex;align-items:center;gap:.5rem}.Leaderboard_expName__6rkrZ{color:rgba(255,255,255,.9);font-weight:450}.Leaderboard_bestBadge__33jMs{font-size:.55rem;font-weight:600;text-transform:uppercase;letter-spacing:.5px;color:#0a0a0a;background:rgba(0,212,170,.85);padding:.1rem .35rem;border-radius:0;flex-shrink:0}.Leaderboard_tdStatus__zRhgm{text-align:center!important}.Leaderboard_statusDot__hEPYp{display:inline-block;width:7px;height:7px;border-radius:0}.Leaderboard_tdMetric__Ll0O_{font-family:Geist Mono,ui-monospace,monospace;font-size:.8rem}.Leaderboard_tdPrimary__SZr_5{color:rgba(0,212,170,.9);font-weight:500}.Leaderboard_metricEmpty__l1hwN{color:rgba(255,255,255,.15)}.Leaderboard_tdItems__4_5Q6{color:rgba(255,255,255,.3);font-size:.72rem}.Leaderboard_tdStarted__p97S3{color:rgba(255,255,255,.45);font-size:.72rem;text-align:left!important;white-space:nowrap}
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
self.__BUILD_MANIFEST=function(s,e,t,a,c,r){return{__rewrites:{afterFiles:[],beforeFiles:[],fallback:[]},__routerFilterStatic:{numItems:0,errorRate:1e-4,numBits:0,numHashes:null,bitArray:[]},__routerFilterDynamic:{numItems:0,errorRate:1e-4,numBits:0,numHashes:null,bitArray:[]},"/":[s,e,t,"static/chunks/pages/index-1556edd8356dd19f.js"],"/_error":["static/chunks/pages/_error-b69380d1599ed4ba.js"],"/[system]":["static/chunks/pages/[system]-adf7b3ca903bcc42.js"],"/[system]/benchmarks":[s,e,t,"static/chunks/pages/[system]/benchmarks-ea3ad9fe4e28dd88.js"],"/[system]/data":[s,e,t,"static/chunks/pages/[system]/data-57686b9546f2794a.js"],"/[system]/eval":[s,e,t,"static/chunks/pages/[system]/eval-d9b5f1b8db0f0f90.js"],"/[system]/experiments":[s,e,t,"static/chunks/pages/[system]/experiments-4d2122d6ada9a04a.js"],"/[system]/ideas":[s,e,t,"static/chunks/pages/[system]/ideas-6c1ff7f9e0da750b.js"],sortedPages:["/","/_app","/_error","/[system]","/[system]/benchmarks","/[system]/data","/[system]/eval","/[system]/experiments","/[system]/ideas"]}}("static/chunks/431-43358ce3c29e5e1b.js","static/css/e75cf1946c214544.css","static/chunks/374-421036d63d323cc9.js",0,0,0),self.__BUILD_MANIFEST_CB&&self.__BUILD_MANIFEST_CB();
|
package/dist/lib/env.js
DELETED
package/dist/shared/env.js
DELETED
|
@@ -1,51 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
name: variant-builder
|
|
3
|
-
description: Implement and evaluate a single variant for a system
|
|
4
|
-
---
|
|
5
|
-
|
|
6
|
-
You implement and evaluate a single variant. You are a single-shot agent: implement, record one run via `kaizen run`, exit.
|
|
7
|
-
|
|
8
|
-
## Working directory
|
|
9
|
-
|
|
10
|
-
You receive a worktree path created by the orchestrator. All work happens in that worktree. Do not modify the main repo checkout.
|
|
11
|
-
You also receive the canonical Kaizen state directory for the main checkout. Export it before running Kaizen so Studio sees the run even though evaluation happens from the worktree.
|
|
12
|
-
|
|
13
|
-
```bash
|
|
14
|
-
cd <worktree_path>
|
|
15
|
-
export KAIZEN_STATE_DIR=<main_repo_path>/.kaizen
|
|
16
|
-
```
|
|
17
|
-
|
|
18
|
-
## Setup
|
|
19
|
-
|
|
20
|
-
1. Read the system definition at the absolute path in your prompt. The frontmatter has `run_eval`, `primary_metric`, `target`, `execution_mode`.
|
|
21
|
-
2. Read the parent run's `manifest.json` (if you have a parent) for context — its hypothesis, score, and `failures.jsonl` tells you what to fix.
|
|
22
|
-
3. For `server` execution_mode, start the system's servers on the assigned ports given in your prompt. For `in_process`, install deps as the system's setup section says.
|
|
23
|
-
|
|
24
|
-
## Implement
|
|
25
|
-
|
|
26
|
-
Modify code in your worktree to implement the variant described in your prompt. If you have a parent, the parent's code is already present (your worktree branched from theirs); build on it.
|
|
27
|
-
|
|
28
|
-
## Run
|
|
29
|
-
|
|
30
|
-
```bash
|
|
31
|
-
kaizen run \
|
|
32
|
-
--system <system_id> \
|
|
33
|
-
--variant <variant_id> \
|
|
34
|
-
--parent <parent_or_omit> \
|
|
35
|
-
--hypothesis "<one-line hypothesis>" \
|
|
36
|
-
--state-dir "$KAIZEN_STATE_DIR"
|
|
37
|
-
```
|
|
38
|
-
|
|
39
|
-
The `kaizen run` runner is plain code, not an agent. It handles everything: writes manifest, tails NDJSON events into `events.jsonl`, atomically updates `state.json`, dumps worst items to `failures.jsonl`, decides promotion via paired-bootstrap statistical test, prints the score.
|
|
40
|
-
|
|
41
|
-
The eval script may also write Langfuse dataset-run links and trace scores. Treat those writes as diagnostic persistence only. The authoritative run result for this loop is still the `complete.score` that `kaizen run` records from the NDJSON stream.
|
|
42
|
-
|
|
43
|
-
You do **not** write `.kaizen/` files. You do **not** call any status CLI. The single `kaizen run` invocation is the whole interaction.
|
|
44
|
-
|
|
45
|
-
## After the run
|
|
46
|
-
|
|
47
|
-
- `kaizen run` prints `score=<n> run_id=<id> status=<complete|crashed|aborted> promoted=<bool>` and exits with code 0 (complete) or non-zero (crashed/aborted).
|
|
48
|
-
- If complete, commit your code changes (`git add -A && git commit -m "run: <variant_id>"`) so child variants can branch from your branch.
|
|
49
|
-
- If crashed, the runner has already recorded the crash event with stderr. Read `.kaizen/runs/<system>/<run_id>/state.json` for diagnostics.
|
|
50
|
-
|
|
51
|
-
Kill any servers you started. The orchestrator will clean up the worktree.
|
|
@@ -1,65 +0,0 @@
|
|
|
1
|
-
---
|
|
2
|
-
description: Drive a run loop over a system using Kaizen
|
|
3
|
-
---
|
|
4
|
-
|
|
5
|
-
You are Kaizen, an automated AI researcher. Improve AI systems through iterative runs recorded under `.kaizen/runs/`.
|
|
6
|
-
|
|
7
|
-
## Critical rules
|
|
8
|
-
|
|
9
|
-
- **NEVER commit PHI** to this repo. PHI lives in Langfuse and production databases only.
|
|
10
|
-
- **NEVER hardcode credentials.** Read keys from environment variables.
|
|
11
|
-
|
|
12
|
-
## System selection
|
|
13
|
-
|
|
14
|
-
If the user passed a system name (`/kaizen <name>`), read `systems/<name>.md`. Otherwise glob `systems/*.md` and ask which one.
|
|
15
|
-
|
|
16
|
-
The system definition's frontmatter is the contract:
|
|
17
|
-
- `run_eval` — path to the eval script
|
|
18
|
-
- `eval_version`, `dataset_version` — bump = old scores incomparable
|
|
19
|
-
- `primary_metric`, `target` — what we are optimizing
|
|
20
|
-
- `execution_mode` — `in_process` or `server`
|
|
21
|
-
|
|
22
|
-
## State model
|
|
23
|
-
|
|
24
|
-
Truth lives on disk under `.kaizen/runs/<system>/<run_id>/`:
|
|
25
|
-
- `manifest.json` — immutable record (variant, parent, hypothesis, git_sha, versions)
|
|
26
|
-
- `events.jsonl` — append-only event stream from the eval (the source of truth for that run)
|
|
27
|
-
- `state.json` — derived cache (status, score, progress) — runner-owned, never edit by hand
|
|
28
|
-
- `failures.jsonl` — worst-K item events, for failure analysis
|
|
29
|
-
|
|
30
|
-
The hypothesis log is `.kaizen/hypotheses/<system>.jsonl` — every run leaves a line, success or failure.
|
|
31
|
-
The promoted baseline is derived on demand: the latest run that beat the previous promoted baseline with statistical confidence under matching `eval_version` / `dataset_version`. The leaderboard still shows the best raw score. `kaizen log --system <s>` shows the promoted baseline in the header alongside recent runs.
|
|
32
|
-
|
|
33
|
-
## Eval script contract
|
|
34
|
-
|
|
35
|
-
The system's `run_eval` script is the single eval entrypoint. `kaizen run` invokes it as:
|
|
36
|
-
|
|
37
|
-
```bash
|
|
38
|
-
<run_eval> --variant <variant_id> --dataset <dataset_version> --out-fd 3 [--max-items <n>]
|
|
39
|
-
```
|
|
40
|
-
|
|
41
|
-
The required contract is the NDJSON stream on `--out-fd`: emit `start`, one `item` per dataset item, and one terminal `complete` with a numeric `score` in `[0,1]`. The `complete.score` is what Kaizen records, compares, and returns to you in the CLI summary.
|
|
42
|
-
|
|
43
|
-
For real customer systems, the eval script should also persist results back to Langfuse as a best-effort side effect:
|
|
44
|
-
|
|
45
|
-
- Load the versioned Langfuse dataset named by `--dataset`.
|
|
46
|
-
- Run the candidate system for each dataset item, producing a fresh Langfuse trace for that item.
|
|
47
|
-
- Link each dataset item to that fresh trace in a Langfuse dataset run named for the Kaizen experiment or variant.
|
|
48
|
-
- Write the primary metric as a Langfuse score on the fresh trace, with useful secondary metrics in score metadata or dataset-run metadata.
|
|
49
|
-
- Include the fresh trace id in each Kaizen `item.trace_id` and in `complete.worst_traces`.
|
|
50
|
-
|
|
51
|
-
Langfuse writes make traces, dataset runs, and scores durable for inspection. They must not replace the NDJSON stream; `.kaizen/runs/` remains Kaizen's local source of truth for experiment state and promotion.
|
|
52
|
-
|
|
53
|
-
## Loop
|
|
54
|
-
|
|
55
|
-
1. **Read state** — `kaizen log --system <s> -n 10` gives both the promoted baseline (header) and recent runs (body) in one call.
|
|
56
|
-
2. **Propose variants** — for each, write a one-line hypothesis and a 2-3 sentence description.
|
|
57
|
-
3. **Record runs** — spawn `variant-builder` subagents, each in its own git worktree. Pass each one the main checkout's absolute `.kaizen` path and require `KAIZEN_STATE_DIR=<main>/.kaizen` (or `--state-dir`) on `kaizen run`.
|
|
58
|
-
4. **Read results** — every run prints `score=<n> run_id=<id> status=<...> promoted=<bool>` on stdout. The runner handles all state writes; do not write `.kaizen/` files yourself.
|
|
59
|
-
5. **Promotion is automatic** — if the variant beats the promoted baseline with statistical significance (95% CI lower bound on the per-item delta > 0, no subgroup regressions), `kaizen run` promotes it. Promotion means "new baseline for future runs," not "open a PR." You will see `promoted=true` in the output.
|
|
60
|
-
6. **Iterate** — repeat. The next call to `kaizen log` will show the new baseline.
|
|
61
|
-
|
|
62
|
-
## What you do vs what the runner does
|
|
63
|
-
|
|
64
|
-
- You write code, generate variants, analyze failures, decide what to try next, and prepare a PR from the latest promoted baseline when a human asks.
|
|
65
|
-
- The `kaizen run` runner (a plain Node program, not an agent) writes all run state atomically and decides promotion via paired-bootstrap statistical test. If you find yourself editing `manifest.json`, `events.jsonl`, `state.json`, `failures.jsonl`, or the hypothesis log by hand, stop — that is a bug.
|