Restore original native instruction comparison for Hard Kinex tasks

hard/app.js

@@ -201,7 +201,7 @@ function instructionPair(task, ep) {
   const record = candidate && candidate.initial_goal === ep.instruction && candidate.task_key === task.key ? candidate : null;
   const recordedNative = record?.observed_native || (ep ? ep.native_instruction : task.native_instruction);
   const measured = record?.text || (ep ? (ep.instruction || (ep.instruction_policy === 'original_native' ? recordedNative : null)) : null);
-  const reference = catalogInstructionAudit.tasks.find(item => item.key === task.key && item.native_id === task.native_id);
+  const reference = catalogInstructionAudit.tasks.find(item => item.published_task_key === task.key && item.id === task.native_id);
   const nativeReference = !recordedNative && reference?.native ? reference : null;
   const native = recordedNative || nativeReference?.native;
   return {record, native, measured, nativeReference};
@@ -216,7 +216,7 @@ function instructionPanel(task,ep) {
   const choices=[['measured',ep?'Measured':'Planned',measured],['native','Original native',native],['current','Current catalog',null]].filter(([key,,text])=>text&&(key!=='current'||(ep&&text!==measured&&text!==initial)));
   const currentChanged=native&&task.catalog_instruction&&native!==task.catalog_instruction;
   const description=!ep?'The current catalog instruction for the planned evaluation.':record?.kind==='scene_bound_goal'?"The measured goal with this episode's objects and targets filled in from the agent's session.":record?.kind==='native_received'?'The native instruction received by the agent during this episode.':'The task instruction supplied for this episode.';
-  return `<section class="panel instruction-panel"><div class="section-heading"><div><h2>Task instruction</h2><p>${description}</p></div><div class="segmented" aria-label="Instruction version">${choices.map(([k,label],i)=>`<button data-instruction="${k}" aria-pressed="${i===0}">${label}</button>`).join('')}</div></div>${nativeReference?`<p class="instruction-diff-note" data-native-reference><span>Native comparison uses a pinned upstream reference${ep?'; native text was not recorded in this episode':''}. ${external(nativeReference.source_url,'Native source')}</span></p>`:''}${choices.map(([k,,text],i)=>{
+  return `<section class="panel instruction-panel"><div class="section-heading"><div><h2>Task instruction</h2><p>${description}</p></div><div class="segmented" aria-label="Instruction version">${choices.map(([k,label],i)=>`<button data-instruction="${k}" aria-pressed="${i===0}">${label}</button>`).join('')}</div></div>${nativeReference?`<p class="instruction-diff-note" data-native-reference><span>Native comparison uses the pinned upstream full_description reference${ep?'; native text was not recorded in this episode':''}. ${external(nativeReference.source_url,'Native source')}</span></p>`:''}${choices.map(([k,,text],i)=>{
     const comparison=k==='current'?(currentChanged?instructionDiff(native,text):null):diff;
     const html=comparison?(k==='native'?comparison.original:comparison.revised):esc(text);
     const note=comparison?(k==='native'?'Struck-through text was removed or replaced in the '+(ep?'measured':'planned')+' instruction.':'Highlighted text was added or replaced relative to the original native instruction.'):(native?'No changes from the original native instruction.':'Original native text is unavailable for comparison.');
@@ -506,7 +506,7 @@ const sourceConfig=await source.json();evidenceRoot=sourceConfig.evidence_root;r
 document.querySelectorAll('[data-result-file]').forEach(link=>{link.href=resultURL(link.dataset.resultFile);});
 const r=await fetch(resultURL('data.json'));if(!r.ok)throw Error('HTTP '+r.status);data=resolveEvidence(await r.json());
 const ir=await fetch(resultURL('measured-instructions.json'));if(!ir.ok)throw Error('Instruction evidence unavailable');measuredInstructions=resolveEvidence(await ir.json());
-const nr=await fetch(resultURL('instruction-audit.json'));if(!nr.ok)throw Error('Instruction audit unavailable');catalogInstructionAudit=resolveEvidence(await nr.json());
+const nr=await fetch('../instruction-audit/hard.json');if(!nr.ok)throw Error('Instruction audit unavailable');catalogInstructionAudit=await nr.json();
 try{const c=await fetch(resultURL('cost-data.json'));if(c.ok)UsageMetrics.init(resolveEvidence(await c.json()));}catch{UsageMetrics.init({episodes:{},sources:[]});}
 for(const task of data.tasks){
   if(task.family==='task02'){

hard/index.html

@@ -10,7 +10,7 @@
 <link rel="stylesheet" href="style.css?v=bd1a59147e1d">
 <script src="usage-metrics.js?v=330e92f8eb96" defer></script>
 <script src="../instruction-audit.js?v=7b5e9a418dec" defer></script>
-<script src="app.js?v=e29b7c36fe28" defer></script>
+<script src="app.js?v=6fde7288fba9" defer></script>
 <link rel="stylesheet" href="policy.css">
 <link rel="stylesheet" href="../instruction-audit.css?v=43163fc879ed">
 </head>

manifest.json

@@ -99,8 +99,8 @@
       "sha256": "a53108594a23b740fbb08623f6fffdda1d73608b552988a9ffda9f0944ae66a1"
     },
     "hard/app.js": {
-      "bytes": 97482,
-      "sha256": "700f94bc25707d08a5fbb25204a47fa6c7a32ade883ef9ad3c370ab07b97317f"
+      "bytes": 97489,
+      "sha256": "6fde7288fba9fc3e8b7990f5fbd8c6b63560888b871bbdd6e113940cb918617c"
     },
     "hard/icon.svg": {
       "bytes": 725,
@@ -112,7 +112,7 @@
     },
     "hard/index.html": {
       "bytes": 3630,
-      "sha256": "c24b627bec54985bdddc7bacc66194771559aa88ac44eb4e7ae7ee8e2f2add76"
+      "sha256": "1488e9f98a843f37801e91464261aa4a76a02531e928193ad3befe57209e31e9"
     },
     "hard/policy.css": {
       "bytes": 1067,