dual-brain 3.8.1 → 4.0.0

package/README.md

@@ -63,7 +63,7 @@ npx -y dual-brain
 | `hooks/gpt-work-dispatcher.mjs` | Dispatch execution tasks to GPT via Codex CLI |
 | `hooks/session-report.mjs` | Session-end summary: activity, compliance, quality |
 | `hooks/health-check.mjs` | Verify all hooks and dependencies are working |
-| `hooks/test-orchestrator.mjs` | Self-test harness (29 tests) |
+| `hooks/test-orchestrator.mjs` | Self-test harness (39 tests) |
 | `hooks/setup-wizard.mjs` | Interactive config (optional — for custom plans) |
 | `hooks/install-git-hooks.mjs` | Git pre-commit hook for quality gate |
 

package/hooks/control-panel.mjs

@@ -437,16 +437,40 @@ function showProfilePicker(rl) {
 
     rl.question('  Choice: ', (answer) => {
       const names = Object.keys(PROFILES);
-      const idx = parseInt(answer, 10) - 1;
+      const trimmed = answer.trim();
+      let selectedName = null;
+
+      // Try numeric selection first
+      const idx = parseInt(trimmed, 10) - 1;
       if (idx >= 0 && idx < names.length) {
+        selectedName = names[idx];
+      }
+
+      // Try natural language alias resolution
+      if (!selectedName && trimmed && trimmed !== 'q') {
+        const PANEL_ALIASES = {
+          'auto': 'auto', 'adaptive': 'auto', 'smart': 'auto', 'default': 'auto', 'normal': 'auto',
+          'balanced': 'balanced', 'even': 'balanced', 'equal': 'balanced',
+          'cost-saver': 'cost-saver', 'cheap': 'cost-saver', 'save': 'cost-saver', 'conservative': 'cost-saver', 'frugal': 'cost-saver', 'budget': 'cost-saver',
+          'quality-first': 'quality-first', 'aggressive': 'quality-first', 'quality': 'quality-first', 'max': 'quality-first', 'full': 'quality-first', 'both': 'quality-first',
+        };
+        const cleaned = trimmed.toLowerCase()
+          .replace(/^(go|be|use|switch to|set|mode)\s+/i, '')
+          .replace(/\s+mode$/i, '');
+        selectedName = PANEL_ALIASES[cleaned] || null;
+      }
+
+      if (selectedName) {
         let customOverrides = null;
         try {
           const existing = JSON.parse(readFileSync(PROFILE_FILE, 'utf8'));
           if (existing.custom_overrides?.budgets) customOverrides = { budgets: existing.custom_overrides.budgets };
         } catch {}
-        saveProfile(names[idx], customOverrides);
-        const pf = PROFILES[names[idx]];
+        saveProfile(selectedName, customOverrides);
+        const pf = PROFILES[selectedName];
         console.log(`  ✅ Switched to ${pf.emoji}  ${pf.uiLabel}`);
+      } else if (trimmed && trimmed !== 'q') {
+        console.log(`  Unknown profile: ${trimmed}. Try: cheap, aggressive, quality, balanced, auto`);
       }
       resolve();
     });

package/hooks/cost-logger.mjs

@@ -8,7 +8,6 @@
  * Output contract: must print "{}" to stdout and exit 0 within ~100 ms.
  */
 
-import { createHash } from "crypto";
 import { appendFileSync, mkdirSync, readFileSync, writeFileSync } from "fs";
 import { dirname, join } from "path";
 import { fileURLToPath } from "url";
@@ -265,8 +264,8 @@ async function main() {
   // Record failures for adaptive routing (failure-loop detection)
   if (status === 'error' && toolName === 'Agent') {
     try {
-      const { recordFailure, pruneOldFailures } = await import('./failure-detector.mjs');
-      const promptHash = createHash('md5').update(JSON.stringify(toolInput)).digest('hex').slice(0, 12);
+      const { computePromptHash, recordFailure, pruneOldFailures } = await import('./failure-detector.mjs');
+      const promptHash = computePromptHash(toolInput);
       recordFailure(promptHash, tier, payload?.error || 'agent_error');
       // Best-effort cleanup of stale failure entries (>24h old)
       try { pruneOldFailures(); } catch {}

package/hooks/enforce-tier.mjs

@@ -1,10 +1,9 @@
 #!/usr/bin/env node
 import { readFileSync, writeFileSync, appendFileSync, renameSync } from 'fs';
-import { createHash } from 'crypto';
 import { dirname, resolve, join } from 'path';
 import { fileURLToPath } from 'url';
 import { classifyRisk, extractPaths } from './risk-classifier.mjs';
-import { checkFailureLoop, recordFailure } from './failure-detector.mjs';
+import { computePromptHash, checkFailureLoop, recordFailure } from './failure-detector.mjs';
 
 const __dirname = dirname(fileURLToPath(import.meta.url));
 const CONFIG_FILE = resolve(__dirname, '..', 'orchestrator.json');
@@ -214,7 +213,7 @@ try {
   const currentModel = (ti.model || '').toLowerCase();
 
   // Compute prompt hash early for duplicate detection and logging
-  const promptHash = createHash('sha256').update(text).digest('hex').slice(0, 12);
+  const promptHash = computePromptHash(ti);
 
   // Burst detection — suppress noise during wave launches (3+ agents in 90s)
   const burstMode = detectBurst();
@@ -227,11 +226,11 @@ try {
     if (burstMode) {
       // In burst mode, only warn on exact hash matches (same description+prompt)
       if (duplicate.prompt_hash === promptHash) {
-        duplicateWarning = `**[Wave] [Duplicate Warning]** A similar agent task was dispatched ${minutesAgo} minute${minutesAgo !== 1 ? 's' : ''} ago. Reuse the prior result unless the scope changed.`;
+        duplicateWarning = `Heads up — a similar task ran ${minutesAgo} minute${minutesAgo !== 1 ? 's' : ''} ago (wave detected). Reuse that result if the scope hasn't changed.`;
       }
       // Otherwise suppress — similar-but-different agents in a wave are expected
     } else {
-      duplicateWarning = `**[Duplicate Warning]** A similar agent task was dispatched ${minutesAgo} minute${minutesAgo !== 1 ? 's' : ''} ago. Reuse the prior result unless the scope changed.`;
+      duplicateWarning = `Heads up — a similar task ran ${minutesAgo} minute${minutesAgo !== 1 ? 's' : ''} ago. Reuse that result if the scope hasn't changed.`;
     }
   }
 
@@ -279,10 +278,10 @@ try {
     ].filter(Boolean);
 
     if (detectedTiers.length > 1) {
-      const splitMsg = `**[Tier Enforcer]** This spans **${detectedTiers.join(' + ')}** work. Consider splitting: ` +
+      const splitMsg = `This spans ${detectedTiers.join(' + ')} work. Consider splitting: ` +
         (hasSearch ? 'search first (haiku), ' : '') +
         (hasExecute ? 'then execute edits (sonnet), ' : '') +
-        (hasThink ? 'keep planning/review on think tier (opus).' : '');
+        (hasThink ? 'keep planning/review on the main session (opus).' : '');
       const fullMsg = prependWarnings(splitMsg.replace(/, $/, '.'));
       logRecommendation({
         tier: detectedTiers.join('+'),
@@ -311,21 +310,21 @@ try {
   if ((riskResult.level === 'critical' || riskResult.level === 'high') && tier !== 'think') {
     tier = 'think';
     autoStatus = riskResult.level === 'critical'
-      ? `Dual-brain: dual-brain review recommended — ${riskResult.reason.split(':')[0]} detected`
-      : `Dual-brain: promoting to think tier — ${riskResult.reason.split(':')[0]}`;
+      ? `This touches ${riskResult.reason.split(':')[0].toLowerCase()} — recommending dual-brain review for safety.`
+      : `Promoting to think tier — this is ${riskResult.reason.split(':')[0].toLowerCase()}.`;
   }
 
   // Failure loop detection
-  const failureCheck = checkFailureLoop(promptHash, tier);
+  const failureCheck = checkFailureLoop(promptHash);
   let failureMessage = null;
   if (failureCheck.isLoop) {
     if (failureCheck.suggestion === 'promote_tier' && tier === 'execute') {
       tier = 'think';
-      autoStatus = 'Dual-brain: escalating to think tier — previous attempt failed';
+      autoStatus = 'Escalating to think tier — this has failed before, let\'s take a different approach.';
     } else if (failureCheck.suggestion === 'escalate_to_dual_brain') {
-      autoStatus = 'Dual-brain: dual-brain review recommended — repeated failures detected';
+      autoStatus = 'Repeated failures detected — recommending dual-brain review to diagnose the issue.';
     }
-    failureMessage = `**[Failure Loop]** ${failureCheck.count} failed attempts in 2hrs. Consider: \`node .claude/hooks/dual-brain-think.mjs --question "why is this failing?"\``;
+    failureMessage = `⚠️ This has failed ${failureCheck.count} times in the last 2 hours. Consider a dual-brain think session to diagnose the root cause.`;
   }
 
   // Apply profile-driven tier adjustments
@@ -345,7 +344,7 @@ try {
       const biasThreshold = profileSettings.bias >= 0 ? 10 : 20;
       if (balance && balance.claudeCalls > balance.openaiCalls * 2 && balance.claudeCalls > biasThreshold) {
         const dispatchModel = tier === 'think' ? 'gpt-5.5' : tier === 'execute' ? 'gpt-5.4' : 'gpt-4.1-mini';
-        balanceHint = `\n\n💡 **Balance tip:** Claude has ${balance.claudeCalls} ${tier} calls vs OpenAI's ${balance.openaiCalls} in the last 5hrs. Consider dispatching isolated work to GPT: \`node .claude/hooks/gpt-work-dispatcher.mjs --task "..." --model ${dispatchModel}\``;
+        balanceHint = `\n\n💡 Claude is handling most work right now (${balance.claudeCalls} ${tier} calls vs ${balance.openaiCalls} GPT). For isolated tasks, consider routing to GPT to balance subscriptions.`;
       }
     }
   }
@@ -375,8 +374,7 @@ try {
     // If we get here, a non-think model is being used for think work
     const thinkBestFor = intelligence[expected || 'opus']?.best_for;
     const thinkBestForSuffix = thinkBestFor ? ` (best for: ${thinkBestFor})` : '';
-    const msg = `**[Tier Enforcer]** This looks like **think** work (architecture/review/planning). ` +
-      `Don't send it to "${currentModel}" — keep it on the main session (${expected || 'opus'}${thinkBestForSuffix}) for best results.`;
+    const msg = `This looks like think-level work (architecture/review/planning) — better kept on the main session (${expected || 'opus'}${thinkBestForSuffix}) rather than delegated to ${currentModel}.`;
     logRecommendation({
       tier,
       recommended: expected,
@@ -407,8 +405,7 @@ try {
     const savings = tier === 'search' ? 'Haiku is 19x cheaper than Opus for read-only lookups.' : 'Sonnet is 5x cheaper than Opus for implementation work.';
     const bestFor = intelligence[expected]?.best_for;
     const bestForSuffix = bestFor ? ` (best for: ${bestFor})` : '';
-    const msg = `**[Tier Enforcer]** This looks like **${tier}** work. ` +
-      `Use \`model: "${expected}"\`${bestForSuffix} instead of "${currentModel || 'opus (inherited)'}". ${savings}`;
+    const msg = `This looks like ${tier} work — use ${expected}${bestForSuffix} instead of ${currentModel || 'opus (inherited)'}. ${savings}`;
     logRecommendation({
       tier,
       recommended: expected,

package/hooks/failure-detector.mjs

@@ -8,6 +8,7 @@
  *   pruneOldFailures() → { pruned, remaining }
  */
 
+import { createHash } from 'crypto';
 import { readFileSync, appendFileSync, writeFileSync, renameSync, unlinkSync } from 'fs';
 import { dirname, join } from 'path';
 import { fileURLToPath } from 'url';
@@ -16,6 +17,19 @@ import { fileURLToPath } from 'url';
 const __dirname = dirname(fileURLToPath(import.meta.url));
 const LEDGER_FILE = join(__dirname, 'decision-ledger.jsonl');
 
+/**
+ * Canonical prompt hash used by all hooks for failure-loop correlation.
+ * Both enforce-tier (PreToolUse) and cost-logger (PostToolUse) must use this
+ * same function so that recorded failures can be matched during escalation.
+ *
+ * @param {object} toolInput — the raw tool_input from the hook payload
+ * @returns {string} 12-char hex hash
+ */
+function computePromptHash(toolInput) {
+  const text = (toolInput?.description || '') + (toolInput?.prompt || '');
+  return createHash('sha256').update(text).digest('hex').slice(0, 12);
+}
+
 /**
  * Compute a decay weight based on failure age.
  * 0-30 min → 1.0, 30-60 min → 0.5, 60-120 min → 0.25, >120 min → 0 (excluded by window)
@@ -121,4 +135,4 @@ function pruneOldFailures() {
   return { pruned, remaining };
 }
 
-export { checkFailureLoop, recordFailure, pruneOldFailures };
+export { computePromptHash, checkFailureLoop, recordFailure, pruneOldFailures };