dual-brain 3.2.0 → 3.4.0

package/hooks/dual-brain-review.mjs

@@ -19,7 +19,11 @@ import { fileURLToPath } from 'url';
 
 const __dirname = dirname(fileURLToPath(import.meta.url));
 
-const REVIEW_PROMPT = `Review the current uncommitted changes in this repo for:
+const REVIEW_PROMPT_R1 = `You are GPT-5.5 performing Round 1 of a dual-brain code review.
+Claude (Opus) will independently review the same changes, then send you their findings
+for a collaborative Round 2 discussion.
+
+Review the current uncommitted changes for:
 1. Correctness — logic errors, off-by-one, null/undefined risks
 2. Security — injection, auth bypass, data exposure
 3. Edge cases — what could break under unusual input
@@ -34,6 +38,24 @@ Required output:
 
 Be concise. Flag only real issues, not style preferences. If the code looks good, say "LGTM" and note any minor suggestions. Output your review as plain text, not JSON.`;
 
+const REVIEW_PROMPT_R2 = `You are GPT-5.5 in Round 2 of a collaborative code review with Claude (Opus).
+You already reviewed this diff in Round 1. Claude has now independently reviewed the same changes.
+This is a professional peer review dialogue — two senior engineers refining their assessment together.
+
+Claude's review findings:
+---CLAUDE_REVIEW---
+
+Now respond as a peer reviewer:
+1. CONFIRMED: Issues you both found — these are high-confidence findings
+2. MISSED: Issues Claude caught that you missed — acknowledge them
+3. DISAGREE: Claude's findings you think are false positives — explain why
+4. ESCALATED: Issues that are MORE severe than either of you initially rated
+5. VERDICT: Combined assessment — LGTM, minor issues, or blocks merge
+
+Be direct. If Claude found something real that you missed, say so.
+If Claude flagged something that isn't actually a problem, explain why with evidence.
+The goal is the most accurate review, not defending your initial take.`;
+
 function loadReviewRules() {
   const rulesFile = resolve(__dirname, '..', 'review-rules.md');
   try {
@@ -127,9 +149,9 @@ function exit(obj) {
 
 /**
  * Try GPT review via Codex CLI (uses ChatGPT subscription auth).
- * Returns review text or null if codex isn't available.
+ * Round 1: independent review. Round 2: respond to Claude's review.
  */
-function tryCodexReview(diff) {
+function tryCodexReview(diff, { round = 1, claudeReview = null } = {}) {
   if (!CODEX_BIN) return null;
   try {
     spawnSync(CODEX_BIN, ['login', 'status'], {
@@ -145,7 +167,14 @@ function tryCodexReview(diff) {
       ? diff.slice(0, MAX_DIFF_CHARS) + '\n[truncated]'
       : diff;
 
-    const fullPrompt = REVIEW_PROMPT + loadReviewRules();
+    let basePrompt;
+    if (round === 2 && claudeReview) {
+      basePrompt = REVIEW_PROMPT_R2.replace('---CLAUDE_REVIEW---', claudeReview);
+    } else {
+      basePrompt = REVIEW_PROMPT_R1;
+    }
+    const fullPrompt = basePrompt + loadReviewRules();
+
     const proc = spawnSync(CODEX_BIN, [
       'exec', '--json', '--ephemeral',
       '-c', `model="${model}"`,
@@ -159,7 +188,6 @@ function tryCodexReview(diff) {
     });
     const result = proc.stdout || '';
 
-    // Parse JSONL output, find agent_message items
     const messages = result
       .split('\n')
       .filter(l => l.trim())
@@ -173,16 +201,17 @@ function tryCodexReview(diff) {
     const usage = messages.find(m => m.type === 'turn.completed')?.usage;
 
     if (agentMessages.length > 0) {
+      const reviewText = agentMessages.join('\n\n');
       return {
-        review: agentMessages.join('\n\n'),
+        round,
+        review: reviewText,
         model,
         auth_type: 'codex_subscription',
-        issues_found: hasIssues(agentMessages.join(' ')),
+        issues_found: hasIssues(reviewText),
         tokens: usage || null,
       };
     }
 
-    // Check for errors
     const errors = messages.filter(m => m.type === 'error' || m.type === 'turn.failed');
     if (errors.length > 0) {
       return {
@@ -205,7 +234,7 @@ function tryCodexReview(diff) {
 /**
  * Try GPT review via direct API call (needs OPENAI_API_KEY).
  */
-async function tryApiReview(diff) {
+async function tryApiReview(diff, { round = 1, claudeReview = null } = {}) {
   const apiKey = process.env.OPENAI_API_KEY;
   if (!apiKey) return null;
 
@@ -214,7 +243,14 @@ async function tryApiReview(diff) {
     ? diff.slice(0, MAX_DIFF_CHARS) + '\n[truncated]'
     : diff;
 
-  const fullPrompt = REVIEW_PROMPT + loadReviewRules();
+  let basePrompt;
+  if (round === 2 && claudeReview) {
+    basePrompt = REVIEW_PROMPT_R2.replace('---CLAUDE_REVIEW---', claudeReview);
+  } else {
+    basePrompt = REVIEW_PROMPT_R1;
+  }
+  const fullPrompt = basePrompt + loadReviewRules();
+
   const controller = new AbortController();
   const timer = setTimeout(() => controller.abort(), 30_000);
 
@@ -245,6 +281,7 @@ async function tryApiReview(diff) {
     if (!text) return null;
 
     return {
+      round,
       review: text,
       model,
       auth_type: 'api_key',
@@ -256,7 +293,37 @@ async function tryApiReview(diff) {
   }
 }
 
+function parseArgs(argv) {
+  const args = {};
+  let i = 0;
+  while (i < argv.length) {
+    const arg = argv[i];
+    if (arg.startsWith('--')) {
+      const eqIdx = arg.indexOf('=');
+      if (eqIdx !== -1) {
+        args[arg.slice(2, eqIdx)] = arg.slice(eqIdx + 1);
+      } else {
+        const key = arg.slice(2);
+        const next = argv[i + 1];
+        if (next !== undefined && !next.startsWith('--')) {
+          args[key] = next;
+          i++;
+        } else {
+          args[key] = true;
+        }
+      }
+    }
+    i++;
+  }
+  return args;
+}
+
 async function main() {
+  const args = parseArgs(process.argv.slice(2));
+  const round = args.round ? parseInt(args.round, 10) : 1;
+  const claudeReview = args['claude-review'] || null;
+  const opts = { round, claudeReview };
+
   // 1. Get diff
   let diff = runGit('git diff --staged') || '';
   if (countLines(diff) < MIN_DIFF_LINES) {
@@ -264,12 +331,11 @@ async function main() {
     if (countLines(headDiff) > countLines(diff)) diff = headDiff;
   }
 
-  // Also gather content of untracked source files
   try {
     const untracked = runGit('git ls-files --others --exclude-standard') || '';
     const sourceExts = /\.(ts|tsx|js|jsx|py|rs|go|java|rb|swift|kt|mjs|cjs)$/;
     const untrackedSrc = untracked.split('\n').filter(f => f && sourceExts.test(f));
-    for (const f of untrackedSrc.slice(0, 10)) { // cap at 10 files
+    for (const f of untrackedSrc.slice(0, 10)) {
       const content = runGit(`git diff --no-index /dev/null "${f}"`);
       if (content) diff += '\n' + content;
     }
@@ -279,13 +345,36 @@ async function main() {
     exit({ review: 'No significant changes to review' });
   }
 
-  // 2. Try Codex CLI first (uses ChatGPT subscription)
-  const codexResult = tryCodexReview(diff);
-  if (codexResult) exit(codexResult);
+  // 2. Try Codex CLI first
+  const codexResult = tryCodexReview(diff, opts);
+  if (codexResult) {
+    if (round === 1) {
+      codexResult.instructions = `Round 1 complete. Now:
+1. Provide YOUR independent code review of the same changes
+2. Then call Round 2 to send your findings back to GPT:
+   node .claude/hooks/dual-brain-review.mjs --round 2 --claude-review "<your findings>"
+3. GPT will respond — confirming shared findings, acknowledging misses, and pushing back on false positives
+4. You then synthesize both rounds into the final review verdict`;
+    } else {
+      codexResult.instructions = `GPT has responded to your review. Synthesize into a FINAL REVIEW:
+- CONFIRMED findings (both found) → high confidence, must fix
+- GPT-only findings you agree with → add to your list
+- Your findings GPT disputed → re-evaluate honestly
+- Final verdict: LGTM, minor issues, or blocks merge`;
+    }
+    exit(codexResult);
+  }
 
   // 3. Try direct API
-  const apiResult = await tryApiReview(diff);
-  if (apiResult) exit(apiResult);
+  const apiResult = await tryApiReview(diff, opts);
+  if (apiResult) {
+    if (round === 1) {
+      apiResult.instructions = `Round 1 complete. Provide YOUR independent review, then call Round 2 with --round 2 --claude-review "<findings>"`;
+    } else {
+      apiResult.instructions = `Synthesize both rounds into a final review verdict.`;
+    }
+    exit(apiResult);
+  }
 
   // 4. No GPT available
   exit({

package/hooks/dual-brain-think.mjs

@@ -60,8 +60,33 @@ function findCodex() {
 // Prompt builder
 // ---------------------------------------------------------------------------
 
-function buildGptPrompt({ question, context, files }) {
+function buildGptPrompt({ question, context, files, round, claudePerspective }) {
+  if (round === 2 && claudePerspective) {
+    return `You are GPT-5.5 in a collaborative architectural discussion with Claude (Opus).
+You gave your initial analysis on a question. Claude has now provided its independent perspective.
+This is a professional dialogue — two experts refining a decision together.
+
+Original question: ${question}
+${context ? `\nContext: ${context}` : ''}
+
+Claude's perspective:
+${claudePerspective}
+
+Now respond as a colleague, not a critic. Structure your response:
+1. AGREEMENTS: Where Claude's analysis strengthens or confirms your thinking
+2. PUSHBACK: Where you disagree — be specific about WHY with evidence or reasoning
+3. NEW INSIGHTS: Anything Claude's perspective surfaced that you missed
+4. REFINED RECOMMENDATION: Your updated recommendation incorporating both perspectives
+5. REMAINING CONCERNS: Open questions neither of you fully resolved
+6. CONFIDENCE DELTA: Has your confidence changed? Why?
+
+Be direct and substantive. If Claude is right about something you got wrong, say so.
+If you still disagree after considering their points, explain what specific evidence would change your mind.`;
+  }
+
   return `You are GPT-5.5, providing an independent architectural perspective.
+This is Round 1 of a dual-brain analysis — Claude (Opus) will independently analyze the same question,
+then send you their perspective for a collaborative discussion in Round 2.
 
 Question: ${question}
 ${context ? `\nContext: ${context}` : ''}
@@ -165,7 +190,7 @@ function logUsage({ durationMs, usage, success }) {
 // Core exported function
 // ---------------------------------------------------------------------------
 
-export async function dualThink({ question, context, files } = {}) {
+export async function dualThink({ question, context, files, round, claudePerspective } = {}) {
   if (!question) {
     return {
       gpt: null,
@@ -174,6 +199,8 @@ export async function dualThink({ question, context, files } = {}) {
     };
   }
 
+  const effectiveRound = (round === 2 && claudePerspective) ? 2 : 1;
+
   const codexBin = findCodex();
   if (!codexBin) {
     return {
@@ -183,7 +210,6 @@ export async function dualThink({ question, context, files } = {}) {
     };
   }
 
-  // Check Codex auth before running
   try {
     execSync(`${codexBin} login status`, {
       encoding: 'utf8',
@@ -198,7 +224,7 @@ export async function dualThink({ question, context, files } = {}) {
     };
   }
 
-  const prompt = buildGptPrompt({ question, context, files });
+  const prompt = buildGptPrompt({ question, context, files, round: effectiveRound, claudePerspective });
   const raw = runGptAnalysis(codexBin, prompt);
 
   logUsage({ durationMs: raw.durationMs, usage: raw.usage, success: raw.success });
@@ -207,18 +233,44 @@ export async function dualThink({ question, context, files } = {}) {
     return {
       gpt: null,
       error: raw.error || 'GPT analysis failed',
-      fallback: 'Proceed with single-brain analysis on Claude Opus',
+      fallback: effectiveRound === 2
+        ? 'GPT rebuttal unavailable — synthesize from Round 1 analysis alone'
+        : 'Proceed with single-brain analysis on Claude Opus',
+    };
+  }
+
+  if (effectiveRound === 2) {
+    return {
+      round: 2,
+      gpt: {
+        rebuttal: raw.text,
+        model: MODEL,
+        durationMs: raw.durationMs,
+        tokens: raw.usage,
+      },
+      instructions: `GPT has responded to your analysis. Now synthesize both rounds into a FINAL DECISION:
+1. Where you both agree → high confidence, proceed
+2. Where GPT pushed back on your points → re-evaluate honestly
+3. Where you still disagree → state why and what evidence would resolve it
+4. Final recommendation with combined confidence level`,
+      question,
     };
   }
 
   return {
+    round: 1,
     gpt: {
       recommendation: raw.text,
       model: MODEL,
       durationMs: raw.durationMs,
       tokens: raw.usage,
     },
-    instructions: 'Now provide YOUR independent analysis of the same question. Then compare both perspectives and make a final decision. If you disagree with GPT, explain why with evidence.',
+    instructions: `Round 1 complete. Now:
+1. Provide YOUR independent analysis of the same question (same structure: recommendation, rationale, alternatives, risks, confidence, verification)
+2. Then call Round 2 to send your perspective back to GPT:
+   node .claude/hooks/dual-brain-think.mjs --question "<same question>" --round 2 --claude-says "<your analysis summary>"
+3. GPT will respond to your specific points — agreements, pushback, and refined recommendation
+4. You then synthesize both rounds into the final decision`,
     question,
     context: context || null,
   };
@@ -268,32 +320,41 @@ function printResult(result, question) {
   const TOP = '╔══════════════════════════════════════════════════╗';
   const BOT = '╚══════════════════════════════════════════════════╝';
 
+  const roundLabel = result.round === 2 ? 'Round 2 — Rebuttal' : 'Round 1 — Initial';
+
   console.log(TOP);
-  console.log('║              Dual-Brain Think                    ║');
+  console.log(`║  🧠 Dual-Brain Think · ${roundLabel}`.padEnd(51) + '║');
   console.log(BAR);
-  // Truncate question to fit the box
   const q = question.length > 44 ? question.slice(0, 41) + '...' : question;
   console.log(`║ Question: ${q.padEnd(38)} ║`);
   console.log(BAR);
 
   if (!result.gpt) {
-    // Failure path
-    console.log(`║ ERROR: ${(result.error || 'Unknown error').padEnd(41)} ║`);
+    console.log(`║ ❌ ${(result.error || 'Unknown error').padEnd(45)} ║`);
     console.log(BAR);
-    console.log(`║ Fallback: ${(result.fallback || '').padEnd(39)} ║`);
+    console.log(`║ ↩️  ${(result.fallback || '').padEnd(45)} ║`);
     console.log(BOT);
     return;
   }
 
-  const durSec = (result.gpt.durationMs / 1000).toFixed(1);
-  console.log(`║ GPT-5.5 Perspective (${MODEL}, ${durSec}s):`.padEnd(51) + '║');
+  const gptData = result.gpt;
+  const durSec = (gptData.durationMs / 1000).toFixed(1);
+  console.log(`║ 🤖 GPT-5.5 (${durSec}s):`.padEnd(51) + '║');
   console.log(BAR);
   console.log('');
-  console.log(result.gpt.recommendation);
+  console.log(gptData.recommendation || gptData.rebuttal);
   console.log('');
   console.log(BAR);
-  console.log('║ Now: Provide YOUR analysis and compare.          ║');
-  console.log('║ If you disagree, explain why with evidence.      ║');
+
+  if (result.round === 2) {
+    console.log('║ 🔄 Synthesize both rounds into final decision.  ║');
+    console.log('║ Where you agree → high confidence.               ║');
+    console.log('║ Where you disagree → state what would resolve it.║');
+  } else {
+    console.log('║ 📝 Your turn: analyze independently, then call   ║');
+    console.log('║    Round 2 with --round 2 --claude-says "..."    ║');
+    console.log('║    for GPT\'s rebuttal to your analysis.          ║');
+  }
   console.log(BOT);
 }
 
@@ -306,7 +367,8 @@ if (import.meta.url === `file://${process.argv[1]}`) {
 
   if (!args.question) {
     console.error(
-      'Usage: node dual-brain-think.mjs --question "<question>" [--context "<context>"] [--files file1,file2]'
+      'Usage: node dual-brain-think.mjs --question "<question>" [--context "<ctx>"] [--files f1,f2]\n' +
+      '       node dual-brain-think.mjs --question "<question>" --round 2 --claude-says "<analysis>"'
     );
     process.exit(1);
   }
@@ -315,6 +377,8 @@ if (import.meta.url === `file://${process.argv[1]}`) {
     question: args.question,
     context: args.context,
     files: args.files,
+    round: args.round ? parseInt(args.round, 10) : 1,
+    claudePerspective: args['claude-says'] || null,
   });
 
   printResult(result, args.question);