diff --git a/package.json b/package.json
index 49743f4..5de2492 100644
--- a/package.json
+++ b/package.json
@@ -2,7 +2,7 @@
   "name": "g1nation",
   "displayName": "G1nation",
   "description": "High-performance autonomous local AI coding agent for VS Code. Features vectorized inference, asynchronous task management, and 100% offline processing.",
-  "version": "2.38.0",
+  "version": "2.39.0",
   "publisher": "connectailab",
   "license": "MIT",
   "icon": "assets/icon.png",
diff --git a/src/agent.ts b/src/agent.ts
index ad73f59..6bf3b32 100644
--- a/src/agent.ts
+++ b/src/agent.ts
@@ -264,6 +264,9 @@ export class AgentExecutor {
                 activeBrain.description ? `Description: ${activeBrain.description}` : '',
                 brainPreview ? `Available file examples:\n${brainPreview}` : 'Files: none found'
             ].filter(Boolean).join('\n');
+            const brainInventoryCtx = prompt && this.isSecondBrainInventoryRequest(prompt)
+                ? `\n\n${this.buildSecondBrainInventoryContext(activeBrain, brainFiles)}`
+                : '';
             const editor = vscode.window.activeTextEditor;
             if (editor && editor.document.uri.scheme === 'file') {
                 const text = editor.document.getText();
@@ -332,7 +335,7 @@ export class AgentExecutor {
                 : '';
             const memoryCtx = this.buildMemoryContext(prompt || '', activeBrain);
 
-            const fullSystemPrompt = `${agentSkillCtx}\n\n${systemPrompt}${internetCtx}${memoryCtx}${designerCtx}${secondBrainTraceCtx}\n\n[CONTEXT]\n${brainContext}\n${contextBlock}${negativeCtx}`;
+            const fullSystemPrompt = `${agentSkillCtx}\n\n${systemPrompt}${internetCtx}${memoryCtx}${designerCtx}${secondBrainTraceCtx}\n\n[CONTEXT]\n${brainContext}${brainInventoryCtx}\n${contextBlock}${negativeCtx}`;
             const messagesForRequest: ChatMessage[] = [
                 { role: 'system', content: fullSystemPrompt, internal: true },
                 ...reqMessages
@@ -428,13 +431,10 @@ export class AgentExecutor {
                 ? renderSecondBrainTraceMarkdown(secondBrainTrace, !!options.secondBrainTraceDebug)
                 : '';
             const finalAssistantContent = traceMarkdown ? `${assistantContent}\n${traceMarkdown}` : assistantContent;
-            const assistantMessage: ChatMessage = { role: 'assistant', content: finalAssistantContent, internal: false, rationale };
-            this.chatHistory.push(assistantMessage);
-            
+
             this.statusBarManager.updateStatus(AgentStatus.Executing);
             const report = await this.executeActions(aiResponseText, rootPath, activeBrain);
             if (!assistantContent.trim() && report.length === 0) {
-                this.chatHistory.pop();
                 logError('Model returned an empty response without actions.', { model: actualModel, engine, apiUrl, loopDepth });
                 this.webview.postMessage({
                     type: 'error',
@@ -476,6 +476,8 @@ export class AgentExecutor {
                 return;
             }
 
+            const assistantMessage: ChatMessage = { role: 'assistant', content: finalAssistantContent, internal: false, rationale };
+            this.chatHistory.push(assistantMessage);
             this.emitHistoryChanged();
             this.statusBarManager.updateStatus(AgentStatus.Success);
             this.webview.postMessage({ type: 'streamChunk', value: finalAssistantContent });
@@ -613,6 +615,47 @@ export class AgentExecutor {
         return /(second brain|2nd brain|제2뇌|브레인|brain|기억|기록|노트|문서|참고해서|사용해서|검색해서|근거|출처)/i.test(prompt);
     }
 
+    private isSecondBrainInventoryRequest(prompt: string): boolean {
+        const normalized = prompt.toLowerCase();
+        const asksBrain = /(second brain|2nd brain|제2뇌|브레인|brain)/i.test(normalized);
+        const asksOverview = /(평가|분석|강점|약점|부족|무엇을 할 수|활용|전체|연결된|현재|inside|overview|inventory|strength|weakness)/i.test(normalized);
+        return asksBrain && asksOverview;
+    }
+
+    private buildSecondBrainInventoryContext(activeBrain: BrainProfile, brainFiles: string[]): string {
+        const relativeFiles = brainFiles.map((file) => path.relative(activeBrain.localBrainPath, file));
+        const directoryCounts = new Map<string, number>();
+        for (const rel of relativeFiles) {
+            const topDir = rel.includes(path.sep) ? rel.split(path.sep)[0] : '(root)';
+            directoryCounts.set(topDir, (directoryCounts.get(topDir) || 0) + 1);
+        }
+
+        const topDirectories = [...directoryCounts.entries()]
+            .sort((a, b) => b[1] - a[1] || a[0].localeCompare(b[0]))
+            .slice(0, 12)
+            .map(([dir, count]) => `- ${dir}: ${count} markdown files`)
+            .join('\n');
+
+        const samples = relativeFiles
+            .slice(0, 40)
+            .map((file) => `- ${file}`)
+            .join('\n');
+
+        return [
+            '[SECOND BRAIN INVENTORY]',
+            'The user is asking about the currently selected Second Brain as a knowledge base. Use this inventory as direct evidence.',
+            `Selected brain name: ${activeBrain.name}`,
+            `Selected brain path: ${activeBrain.localBrainPath}`,
+            `Markdown file count: ${brainFiles.length}`,
+            brainFiles.length > 0
+                ? 'Do not say the Second Brain has no data, no files, or cannot be evaluated because files were not provided.'
+                : 'No Markdown files were found in the selected Second Brain path.',
+            topDirectories ? `Top-level distribution:\n${topDirectories}` : 'Top-level distribution: none',
+            samples ? `Sample files:\n${samples}` : 'Sample files: none',
+            'For strengths and weaknesses, infer from the inventory and selected note excerpts. Mark broad conclusions as inference when they are not directly proven.'
+        ].join('\n');
+    }
+
     private isStaleRun(runId: number): boolean {
         return runId !== this.activeRunId;
     }