@xagent-ai/cli 1.2.0 → 1.2.2

package/src/ai-client.ts

@@ -1,1459 +1,1560 @@
-import axios, { AxiosInstance } from 'axios';
-import https from 'https';
-import { AuthConfig } from './types.js';
-
-// Message content block type for Anthropic format
-export interface AnthropicContentBlock {
-  type: 'text' | 'tool_use' | 'tool_result' | 'thinking';
-  text?: string;
-  id?: string;
-  name?: string;
-  input?: any;
-  tool_use_id?: string;
-  content?: string;
-  thinking?: string;
-}
-
-// Markdown rendering helper function
-function renderMarkdown(text: string): string {
-  // Code block rendering
-  text = text.replace(/```(\w*)\n([\s\S]*?)```/g, (_, lang, code) => {
-    return `\n┌─[${lang || 'code'}]\n${code.trim().split('\n').map((l: string) => '│ ' + l).join('\n')}\n└─\n`;
-  });
-
-  // Inline code rendering
-  text = text.replace(/`([^`]+)`/g, '`$1`');
-
-  // Bold rendering
-  text = text.replace(/\*\*([^*]+)\*\*/g, '●$1○');
-
-  // Italic rendering
-  text = text.replace(/\*([^*]+)\*/g, '/$1/');
-
-  // List rendering
-  text = text.replace(/^- (.*$)/gm, '○ $1');
-  text = text.replace(/^\d+\. (.*$)/gm, '• $1');
-
-  // Heading rendering
-  text = text.replace(/^### (.*$)/gm, '\n━━━ $1 ━━━\n');
-  text = text.replace(/^## (.*$)/gm, '\n━━━━━ $1 ━━━━━\n');
-  text = text.replace(/^# (.*$)/gm, '\n━━━━━━━ $1 ━━━━━━━\n');
-
-  // Quote rendering
-  text = text.replace(/^> (.*$)/gm, '│ │ $1');
-
-  // Link rendering
-  text = text.replace(/\[([^\]]+)\]\(([^)]+)\)/g, '[$1]($2)');
-
-  return text;
-}
-
-// Format message content
-function formatMessageContent(content: string | Array<any>): string {
-  if (typeof content === 'string') {
-    return renderMarkdown(content);
-  }
-
-  const parts: string[] = [];
-  let hasToolUse = false;
-
-  for (const block of content) {
-    if (block.type === 'text') {
-      parts.push(renderMarkdown(block.text || ''));
-    } else if (block.type === 'tool_use') {
-      hasToolUse = true;
-      parts.push(`[🔧 TOOL CALL PENDING: ${block.name}]`);
-    } else if (block.type === 'tool_result') {
-      const result = typeof block.content === 'string' ? block.content : JSON.stringify(block.content);
-      parts.push(`[✅ TOOL RESULT]\n${result}`);
-    } else if (block.type === 'thinking') {
-      parts.push(`[🧠 THINKING]\n${block.thinking || ''}`);
-    }
-  }
-
-  if (hasToolUse) {
-    parts.push('\n[⚠️  Note: Tool calls are executed by the framework, not displayed here]');
-  }
-
-  return parts.join('\n');
-}
-
-// Display messages by category
-function displayMessages(messages: any[], systemPrompt?: string): void {
-  const roleColors: Record<string, string> = {
-    system: '🟫 SYSTEM',
-    user: '👤 USER',
-    assistant: '🤖 ASSISTANT',
-    tool: '🔧 TOOL'
-  };
-
-  // Display system message first (if there's a separate systemPrompt parameter)
-  if (systemPrompt) {
-    console.log('\n┌─────────────────────────────────────────────────────────────┐');
-    console.log('│ 🟫 SYSTEM                                                     │');
-    console.log('├─────────────────────────────────────────────────────────────┤');
-    console.log(renderMarkdown(systemPrompt).split('\n').map((l: string) => '│ ' + l).join('\n'));
-    console.log('└─────────────────────────────────────────────────────────────┘');
-  }
-
-  // Iterate through all messages
-  for (let i = 0; i < messages.length; i++) {
-    const msg = messages[i];
-    const role = msg.role as string;
-    const roleLabel = roleColors[role] || `● ${role.toUpperCase()}`;
-
-    console.log(`\n┌─────────────────────────────────────────────────────────────┐`);
-    console.log(`│ ${roleLabel} (${i + 1}/${messages.length})                                          │`);
-    console.log('├─────────────────────────────────────────────────────────────┤');
-
-    // Display reasoning_content (if present) - check both camelCase and snake_case
-    const reasoningContent = (msg as any).reasoningContent || (msg as any).reasoning_content;
-    if (reasoningContent) {
-      console.log('│ 🧠 REASONING:');
-      console.log('│ ───────────────────────────────────────────────────────────');
-      const reasoningLines = renderMarkdown(reasoningContent).split('\n');
-      for (const line of reasoningLines.slice(0, 20)) {
-        console.log('│ ' + line.slice(0, 62));
-      }
-      if (reasoningContent.length > 1000) console.log('│ ... (truncated)');
-      console.log('│ ───────────────────────────────────────────────────────────');
-    }
-
-    // Display main content
-    const content = formatMessageContent(msg.content);
-    const lines = content.split('\n');
-
-    for (const line of lines.slice(0, 50)) {
-      console.log('│ ' + line.slice(0, 62));
-    }
-    if (lines.length > 50) {
-      console.log('│ ... (' + (lines.length - 50) + ' more lines)');
-    }
-
-    console.log('└─────────────────────────────────────────────────────────────┘');
-  }
-}
-
-// Format response content
-function formatResponseContent(content: string | Array<any>): string {
-  if (typeof content === 'string') {
-    return renderMarkdown(content);
-  }
-
-  const parts: string[] = [];
-  let hasToolUse = false;
-
-  for (const block of content) {
-    if (block.type === 'text') {
-      parts.push(renderMarkdown(block.text || ''));
-    } else if (block.type === 'tool_use') {
-      hasToolUse = true;
-      // Tool calls are handled via tool_calls field, not shown here
-    } else if (block.type === 'tool_result') {
-      const result = typeof block.content === 'string' ? block.content : JSON.stringify(block.content);
-      parts.push(`[✅ TOOL RESULT]\n${result}`);
-    } else if (block.type === 'thinking') {
-      parts.push(`[🧠 THINKING]\n${block.thinking || ''}`);
-    } else if (block.type === 'image') {
-      parts.push('[IMAGE]');
-    }
-  }
-
-  if (hasToolUse) {
-    parts.push('\n[⚠️  Note: Tool calls are executed via tool_calls field, not shown here]');
-  }
-
-  return parts.join('\n');
-}
-
-export interface Message {
-  role: 'system' | 'user' | 'assistant' | 'tool';
-  content: string | Array<AnthropicContentBlock | { type: string; text?: string; image_url?: { url: string } }>;
-  reasoning_content?: string;
-  tool_calls?: any[];
-  tool_call_id?: string;
-}
-
-export interface ToolDefinition {
-  type: 'function';
-  function: {
-    name: string;
-    description: string;
-    parameters?: any;
-  };
-}
-
-export interface ChatCompletionOptions {
-  model?: string;
-  temperature?: number;
-  maxTokens?: number;
-  tools?: ToolDefinition[];
-  toolChoice?: 'auto' | 'none' | { type: string; function: { name: string } };
-  stream?: boolean;
-  thinkingTokens?: number;
-}
-
-export interface ChatCompletionResponse {
-  id: string;
-  object: string;
-  created: number;
-  model: string;
-  choices: Array<{
-    index: number;
-    message: Message;
-    finish_reason: string;
-  }>;
-  usage?: {
-    prompt_tokens: number;
-    completion_tokens: number;
-    total_tokens: number;
-  };
-}
-
-// Detect if it's Anthropic compatible API（Use x-api-key authentication header）
-function isAnthropicCompatible(baseUrl: string): boolean {
-  return baseUrl.includes('anthropic') || 
-         baseUrl.includes('minimaxi.com') ||
-         baseUrl.includes('minimax.chat');
-}
-
-// MiniMax API path detection
-function detectMiniMaxAPI(baseUrl: string): boolean {
-  return baseUrl.includes('minimax.chat') || 
-         baseUrl.includes('minimaxi.com');
-}
-
-// Get correct endpoint path for MiniMax
-function getMiniMaxEndpoint(baseUrl: string): { endpoint: string; format: 'anthropic' | 'openai' } {
-  // MiniMax Anthropic format: https://api.minimax.chat/anthropic + /v1/messages
-  if (baseUrl.includes('/anthropic')) {
-    return { endpoint: '/v1/messages', format: 'anthropic' };
-  }
-  // MiniMax OpenAI format: https://api.minimaxi.com/v1 + /chat/completions
-  if (baseUrl.includes('/v1') && !baseUrl.includes('/anthropic')) {
-    return { endpoint: '/chat/completions', format: 'openai' };
-  }
-  // Default to Anthropic format
-  return { endpoint: '/v1/messages', format: 'anthropic' };
-}
-
-export class AIClient {
-  private client: AxiosInstance;
-  private authConfig: AuthConfig;
-
-  constructor(authConfig: AuthConfig) {
-    this.authConfig = authConfig;
-    const isMiniMax = detectMiniMaxAPI(authConfig.baseUrl || '');
-    const isAnthropicOfficial = !isMiniMax && isAnthropicCompatible(authConfig.baseUrl || '');
-    
-    const headers: Record<string, string> = {
-      'Content-Type': 'application/json'
-    };
-    
-    if (isMiniMax) {
-      // MiniMax: Use x-api-key authentication header
-      headers['x-api-key'] = authConfig.apiKey || '';
-      headers['anthropic-version'] = '2023-06-01';
-    } else if (isAnthropicOfficial) {
-      // Anthropic official: Use x-api-key authentication header
-      headers['x-api-key'] = authConfig.apiKey || '';
-      headers['anthropic-version'] = '2023-06-01';
-      headers['anthropic-dangerous-direct-browser-access'] = 'true';
-    } else {
-      // Other OpenAI compatible: 使用 Bearer token
-      headers['Authorization'] = `Bearer ${authConfig.apiKey}`;
-    }
-    
-    this.client = axios.create({
-      baseURL: authConfig.baseUrl,
-      headers,
-      timeout: 240000,
-      httpsAgent: new https.Agent({ rejectUnauthorized: false })
-    });
-  }
-
-  // Convert OpenAI format messages to Anthropic format
-  private convertToAnthropicFormat(
-    messages: Message[],
-    systemPrompt?: string
-  ): { system: string; messages: Array<{ role: string; content: AnthropicContentBlock[] }> } {
-    const systemMessages = messages.filter(m => m.role === 'system');
-    const otherMessages = messages.filter(m => m.role !== 'system');
-    
-    const systemContent = systemMessages[0]?.content;
-    const system = systemPrompt || (typeof systemContent === 'string' ? systemContent : '');
-
-    const anthropicMessages: Array<{ role: string; content: AnthropicContentBlock[] }> = [];
-
-    for (const msg of otherMessages) {
-      const blocks: AnthropicContentBlock[] = [];
-
-      if (typeof msg.content === 'string') {
-        blocks.push({ type: 'text', text: msg.content });
-      } else if (Array.isArray(msg.content)) {
-        for (const block of msg.content) {
-          if (block.type === 'text' && 'text' in block) {
-            blocks.push({ type: 'text', text: (block as any).text });
-          } else if (block.type === 'tool_use') {
-            blocks.push({
-              type: 'tool_use',
-              id: (block as any).id,
-              name: (block as any).function?.name || (block as any).name,
-              input: (block as any).function?.arguments || (block as any).input
-            });
-          } else if (block.type === 'tool_result') {
-            blocks.push({
-              type: 'tool_result',
-              tool_use_id: (block as any).tool_call_id || (block as any).tool_use_id,
-              content: typeof (block as any).content === 'string' 
-                ? (block as any).content 
-                : JSON.stringify((block as any).content)
-            });
-          } else if (block.type === 'thinking') {
-            blocks.push({ type: 'thinking', thinking: (block as any).thinking });
-          }
-        }
-      }
-
-      // Handle tool_calls (OpenAI 格式)
-      if (msg.tool_calls) {
-        for (const tc of msg.tool_calls) {
-          blocks.push({
-            type: 'tool_use',
-            id: tc.id,
-            name: tc.function?.name,
-            input: tc.function?.arguments ? (typeof tc.function.arguments === 'string' ? JSON.parse(tc.function.arguments) : tc.function.arguments) : {}
-          });
-        }
-      }
-
-      if (blocks.length > 0) {
-        anthropicMessages.push({
-          role: msg.role === 'tool' ? 'user' : msg.role,
-          content: blocks as AnthropicContentBlock[]
-        });
-      }
-    }
-
-    return { system, messages: anthropicMessages };
-  }
-
-  async chatCompletion(
-    messages: Message[],
-    options: ChatCompletionOptions = {}
-  ): Promise<ChatCompletionResponse> {
-    const model = options.model || this.authConfig.modelName || 'gpt-4';
-    const isMiniMax = detectMiniMaxAPI(this.authConfig.baseUrl || '');
-
-    if (isMiniMax) {
-      return this.minimaxChatCompletion(messages, options);
-    }
-
-    const isAnthropic = isAnthropicCompatible(this.authConfig.baseUrl || '');
-    if (isAnthropic) {
-      return this.anthropicNativeChatCompletion(messages, options);
-    }
-
-    // OpenAI format request
-    const requestBody: any = {
-      model,
-      messages,
-      temperature: options.temperature ?? 0.7,
-      stream: options.stream ?? false
-    };
-
-    if (options.maxTokens && options.maxTokens > 0) {
-      requestBody.max_tokens = options.maxTokens;
-    }
-
-    if (options.tools && options.tools.length > 0) {
-      requestBody.tools = options.tools;
-      requestBody.tool_choice = options.toolChoice || 'auto';
-    }
-
-    if (options.thinkingTokens && options.thinkingTokens > 0) {
-      requestBody.max_completion_tokens = options.thinkingTokens;
-    }
-
-    // Debug output（受showAIDebugInfo配置控制）
-    const showDebug = this.authConfig.showAIDebugInfo ?? false;
-    
-    if (showDebug) {
-      console.log('\n╔══════════════════════════════════════════════════════════╗');
-      console.log('║                    AI REQUEST DEBUG                      ║');
-      console.log('╚══════════════════════════════════════════════════════════╝');
-      console.log(`📦 Model: ${model}`);
-      console.log(`🌐 Base URL: ${this.authConfig.baseUrl}`);
-      console.log(`💬 Total Messages: ${messages.length} items`);
-      if (options.temperature !== undefined) console.log(`🌡️  Temperature: ${options.temperature}`);
-      if (options.maxTokens) console.log(`📏 Max Tokens: ${options.maxTokens}`);
-      if (options.tools?.length) console.log(`🔧 Tools: ${options.tools.length} items`);
-      if (options.thinkingTokens) console.log(`🧠 Thinking Tokens: ${options.thinkingTokens}`);
-      console.log('─'.repeat(60));
-      
-      // Separate system messages
-      const systemMsgs = messages.filter(m => m.role === 'system');
-      const otherMsgs = messages.filter(m => m.role !== 'system');
-      
-      if (systemMsgs.length > 0) {
-        const systemContent = typeof systemMsgs[0].content === 'string' 
-          ? systemMsgs[0].content 
-          : formatMessageContent(systemMsgs[0].content);
-        console.log('\n┌─────────────────────────────────────────────────────────────┐');
-        console.log('│ 🟫 SYSTEM                                                     │');
-        console.log('├─────────────────────────────────────────────────────────────┤');
-        console.log(renderMarkdown(systemContent).split('\n').map(l => '│ ' + l).join('\n'));
-        console.log('└─────────────────────────────────────────────────────────────┘');
-      }
-      
-      displayMessages(otherMsgs);
-      
-      console.log('\n📤 Sending request to API...\n');
-    }
-
-    try {
-      const response = await this.client.post('/chat/completions', requestBody);
-      
-      if (showDebug) {
-        console.log('\n╔══════════════════════════════════════════════════════════╗');
-        console.log('║                   AI RESPONSE DEBUG                      ║');
-        console.log('╚══════════════════════════════════════════════════════════╝');
-        console.log(`🆔 ID: ${response.data.id}`);
-        console.log(`🤖 Model: ${response.data.model}`);
-        const usage = response.data.usage;
-        if (usage) {
-          console.log(`📊 Tokens: ${usage.prompt_tokens} (prompt) + ${usage.completion_tokens} (completion) = ${usage.total_tokens} (total)`);
-        }
-        const choice = response.data.choices?.[0];
-        if (choice) {
-          console.log(`🏁 Finish Reason: ${choice.finish_reason}`);
-          
-          console.log('\n┌─────────────────────────────────────────────────────────────┐');
-          console.log('│ 🤖 ASSISTANT                                                 │');
-          console.log('├─────────────────────────────────────────────────────────────┤');
-          
-          // Display reasoning_content（如果有）
-          if (choice.message.reasoning_content) {
-            console.log('│ 🧠 REASONING:');
-            console.log('│ ───────────────────────────────────────────────────────────');
-            const reasoningLines = renderMarkdown(choice.message.reasoning_content).split('\n');
-            for (const line of reasoningLines.slice(0, 15)) {
-              console.log('│ ' + line.slice(0, 62));
-            }
-            if (choice.message.reasoning_content.length > 800) console.log('│ ... (truncated)');
-            console.log('│ ───────────────────────────────────────────────────────────');
-          }
-          
-          // Display main content
-          const content = formatResponseContent(choice.message.content);
-          const lines = content.split('\n');
-          console.log('│ 💬 CONTENT:');
-          console.log('│ ───────────────────────────────────────────────────────────');
-          for (const line of lines.slice(0, 40)) {
-            console.log('│ ' + line.slice(0, 62));
-          }
-          if (lines.length > 40) {
-            console.log(`│ ... (${lines.length - 40} more lines)`);
-          }
-          console.log('└─────────────────────────────────────────────────────────────┘');
-        }
-        console.log('╔══════════════════════════════════════════════════════════╗');
-        console.log('║                    RESPONSE ENDED                        ║');
-        console.log('╚══════════════════════════════════════════════════════════╝\n');
-      }
-      
-      return response.data;
-    } catch (error: any) {
-      if (error.response) {
-        throw new Error(
-          `API Error: ${error.response.status} - ${JSON.stringify(error.response.data)}`
-        );
-      } else if (error.request) {
-        throw new Error('Network error: No response received from server');
-      } else {
-        throw new Error(`Request error: ${error.message}`);
-      }
-    }
-  }
-
-  // Anthropic official原生 API（使用 /v1/messages 端点）
-  private async anthropicNativeChatCompletion(
-    messages: Message[],
-    options: ChatCompletionOptions = {}
-  ): Promise<ChatCompletionResponse> {
-    const { system, messages: anthropicMessages } = this.convertToAnthropicFormat(messages);
-
-    const requestBody: any = {
-      model: options.model || this.authConfig.modelName || 'claude-sonnet-4-20250514',
-      messages: anthropicMessages,
-      temperature: options.temperature ?? 1.0,
-      stream: false,
-      max_tokens: options.maxTokens || 4096
-    };
-
-    if (system) {
-      requestBody.system = system;
-    }
-
-    // Anthropic native tool format
-    if (options.tools && options.tools.length > 0) {
-      requestBody.tools = options.tools.map(tool => ({
-        name: tool.function.name,
-        description: tool.function.description,
-        input_schema: tool.function.parameters || { type: 'object', properties: {} }
-      }));
-      
-      // Convert tool_choice 从 OpenAI 格式到 Anthropic 格式
-      const toolChoice = options.toolChoice;
-      if (toolChoice === 'none') {
-        requestBody.tool_choice = { type: 'auto' };
-      } else if (toolChoice && typeof toolChoice === 'object') {
-        if (toolChoice.type === 'function' && toolChoice.function) {
-          requestBody.tool_choice = { type: 'tool', tool: { name: toolChoice.function.name } };
-        } else {
-          requestBody.tool_choice = { type: 'auto' };
-        }
-      } else {
-        requestBody.tool_choice = { type: 'auto' };
-      }
-    }
-
-    // Anthropic thinking mode
-    if (options.thinkingTokens && options.thinkingTokens > 0) {
-      requestBody.thinking = { type: 'enabled', budget_tokens: options.thinkingTokens };
-    }
-
-    // Debug output（受showAIDebugInfo配置控制）
-    const showDebug = this.authConfig.showAIDebugInfo ?? false;
-    
-    if (showDebug) {
-      console.log('\n╔══════════════════════════════════════════════════════════╗');
-      console.log('║              AI REQUEST DEBUG (ANTHROPIC)                ║');
-      console.log('╚══════════════════════════════════════════════════════════╝');
-      console.log(`📦 Model: ${requestBody.model}`);
-      console.log(`🌐 Base URL: ${this.authConfig.baseUrl}`);
-      console.log(`💬 Total Messages: ${anthropicMessages.length} items`);
-      if (requestBody.temperature) console.log(`🌡️  Temperature: ${requestBody.temperature}`);
-      if (requestBody.max_tokens) console.log(`📏 Max Tokens: ${requestBody.max_tokens}`);
-      if (requestBody.tools) console.log(`🔧 Tools: ${requestBody.tools.length} items`);
-      if (requestBody.thinking) console.log(`🧠 Thinking Budget: ${requestBody.thinking.budget_tokens}`);
-      console.log('─'.repeat(60));
-      
-      // Display system messages
-      if (system) {
-        console.log('\n┌─────────────────────────────────────────────────────────────┐');
-        console.log('│ 🟫 SYSTEM                                                     │');
-        console.log('├─────────────────────────────────────────────────────────────┤');
-        console.log(renderMarkdown(system).split('\n').map(l => '│ ' + l).join('\n'));
-        console.log('└─────────────────────────────────────────────────────────────┘');
-      }
-      
-      // Display user and assistant messages
-      displayMessages(anthropicMessages);
-      
-      console.log('\n📤 Sending to Anthropic API (v1/messages)...\n');
-    }
-
-    try {
-      // Use Anthropic native endpoint /v1/messages
-      const response = await this.client.post('/v1/messages', requestBody);
-      
-      if (showDebug) {
-        console.log('\n╔══════════════════════════════════════════════════════════╗');
-        console.log('║             AI RESPONSE DEBUG (ANTHROPIC)                ║');
-        console.log('╚══════════════════════════════════════════════════════════╝');
-        console.log(`🆔 ID: ${response.data.id}`);
-        console.log(`🤖 Model: ${response.data.model}`);
-        const usage = response.data.usage;
-        if (usage) {
-          console.log(`📊 Tokens: ${usage.input_tokens} (input) + ${usage.output_tokens} (output) = ${usage.input_tokens + usage.output_tokens} (total)`);
-        }
-        console.log(`🏁 Stop Reason: ${response.data.stop_reason}`);
-        
-        console.log('\n┌─────────────────────────────────────────────────────────────┐');
-        console.log('│ 🤖 ASSISTANT                                                 │');
-        console.log('├─────────────────────────────────────────────────────────────┤');
-        
-        const content = response.data.content || [];
-        const reasoning = content.filter((c: any) => c.type === 'thinking').map((c: any) => c.thinking).join('');
-        const textContent = content.filter((c: any) => c.type === 'text').map((c: any) => c.text).join('');
-        
-        // Display thinking
-        if (reasoning) {
-          console.log('│ 🧠 REASONING:');
-          console.log('│ ───────────────────────────────────────────────────────────');
-          const reasoningLines = renderMarkdown(reasoning).split('\n');
-          for (const line of reasoningLines.slice(0, 15)) {
-            console.log('│ ' + line.slice(0, 62));
-          }
-          if (reasoning.length > 800) console.log('│ ... (truncated)');
-          console.log('│ ───────────────────────────────────────────────────────────');
-        }
-        
-        // Display content
-        console.log('│ 💬 CONTENT:');
-        console.log('│ ───────────────────────────────────────────────────────────');
-        const lines = renderMarkdown(textContent).split('\n');
-        for (const line of lines.slice(0, 40)) {
-          console.log('│ ' + line.slice(0, 62));
-        }
-        if (lines.length > 40) {
-          console.log(`│ ... (${lines.length - 40} more lines)`);
-        }
-        console.log('└─────────────────────────────────────────────────────────────┘');
-        
-        console.log('\n╔══════════════════════════════════════════════════════════╗');
-        console.log('║                    RESPONSE ENDED                        ║');
-        console.log('╚══════════════════════════════════════════════════════════╝\n');
-      }
-      
-      return this.convertFromAnthropicNativeResponse(response.data);
-    } catch (error: any) {
-      if (error.response) {
-        throw new Error(
-          `API Error: ${error.response.status} - ${JSON.stringify(error.response.data)}`
-        );
-      } else if (error.request) {
-        throw new Error('Network error: No response received from server');
-      } else {
-        throw new Error(`Request error: ${error.message}`);
-      }
-    }
-  }
-
-  // MiniMax API（Automatically select based on baseUrl Anthropic 或 OpenAI 格式）
-  private async minimaxChatCompletion(
-    messages: Message[],
-    options: ChatCompletionOptions = {}
-  ): Promise<ChatCompletionResponse> {
-    const { system, messages: anthropicMessages } = this.convertToAnthropicFormat(messages);
-    const { endpoint, format } = getMiniMaxEndpoint(this.authConfig.baseUrl || '');
-
-    const requestBody: any = {
-      model: options.model || this.authConfig.modelName || 'MiniMax-M2',
-      messages: format === 'anthropic' ? anthropicMessages : messages,
-      temperature: options.temperature ?? 1.0,
-      stream: false,
-      max_tokens: options.maxTokens || 4096
-    };
-
-    if (system && format === 'anthropic') {
-      requestBody.system = system;
-    }
-
-    if (format === 'anthropic') {
-      // Anthropic format tools
-      if (options.tools && options.tools.length > 0) {
-        requestBody.tools = options.tools.map(tool => ({
-          name: tool.function.name,
-          description: tool.function.description,
-          input_schema: tool.function.parameters || { type: 'object', properties: {} }
-        }));
-        
-        const toolChoice = options.toolChoice;
-        if (toolChoice === 'none') {
-          requestBody.tool_choice = { type: 'auto' };
-        } else if (toolChoice && typeof toolChoice === 'object') {
-          if (toolChoice.type === 'function' && toolChoice.function) {
-            requestBody.tool_choice = { type: 'tool', tool: { name: toolChoice.function.name } };
-          } else {
-            requestBody.tool_choice = { type: 'auto' };
-          }
-        } else {
-          requestBody.tool_choice = { type: 'auto' };
-        }
-      }
-    } else {
-      // OpenAI format tools
-      if (options.tools && options.tools.length > 0) {
-        requestBody.tools = options.tools;
-        requestBody.tool_choice = options.toolChoice || 'auto';
-      }
-    }
-
-    // Debug output（受showAIDebugInfo配置控制）
-    const showDebug = this.authConfig.showAIDebugInfo ?? false;
-    
-    if (showDebug) {
-      console.log('\n╔══════════════════════════════════════════════════════════╗');
-      console.log('║               AI REQUEST DEBUG (MINIMAX)                 ║');
-      console.log('╚══════════════════════════════════════════════════════════╝');
-      console.log(`📦 Model: ${requestBody.model}`);
-      console.log(`🔗 Format: ${format.toUpperCase()} | Endpoint: ${endpoint}`);
-      console.log(`🌐 Base URL: ${this.authConfig.baseUrl}`);
-      console.log(`💬 Total Messages: ${requestBody.messages.length} items`);
-      if (requestBody.temperature) console.log(`🌡️  Temperature: ${requestBody.temperature}`);
-      if (requestBody.max_tokens) console.log(`📏 Max Tokens: ${requestBody.max_tokens}`);
-      if (requestBody.tools) console.log(`🔧 Tools: ${requestBody.tools.length} items`);
-      console.log('─'.repeat(60));
-      
-      // Display system messages
-      if (system && format === 'anthropic') {
-        console.log('\n┌─────────────────────────────────────────────────────────────┐');
-        console.log('│ 🟫 SYSTEM                                                     │');
-        console.log('├─────────────────────────────────────────────────────────────┤');
-        console.log(renderMarkdown(system).split('\n').map(l => '│ ' + l).join('\n'));
-        console.log('└─────────────────────────────────────────────────────────────┘');
-      }
-      
-      // Display other messages
-      displayMessages(requestBody.messages);
-      
-      console.log('\n📤 Sending to MiniMax API...\n');
-    }
-
-    try {
-      // MiniMax uses correct endpoint
-      const response = await this.client.post(endpoint, requestBody);
-      
-      if (showDebug) {
-        console.log('\n╔══════════════════════════════════════════════════════════╗');
-        console.log('║              AI RESPONSE DEBUG (MINIMAX)                 ║');
-        console.log('╚══════════════════════════════════════════════════════════╝');
-        console.log(`🆔 ID: ${response.data.id}`);
-        console.log(`🤖 Model: ${response.data.model}`);
-        const usage = response.data.usage;
-        if (usage) {
-          console.log(`📊 Tokens: ${usage.prompt_tokens} (prompt) + ${usage.completion_tokens} (completion) = ${usage.total_tokens} (total)`);
-        }
-        console.log(`🏁 Stop Reason: ${response.data.stop_reason}`);
-        
-        console.log('\n┌─────────────────────────────────────────────────────────────┐');
-        console.log('│ 🤖 ASSISTANT                                                 │');
-        console.log('├─────────────────────────────────────────────────────────────┤');
-        
-        const message = response.data.choices?.[0]?.message;
-        const content = typeof message?.content === 'string' ? message.content : JSON.stringify(message?.content);
-        
-        console.log('│ 💬 CONTENT:');
-        console.log('│ ───────────────────────────────────────────────────────────');
-        const lines = renderMarkdown(content).split('\n');
-        for (const line of lines.slice(0, 40)) {
-          console.log('│ ' + line.slice(0, 62));
-        }
-        if (lines.length > 40) {
-          console.log(`│ ... (${lines.length - 40} more lines)`);
-        }
-        console.log('└─────────────────────────────────────────────────────────────┘');
-        
-        console.log('\n╔══════════════════════════════════════════════════════════╗');
-        console.log('║                    RESPONSE ENDED                        ║');
-        console.log('╚══════════════════════════════════════════════════════════╝\n');
-      }
-      
-      if (format === 'anthropic') {
-        return this.convertFromAnthropicNativeResponse(response.data);
-      } else {
-        return this.convertFromMiniMaxResponse(response.data);
-      }
-    } catch (error: any) {
-      if (error.response) {
-        throw new Error(
-          `API Error: ${error.response.status} - ${JSON.stringify(error.response.data)}`
-        );
-      } else if (error.request) {
-        throw new Error('Network error: No response received from server');
-      } else {
-        throw new Error(`Request error: ${error.message}`);
-      }
-    }
-  }
-
-  // Convert Anthropic native response to unified format
-  private convertFromAnthropicNativeResponse(anthropicResponse: any): ChatCompletionResponse {
-    const content = anthropicResponse.content || [];
-    let textContent = '';
-    let reasoningContent = '';
-    const toolCalls: any[] = [];
-
-    for (const block of content) {
-      if (block.type === 'text') {
-        textContent += block.text || '';
-      } else if (block.type === 'thinking') {
-        reasoningContent += block.thinking || '';
-      } else if (block.type === 'tool_use') {
-        toolCalls.push({
-          id: block.id,
-          type: 'function',
-          function: {
-            name: block.name,
-            arguments: JSON.stringify(block.input || {})
-          }
-        });
-      }
-    }
-
-    return {
-      id: anthropicResponse.id || `anthropic-${Date.now()}`,
-      object: 'chat.completion',
-      created: Math.floor(Date.now() / 1000),
-      model: anthropicResponse.model || this.authConfig.modelName || 'claude-sonnet-4-20250514',
-      choices: [{
-        index: 0,
-        message: {
-          role: 'assistant',
-          content: textContent,
-          reasoning_content: reasoningContent || undefined,
-          tool_calls: toolCalls.length > 0 ? toolCalls : undefined
-        },
-        finish_reason: anthropicResponse.stop_reason === 'end_turn' ? 'stop' : 
-                       anthropicResponse.stop_reason === 'max_tokens' ? 'length' : 'stop'
-      }],
-      usage: anthropicResponse.usage ? {
-        prompt_tokens: anthropicResponse.usage.input_tokens || 0,
-        completion_tokens: anthropicResponse.usage.output_tokens || 0,
-        total_tokens: (anthropicResponse.usage.input_tokens || 0) + (anthropicResponse.usage.output_tokens || 0)
-      } : undefined
-    };
-  }
-
-  // Convert MiniMax response to unified format
-  private convertFromMiniMaxResponse(minimaxResponse: any): ChatCompletionResponse {
-    const message = minimaxResponse.choices?.[0]?.message;
-    const content = message?.content;
-    let textContent = '';
-    let reasoningContent = '';
-    const toolCalls: any[] = [];
-
-    if (typeof content === 'string') {
-      textContent = content.trim();
-    } else if (Array.isArray(content)) {
-      for (const block of content) {
-        if (block.type === 'text') {
-          textContent += block.text || '';
-        } else if (block.type === 'thinking') {
-          reasoningContent += block.thinking || '';
-        } else if (block.type === 'tool_use') {
-          toolCalls.push({
-            id: block.id,
-            type: 'function',
-            function: {
-              name: block.name,
-              arguments: JSON.stringify(block.input || {})
-            }
-          });
-        }
-      }
-    }
-
-    return {
-      id: minimaxResponse.id || `minimax-${Date.now()}`,
-      object: 'chat.completion',
-      created: Math.floor(Date.now() / 1000),
-      model: minimaxResponse.model || this.authConfig.modelName || 'MiniMax-M2',
-      choices: [{
-        index: 0,
-        message: {
-          role: 'assistant',
-          content: textContent,
-          reasoning_content: reasoningContent || undefined,
-          tool_calls: toolCalls.length > 0 ? toolCalls : undefined
-        },
-        finish_reason: minimaxResponse.stop_reason === 'end_turn' ? 'stop' : 
-                       minimaxResponse.stop_reason === 'max_tokens' ? 'length' : 'stop'
-      }],
-      usage: minimaxResponse.usage
-    };
-  }
-
-  async *streamChatCompletion(
-    messages: Message[],
-    options: ChatCompletionOptions = {}
-  ): AsyncGenerator<string, void, unknown> {
-    const isMiniMax = detectMiniMaxAPI(this.authConfig.baseUrl || '');
-
-    if (isMiniMax) {
-      yield* this.minimaxStreamChatCompletion(messages, options);
-      return;
-    }
-
-    const isAnthropic = isAnthropicCompatible(this.authConfig.baseUrl || '');
-    if (isAnthropic) {
-      yield* this.anthropicNativeStreamChatCompletion(messages, options);
-      return;
-    }
-
-    // OpenAI streaming response
-    const model = options.model || this.authConfig.modelName || 'gpt-4';
-
-    const requestBody: any = {
-      model,
-      messages,
-      temperature: options.temperature ?? 0.7,
-      stream: true
-    };
-
-    if (options.maxTokens && options.maxTokens > 0) {
-      requestBody.max_tokens = options.maxTokens;
-    }
-
-    if (options.tools && options.tools.length > 0) {
-      requestBody.tools = options.tools;
-      requestBody.tool_choice = options.toolChoice || 'auto';
-    }
-
-    if (options.thinkingTokens && options.thinkingTokens > 0) {
-      requestBody.max_completion_tokens = options.thinkingTokens;
-    }
-
-    // Debug output（受showAIDebugInfo配置控制）
-    const showDebug = this.authConfig.showAIDebugInfo ?? false;
-    
-    if (showDebug) {
-      console.log('\n╔══════════════════════════════════════════════════════════╗');
-      console.log('║              AI REQUEST DEBUG (STREAM)                   ║');
-      console.log('╚══════════════════════════════════════════════════════════╝');
-      console.log(`📦 Model: ${model}`);
-      console.log(`🌐 Base URL: ${this.authConfig.baseUrl}`);
-      console.log(`💬 Total Messages: ${messages.length} items`);
-      if (options.temperature) console.log(`🌡️  Temperature: ${options.temperature}`);
-      if (options.maxTokens) console.log(`📏 Max Tokens: ${options.maxTokens}`);
-      if (options.tools?.length) console.log(`🔧 Tools: ${options.tools.length} items`);
-      console.log('─'.repeat(60));
-      
-      // Separate and display messages
-      const systemMsgs = messages.filter(m => m.role === 'system');
-      const otherMsgs = messages.filter(m => m.role !== 'system');
-      
-      if (systemMsgs.length > 0) {
-        const systemContent = typeof systemMsgs[0].content === 'string' 
-          ? systemMsgs[0].content 
-          : formatMessageContent(systemMsgs[0].content);
-        console.log('\n┌─────────────────────────────────────────────────────────────┐');
-        console.log('│ 🟫 SYSTEM                                                     │');
-        console.log('├─────────────────────────────────────────────────────────────┤');
-        console.log(renderMarkdown(systemContent).split('\n').map(l => '│ ' + l).join('\n'));
-        console.log('└─────────────────────────────────────────────────────────────┘');
-      }
-      
-      displayMessages(otherMsgs);
-      
-      console.log('\n📤 Starting stream...\n');
-    }
-
-    try {
-      const response = await this.client.post('/chat/completions', requestBody, {
-        responseType: 'stream'
-      });
-
-      console.log('📥 Receiving stream chunks...\n');
-
-      let buffer = '';
-      let chunkCount = 0;
-      let outputBuffer = '';
-
-      for await (const chunk of response.data) {
-        buffer += chunk.toString();
-        const lines = buffer.split('\n');
-
-        buffer = lines.pop() || '';
-
-        for (const line of lines) {
-          const trimmedLine = line.trim();
-          if (!trimmedLine) continue;
-
-          if (trimmedLine.startsWith('data: ')) {
-            const data = trimmedLine.slice(6);
-            if (data === '[DONE]') {
-              if (showDebug) {
-                console.log('\n╔══════════════════════════════════════════════════════════╗');
-                console.log('║              STREAM COMPLETED                            ║');
-                console.log('╚══════════════════════════════════════════════════════════╝');
-                console.log(`📦 Total chunks: ${chunkCount}`);
-                console.log(`📏 Total output: ${outputBuffer.length} chars`);
-                
-                console.log('\n┌─────────────────────────────────────────────────────────────┐');
-                console.log('│ 🤖 ASSISTANT OUTPUT                                        │');
-                console.log('├─────────────────────────────────────────────────────────────┤');
-                console.log('│ 💬 CONTENT:');
-                console.log('│ ───────────────────────────────────────────────────────────');
-                const lines = renderMarkdown(outputBuffer).split('\n');
-                for (const line of lines.slice(0, 30)) {
-                  console.log('│ ' + line.slice(0, 62));
-                }
-                if (lines.length > 30) {
-                  console.log(`│ ... (${lines.length - 30} more lines)`);
-                }
-                console.log('└─────────────────────────────────────────────────────────────┘');
-                console.log('');
-              }
-              return;
-            }
-
-            try {
-              const parsed = JSON.parse(data);
-              const delta = parsed.choices?.[0]?.delta;
-              if (delta?.content) {
-                chunkCount++;
-                outputBuffer += delta.content;
-                yield delta.content;
-              } else if (delta?.reasoning_content) {
-                chunkCount++;
-                outputBuffer += delta.reasoning_content;
-                yield delta.reasoning_content;
-              }
-            } catch (e) {
-              // Silently ignore parsing errors
-            }
-          }
-        }
-      }
-
-      if (buffer.trim()) {
-        const trimmedLine = buffer.trim();
-        if (trimmedLine.startsWith('data: ')) {
-          const data = trimmedLine.slice(6);
-          if (data !== '[DONE]') {
-            try {
-              const parsed = JSON.parse(data);
-              const delta = parsed.choices?.[0]?.delta;
-              if (delta?.content) {
-                yield delta.content;
-              } else if (delta?.reasoning_content) {
-                yield delta.reasoning_content;
-              }
-            } catch (e) {
-              // Ignore final parsing errors
-            }
-          }
-        }
-      }
-      
-      console.log('\n╔══════════════════════════════════════════════════════════╗');
-      console.log('║              STREAM COMPLETED                            ║');
-      console.log('╚══════════════════════════════════════════════════════════╝\n');
-    } catch (error: any) {
-      if (error.response) {
-        throw new Error(
-          `API Error: ${error.response.status} - ${JSON.stringify(error.response.data)}`
-        );
-      } else if (error.request) {
-        throw new Error('Network error: No response received from server');
-      } else {
-        throw new Error(`Request error: ${error.message}`);
-      }
-    }
-  }
-
-  // Anthropic native streaming response（/v1/messages 端点）
-  private async *anthropicNativeStreamChatCompletion(
-    messages: Message[],
-    options: ChatCompletionOptions = {}
-  ): AsyncGenerator<string, void, unknown> {
-    const { system, messages: anthropicMessages } = this.convertToAnthropicFormat(messages);
-
-    const requestBody: any = {
-      model: options.model || this.authConfig.modelName || 'claude-sonnet-4-20250514',
-      messages: anthropicMessages,
-      temperature: options.temperature ?? 1.0,
-      stream: true,
-      max_tokens: options.maxTokens || 4096
-    };
-
-    if (system) {
-      requestBody.system = system;
-    }
-
-    // Anthropic native tool format
-    if (options.tools && options.tools.length > 0) {
-      requestBody.tools = options.tools.map(tool => ({
-        name: tool.function.name,
-        description: tool.function.description,
-        input_schema: tool.function.parameters || { type: 'object', properties: {} }
-      }));
-      
-      const toolChoice = options.toolChoice;
-      if (toolChoice === 'none') {
-        requestBody.tool_choice = { type: 'auto' };
-      } else if (toolChoice && typeof toolChoice === 'object') {
-        if (toolChoice.type === 'function' && toolChoice.function) {
-          requestBody.tool_choice = { type: 'tool', tool: { name: toolChoice.function.name } };
-        } else {
-          requestBody.tool_choice = { type: 'auto' };
-        }
-      } else {
-        requestBody.tool_choice = { type: 'auto' };
-      }
-    }
-
-    if (options.thinkingTokens && options.thinkingTokens > 0) {
-      requestBody.thinking = { type: 'enabled', budget_tokens: options.thinkingTokens };
-    }
-
-    // Debug output（受showAIDebugInfo配置控制）
-    const showDebug = this.authConfig.showAIDebugInfo ?? false;
-    
-    if (showDebug) {
-      console.log('\n╔══════════════════════════════════════════════════════════╗');
-      console.log('║         AI REQUEST DEBUG (ANTHROPIC STREAM)             ║');
-      console.log('╚══════════════════════════════════════════════════════════╝');
-      console.log(`📦 Model: ${requestBody.model}`);
-      console.log(`🌐 Base URL: ${this.authConfig.baseUrl}`);
-      console.log(`💬 Total Messages: ${anthropicMessages.length} items`);
-      if (requestBody.temperature) console.log(`🌡️  Temperature: ${requestBody.temperature}`);
-      if (requestBody.max_tokens) console.log(`📏 Max Tokens: ${requestBody.max_tokens}`);
-      if (requestBody.thinking) console.log(`🧠 Thinking Budget: ${requestBody.thinking.budget_tokens}`);
-      console.log('─'.repeat(60));
-      
-      // Display system messages
-      if (system) {
-        console.log('\n┌─────────────────────────────────────────────────────────────┐');
-        console.log('│ 🟫 SYSTEM                                                     │');
-        console.log('├─────────────────────────────────────────────────────────────┤');
-        console.log(renderMarkdown(system).split('\n').map(l => '│ ' + l).join('\n'));
-        console.log('└─────────────────────────────────────────────────────────────┘');
-      }
-      
-      displayMessages(anthropicMessages);
-      
-      console.log('\n📤 Starting Anthropic stream...\n');
-    }
-
-    try {
-      // Anthropic native streaming endpoint /v1/messages
-      const response = await this.client.post('/v1/messages', requestBody, {
-        responseType: 'stream'
-      });
-
-      console.log('📥 Receiving Anthropic stream chunks...\n');
-
-      let buffer = '';
-      let outputBuffer = '';
-
-      for await (const chunk of response.data) {
-        buffer += chunk.toString();
-        const lines = buffer.split('\n');
-        buffer = lines.pop() || '';
-
-        for (const line of lines) {
-          const trimmedLine = line.trim();
-          if (!trimmedLine) continue;
-
-          // Anthropic streaming format: data: {"type":"content_block_delta",...}
-          if (trimmedLine.startsWith('data: ')) {
-            const data = trimmedLine.slice(6);
-
-            try {
-              const parsed = JSON.parse(data);
-
-              // Anthropic event types
-              if (parsed.type === 'content_block_delta') {
-                if (parsed.delta?.type === 'text_delta' && parsed.delta.text) {
-                  outputBuffer += parsed.delta.text;
-                  yield parsed.delta.text;
-                } else if (parsed.delta?.type === 'thinking_delta' && parsed.delta.thinking) {
-                  outputBuffer += parsed.delta.thinking;
-                  yield parsed.delta.thinking;
-                }
-              } else if (parsed.type === 'message_delta') {
-                if (parsed.delta?.stop_reason) {
-                  // Message end
-                  if (showDebug) {
-                    console.log('\n╔══════════════════════════════════════════════════════════╗');
-                    console.log('║              STREAM COMPLETED                            ║');
-                    console.log('╚══════════════════════════════════════════════════════════╝');
-                    console.log(`📏 Total output: ${outputBuffer.length} chars`);
-                    
-                    console.log('\n┌─────────────────────────────────────────────────────────────┐');
-                    console.log('│ 🤖 ASSISTANT OUTPUT                                        │');
-                    console.log('├─────────────────────────────────────────────────────────────┤');
-                    console.log('│ 💬 CONTENT:');
-                    console.log('│ ───────────────────────────────────────────────────────────');
-                    const lines = renderMarkdown(outputBuffer).split('\n');
-                    for (const line of lines.slice(0, 30)) {
-                      console.log('│ ' + line.slice(0, 62));
-                    }
-                    if (lines.length > 30) {
-                      console.log(`│ ... (${lines.length - 30} more lines)`);
-                    }
-                    console.log('└─────────────────────────────────────────────────────────────┘');
-                    console.log('');
-                  }
-                  return;
-                }
-              }
-            } catch (e) {
-              // Ignore parsing errors
-            }
-          }
-        }
-      }
-      
-      console.log('\n╔══════════════════════════════════════════════════════════╗');
-      console.log('║              STREAM COMPLETED                            ║');
-      console.log('╚══════════════════════════════════════════════════════════╝\n');
-    } catch (error: any) {
-      if (error.response) {
-        throw new Error(
-          `API Error: ${error.response.status} - ${JSON.stringify(error.response.data)}`
-        );
-      } else if (error.request) {
-        throw new Error('Network error: No response received from server');
-      } else {
-        throw new Error(`Request error: ${error.message}`);
-      }
-    }
-  }
-
-  // MiniMax streaming response（Automatically select based on baseUrl格式）
-  private async *minimaxStreamChatCompletion(
-    messages: Message[],
-    options: ChatCompletionOptions = {}
-  ): AsyncGenerator<string, void, unknown> {
-    const { system, messages: anthropicMessages } = this.convertToAnthropicFormat(messages);
-    const { endpoint, format } = getMiniMaxEndpoint(this.authConfig.baseUrl || '');
-
-    const requestBody: any = {
-      model: options.model || this.authConfig.modelName || 'MiniMax-M2',
-      messages: format === 'anthropic' ? anthropicMessages : messages,
-      temperature: options.temperature ?? 1.0,
-      stream: true,
-      max_tokens: options.maxTokens || 4096
-    };
-
-    if (system && format === 'anthropic') {
-      requestBody.system = system;
-    }
-
-    if (format === 'anthropic') {
-      // Anthropic 格式的工具
-      if (options.tools && options.tools.length > 0) {
-        requestBody.tools = options.tools.map(tool => ({
-          name: tool.function.name,
-          description: tool.function.description,
-          input_schema: tool.function.parameters || { type: 'object', properties: {} }
-        }));
-        
-        const toolChoice = options.toolChoice;
-        if (toolChoice === 'none') {
-          requestBody.tool_choice = { type: 'auto' };
-        } else if (toolChoice && typeof toolChoice === 'object') {
-          if (toolChoice.type === 'function' && toolChoice.function) {
-            requestBody.tool_choice = { type: 'tool', tool: { name: toolChoice.function.name } };
-          } else {
-            requestBody.tool_choice = { type: 'auto' };
-          }
-        } else {
-          requestBody.tool_choice = { type: 'auto' };
-        }
-      }
-    } else {
-      // OpenAI format tools
-      if (options.tools && options.tools.length > 0) {
-        requestBody.tools = options.tools;
-        requestBody.tool_choice = options.toolChoice || 'auto';
-      }
-    }
-
-    console.log('\n╔══════════════════════════════════════════════════════════╗');
-    console.log('║            AI REQUEST DEBUG (MINIMAX STREAM)             ║');
-    console.log('╚══════════════════════════════════════════════════════════╝');
-    console.log(`📦 Model: ${requestBody.model}`);
-    console.log(`🔗 Format: ${format.toUpperCase()} | Endpoint: ${endpoint}`);
-    console.log(`🌐 Base URL: ${this.authConfig.baseUrl}`);
-    console.log(`💬 Total Messages: ${requestBody.messages.length} items`);
-    if (requestBody.temperature) console.log(`🌡️  Temperature: ${requestBody.temperature}`);
-    if (requestBody.max_tokens) console.log(`📏 Max Tokens: ${requestBody.max_tokens}`);
-    console.log('─'.repeat(60));
-    
-    // Display system messages
-    if (system && format === 'anthropic') {
-      console.log('\n┌─────────────────────────────────────────────────────────────┐');
-      console.log('│ 🟫 SYSTEM                                                     │');
-      console.log('├─────────────────────────────────────────────────────────────┤');
-      console.log(renderMarkdown(system).split('\n').map(l => '│ ' + l).join('\n'));
-      console.log('└─────────────────────────────────────────────────────────────┘');
-    }
-    
-    displayMessages(requestBody.messages);
-    
-    console.log('\n📤 Starting MiniMax stream...\n');
-
-    try {
-      // MiniMax uses correct endpoint
-      const response = await this.client.post(endpoint, requestBody, {
-        responseType: 'stream'
-      });
-
-      console.log('📥 Receiving MiniMax stream chunks...\n');
-
-      let buffer = '';
-      let outputBuffer = '';
-
-      for await (const chunk of response.data) {
-        buffer += chunk.toString();
-        const lines = buffer.split('\n');
-        buffer = lines.pop() || '';
-
-        for (const line of lines) {
-          const trimmedLine = line.trim();
-          if (!trimmedLine) continue;
-
-          // Parse different streaming responses based on format
-          if (format === 'anthropic') {
-            // Anthropic SSE format: data: {"type":"content_block_delta",...}
-            if (trimmedLine.startsWith('data: ')) {
-              const data = trimmedLine.slice(6);
-
-              try {
-                const parsed = JSON.parse(data);
-
-                if (parsed.type === 'content_block_delta') {
-                  if (parsed.delta?.type === 'text_delta' && parsed.delta.text) {
-                    yield parsed.delta.text;
-                  } else if (parsed.delta?.type === 'thinking_delta' && parsed.delta.thinking) {
-                    yield parsed.delta.thinking;
-                  }
-                } else if (parsed.type === 'message_delta') {
-                  if (parsed.delta?.stop_reason) {
-                    console.log('\n╔══════════════════════════════════════════════════════════╗');
-                    console.log('║              STREAM COMPLETED                            ║');
-                    console.log('╚══════════════════════════════════════════════════════════╝');
-                    console.log(`📏 Total output: ${outputBuffer.length} chars`);
-                    
-                    console.log('\n┌─────────────────────────────────────────────────────────────┐');
-                    console.log('│ 🤖 ASSISTANT OUTPUT                                        │');
-                    console.log('├─────────────────────────────────────────────────────────────┤');
-                    console.log('│ 💬 CONTENT:');
-                    console.log('│ ───────────────────────────────────────────────────────────');
-                    const lines = renderMarkdown(outputBuffer).split('\n');
-                    for (const line of lines.slice(0, 30)) {
-                      console.log('│ ' + line.slice(0, 62));
-                    }
-                    if (lines.length > 30) {
-                      console.log(`│ ... (${lines.length - 30} more lines)`);
-                    }
-                    console.log('└─────────────────────────────────────────────────────────────┘');
-                    console.log('');
-                    return;
-                  }
-                }
-              } catch (e) {
-                // Ignore parsing errors
-              }
-            }
-          } else {
-            // OpenAI SSE format: data: {...}
-            if (trimmedLine.startsWith('data: ')) {
-              const data = trimmedLine.slice(6);
-              if (data === '[DONE]') continue;
-
-              try {
-                const parsed = JSON.parse(data);
-                const delta = parsed.choices?.[0]?.delta;
-                if (delta?.content) {
-                  outputBuffer += delta.content;
-                  yield delta.content;
-                } else if (delta?.reasoning_content) {
-                  outputBuffer += delta.reasoning_content;
-                  yield delta.reasoning_content;
-                }
-              } catch (e) {
-                // Ignore parsing errors
-              }
-            }
-          }
-        }
-      }
-      
-      console.log('\n╔══════════════════════════════════════════════════════════╗');
-      console.log('║              STREAM COMPLETED                            ║');
-      console.log('╚══════════════════════════════════════════════════════════╝\n');
-    } catch (error: any) {
-      if (error.response) {
-        throw new Error(
-          `API Error: ${error.response.status} - ${JSON.stringify(error.response.data)}`
-        );
-      } else if (error.request) {
-        throw new Error('Network error: No response received from server');
-      } else {
-        throw new Error(`Request error: ${error.message}`);
-      }
-    }
-  }
-
-  async listModels(): Promise<any[]> {
-    try {
-      const response = await this.client.get('/models');
-      return response.data.data || [];
-    } catch (error: any) {
-      console.error('Failed to list models:', error);
-      return [];
-    }
-  }
-
-  updateAuthConfig(authConfig: AuthConfig): void {
-    this.authConfig = authConfig;
-    this.client.defaults.baseURL = authConfig.baseUrl;
-    
-    const isMiniMax = detectMiniMaxAPI(authConfig.baseUrl || '');
-    const isAnthropic = !isMiniMax && isAnthropicCompatible(authConfig.baseUrl || '');
-    
-    if (isMiniMax || isAnthropic) {
-      // MiniMax/Anthropic: Use x-api-key auth header
-      this.client.defaults.headers['x-api-key'] = authConfig.apiKey || '';
-      this.client.defaults.headers['anthropic-version'] = '2023-06-01';
-      // Clear Bearer header
-      delete this.client.defaults.headers['Authorization'];
-    } else {
-      // OpenAI compatible: Use Bearer token
-      this.client.defaults.headers['Authorization'] = `Bearer ${authConfig.apiKey}`;
-      // Clear x-api-key header
-      delete this.client.defaults.headers['x-api-key'];
-      delete this.client.defaults.headers['anthropic-version'];
-    }
-  }
-
-  getAuthConfig(): AuthConfig {
-    return { ...this.authConfig };
-  }
-
-  // Check if messages contain tool calls
-  hasToolCalls(messages: Message[]): boolean {
-    return messages.some(msg => {
-      if (msg.tool_calls && msg.tool_calls.length > 0) return true;
-      if (Array.isArray(msg.content)) {
-        return msg.content.some(block => 
-          block.type === 'tool_use' || 
-          (block as any).type === 'tool_result'
-        );
-      }
-      return false;
-    });
-  }
-}
-
-export function detectThinkingKeywords(text: string): 'none' | 'normal' | 'hard' | 'mega' | 'ultra' {
-  const ultraKeywords = ['super think', 'extreme think', 'deep think', 'full think', 'ultra think', 'careful think',
-    'ultrathink', 'think really super hard', 'think intensely'];
-  const megaKeywords = ['strong think', 'powerful think', 'think hard', 'try hard to think', 'think well', 'think carefully',
-    'megathink', 'think really hard', 'think a lot'];
-  const hardKeywords = ['think again', 'think more', 'think clearly', 'think thoroughly', 'consider carefully',
-    'think about it', 'think more', 'think harder'];
-  const normalKeywords = ['think', 'think', 'consider', 'think'];
-
-  const lowerText = text.toLowerCase();
-
-  if (ultraKeywords.some(keyword => lowerText.includes(keyword.toLowerCase()))) {
-    return 'ultra';
-  } else if (megaKeywords.some(keyword => lowerText.includes(keyword.toLowerCase()))) {
-    return 'mega';
-  } else if (hardKeywords.some(keyword => lowerText.includes(keyword.toLowerCase()))) {
-    return 'hard';
-  } else if (normalKeywords.some(keyword => lowerText.includes(keyword.toLowerCase()))) {
-    return 'normal';
-  }
-
-  return 'none';
-}
-
-export function getThinkingTokens(mode: 'none' | 'normal' | 'hard' | 'mega' | 'ultra'): number {
-  const tokensMap = {
-    none: 0,
-    normal: 2000,
-    hard: 4000,
-    mega: 10000,
-    ultra: 32000
-  };
-  return tokensMap[mode];
+import axios, { AxiosInstance } from 'axios';
+import https from 'https';
+import { AuthConfig } from './types.js';
+import { withRetry, RetryConfig } from './retry.js';
+
+// Message content block type for Anthropic format
+export interface AnthropicContentBlock {
+  type: 'text' | 'tool_use' | 'tool_result' | 'thinking';
+  text?: string;
+  id?: string;
+  name?: string;
+  input?: any;
+  tool_use_id?: string;
+  content?: string;
+  thinking?: string;
+}
+
+// Markdown rendering helper function
+export function renderMarkdown(text: string): string {
+  // Code block rendering
+  text = text.replace(/```(\w*)\n([\s\S]*?)```/g, (_, lang, code) => {
+    return `\n┌─[${lang || 'code'}]\n${code.trim().split('\n').map((l: string) => '│ ' + l).join('\n')}\n└─\n`;
+  });
+
+  // Inline code rendering
+  text = text.replace(/`([^`]+)`/g, '`$1`');
+
+  // Bold rendering
+  text = text.replace(/\*\*([^*]+)\*\*/g, '●$1○');
+
+  // Italic rendering
+  text = text.replace(/\*([^*]+)\*/g, '/$1/');
+
+  // List rendering
+  text = text.replace(/^- (.*$)/gm, '○ $1');
+  text = text.replace(/^\d+\. (.*$)/gm, '• $1');
+
+  // Heading rendering
+  text = text.replace(/^### (.*$)/gm, '\n━━━ $1 ━━━\n');
+  text = text.replace(/^## (.*$)/gm, '\n━━━━━ $1 ━━━━━\n');
+  text = text.replace(/^# (.*$)/gm, '\n━━━━━━━ $1 ━━━━━━━\n');
+
+  // Quote rendering
+  text = text.replace(/^> (.*$)/gm, '│ │ $1');
+
+  // Link rendering
+  text = text.replace(/\[([^\]]+)\]\(([^)]+)\)/g, '[$1]($2)');
+
+  return text;
+}
+
+// Format message content
+function formatMessageContent(content: string | Array<any>): string {
+  if (typeof content === 'string') {
+    return renderMarkdown(content);
+  }
+
+  const parts: string[] = [];
+  let hasToolUse = false;
+
+  for (const block of content) {
+    if (block.type === 'text') {
+      parts.push(renderMarkdown(block.text || ''));
+    } else if (block.type === 'tool_use') {
+      hasToolUse = true;
+      parts.push(`[🔧 TOOL CALL PENDING: ${block.name}]`);
+    } else if (block.type === 'tool_result') {
+      const result = typeof block.content === 'string' ? block.content : JSON.stringify(block.content);
+      parts.push(`[✅ TOOL RESULT]\n${result}`);
+    } else if (block.type === 'thinking') {
+      parts.push(`[🧠 THINKING]\n${block.thinking || ''}`);
+    }
+  }
+
+  if (hasToolUse) {
+    parts.push('\n[⚠️  Note: Tool calls are executed by the framework, not displayed here]');
+  }
+
+  return parts.join('\n');
+}
+
+// Display messages by category
+export function displayMessages(messages: any[], systemPrompt?: string): void {
+  const roleColors: Record<string, string> = {
+    system: '🟫 SYSTEM',
+    user: '👤 USER',
+    assistant: '🤖 ASSISTANT',
+    tool: '🔧 TOOL'
+  };
+
+  // Display system message first (if there's a separate systemPrompt parameter)
+  if (systemPrompt) {
+    console.log('\n┌─────────────────────────────────────────────────────────────┐');
+    console.log('│ 🟫 SYSTEM                                                     │');
+    console.log('├─────────────────────────────────────────────────────────────┤');
+    console.log(renderMarkdown(systemPrompt).split('\n').map((l: string) => '│ ' + l).join('\n'));
+    console.log('└─────────────────────────────────────────────────────────────┘');
+  }
+
+  // Iterate through all messages
+  for (let i = 0; i < messages.length; i++) {
+    const msg = messages[i];
+    const role = msg.role as string;
+    const roleLabel = roleColors[role] || `● ${role.toUpperCase()}`;
+
+    console.log(`\n┌─────────────────────────────────────────────────────────────┐`);
+    console.log(`│ ${roleLabel} (${i + 1}/${messages.length})                                          │`);
+    console.log('├─────────────────────────────────────────────────────────────┤');
+
+    // Display reasoning_content (if present) - check both camelCase and snake_case
+    const reasoningContent = (msg as any).reasoningContent || (msg as any).reasoning_content;
+    if (reasoningContent) {
+      console.log('│ 🧠 REASONING:');
+      console.log('│ ───────────────────────────────────────────────────────────');
+      const reasoningLines = renderMarkdown(reasoningContent).split('\n');
+      for (const line of reasoningLines.slice(0, 20)) {
+        console.log('│ ' + line.slice(0, 62));
+      }
+      if (reasoningContent.length > 1000) console.log('│ ... (truncated)');
+      console.log('│ ───────────────────────────────────────────────────────────');
+    }
+
+    // Display main content
+    const content = formatMessageContent(msg.content);
+    const lines = content.split('\n');
+
+    for (const line of lines.slice(0, 50)) {
+      console.log('│ ' + line.slice(0, 62));
+    }
+    if (lines.length > 50) {
+      console.log('│ ... (' + (lines.length - 50) + ' more lines)');
+    }
+
+    console.log('└─────────────────────────────────────────────────────────────┘');
+  }
+}
+
+// Format response content
+function formatResponseContent(content: string | Array<any>): string {
+  if (typeof content === 'string') {
+    return renderMarkdown(content);
+  }
+
+  const parts: string[] = [];
+  let hasToolUse = false;
+
+  for (const block of content) {
+    if (block.type === 'text') {
+      parts.push(renderMarkdown(block.text || ''));
+    } else if (block.type === 'tool_use') {
+      hasToolUse = true;
+      // Tool calls are handled via tool_calls field, not shown here
+    } else if (block.type === 'tool_result') {
+      const result = typeof block.content === 'string' ? block.content : JSON.stringify(block.content);
+      parts.push(`[✅ TOOL RESULT]\n${result}`);
+    } else if (block.type === 'thinking') {
+      parts.push(`[🧠 THINKING]\n${block.thinking || ''}`);
+    } else if (block.type === 'image') {
+      parts.push('[IMAGE]');
+    }
+  }
+
+  if (hasToolUse) {
+    parts.push('\n[⚠️  Note: Tool calls are executed via tool_calls field, not shown here]');
+  }
+
+  return parts.join('\n');
+}
+
+export interface Message {
+  role: 'system' | 'user' | 'assistant' | 'tool';
+  content: string | Array<AnthropicContentBlock | { type: string; text?: string; image_url?: { url: string } }>;
+  reasoning_content?: string;
+  tool_calls?: any[];
+  tool_call_id?: string;
+}
+
+export interface ToolDefinition {
+  type: 'function';
+  function: {
+    name: string;
+    description: string;
+    parameters?: any;
+  };
+}
+
+export interface ChatCompletionOptions {
+  model?: string;
+  temperature?: number;
+  maxTokens?: number;
+  tools?: ToolDefinition[];
+  toolChoice?: 'auto' | 'none' | { type: string; function: { name: string } };
+  stream?: boolean;
+  thinkingTokens?: number;
+}
+
+export interface ChatCompletionResponse {
+  id: string;
+  object: string;
+  created: number;
+  model: string;
+  choices: Array<{
+    index: number;
+    message: Message;
+    finish_reason: string;
+  }>;
+  usage?: {
+    prompt_tokens: number;
+    completion_tokens: number;
+    total_tokens: number;
+  };
+}
+
+// Detect if it's Anthropic compatible API（Use x-api-key authentication header）
+function isAnthropicCompatible(baseUrl: string): boolean {
+  return baseUrl.includes('anthropic') || 
+         baseUrl.includes('minimaxi.com') ||
+         baseUrl.includes('minimax.chat');
+}
+
+// MiniMax API path detection
+function detectMiniMaxAPI(baseUrl: string): boolean {
+  return baseUrl.includes('minimax.chat') || 
+         baseUrl.includes('minimaxi.com');
+}
+
+// Get correct endpoint path for MiniMax
+function getMiniMaxEndpoint(baseUrl: string): { endpoint: string; format: 'anthropic' | 'openai' } {
+  // MiniMax Anthropic format: https://api.minimax.chat/anthropic + /v1/messages
+  if (baseUrl.includes('/anthropic')) {
+    return { endpoint: '/v1/messages', format: 'anthropic' };
+  }
+  // MiniMax OpenAI format: https://api.minimaxi.com/v1 + /chat/completions
+  if (baseUrl.includes('/v1') && !baseUrl.includes('/anthropic')) {
+    return { endpoint: '/chat/completions', format: 'openai' };
+  }
+  // Default to Anthropic format
+  return { endpoint: '/v1/messages', format: 'anthropic' };
+}
+
+export class AIClient {
+  private client: AxiosInstance;
+  private authConfig: AuthConfig;
+
+  constructor(authConfig: AuthConfig) {
+    this.authConfig = authConfig;
+    const isMiniMax = detectMiniMaxAPI(authConfig.baseUrl || '');
+    const isAnthropicOfficial = !isMiniMax && isAnthropicCompatible(authConfig.baseUrl || '');
+    
+    const headers: Record<string, string> = {
+      'Content-Type': 'application/json'
+    };
+    
+    if (isMiniMax) {
+      // MiniMax: Use x-api-key authentication header
+      headers['x-api-key'] = authConfig.apiKey || '';
+      headers['anthropic-version'] = '2023-06-01';
+    } else if (isAnthropicOfficial) {
+      // Anthropic official: Use x-api-key authentication header
+      headers['x-api-key'] = authConfig.apiKey || '';
+      headers['anthropic-version'] = '2023-06-01';
+      headers['anthropic-dangerous-direct-browser-access'] = 'true';
+    } else {
+      // Other OpenAI compatible: 使用 Bearer token
+      headers['Authorization'] = `Bearer ${authConfig.apiKey}`;
+    }
+    
+    this.client = axios.create({
+      baseURL: authConfig.baseUrl,
+      headers,
+      timeout: 300000,
+      httpsAgent: new https.Agent({ rejectUnauthorized: false })
+    });
+  }
+
+  // Convert OpenAI format messages to Anthropic format
+  private convertToAnthropicFormat(
+    messages: Message[],
+    systemPrompt?: string
+  ): { system: string; messages: Array<{ role: string; content: AnthropicContentBlock[] }> } {
+    const systemMessages = messages.filter(m => m.role === 'system');
+    const otherMessages = messages.filter(m => m.role !== 'system');
+    
+    const systemContent = systemMessages[0]?.content;
+    const system = systemPrompt || (typeof systemContent === 'string' ? systemContent : '');
+
+    const anthropicMessages: Array<{ role: string; content: AnthropicContentBlock[] }> = [];
+
+    for (const msg of otherMessages) {
+      const blocks: AnthropicContentBlock[] = [];
+
+      if (typeof msg.content === 'string') {
+        blocks.push({ type: 'text', text: msg.content });
+      } else if (Array.isArray(msg.content)) {
+        for (const block of msg.content) {
+          if (block.type === 'text' && 'text' in block) {
+            blocks.push({ type: 'text', text: (block as any).text });
+          } else if (block.type === 'tool_use') {
+            blocks.push({
+              type: 'tool_use',
+              id: (block as any).id,
+              name: (block as any).function?.name || (block as any).name,
+              input: (block as any).function?.arguments || (block as any).input
+            });
+          } else if (block.type === 'tool_result') {
+            blocks.push({
+              type: 'tool_result',
+              tool_use_id: (block as any).tool_call_id || (block as any).tool_use_id,
+              content: typeof (block as any).content === 'string' 
+                ? (block as any).content 
+                : JSON.stringify((block as any).content)
+            });
+          } else if (block.type === 'thinking') {
+            blocks.push({ type: 'thinking', thinking: (block as any).thinking });
+          }
+        }
+      }
+
+      // Handle tool_calls (OpenAI 格式)
+      if (msg.tool_calls) {
+        for (const tc of msg.tool_calls) {
+          blocks.push({
+            type: 'tool_use',
+            id: tc.id,
+            name: tc.function?.name,
+            input: tc.function?.arguments ? (typeof tc.function.arguments === 'string' ? JSON.parse(tc.function.arguments) : tc.function.arguments) : {}
+          });
+        }
+      }
+
+      if (blocks.length > 0) {
+        anthropicMessages.push({
+          role: msg.role === 'tool' ? 'user' : msg.role,
+          content: blocks as AnthropicContentBlock[]
+        });
+      }
+    }
+
+    return { system, messages: anthropicMessages };
+  }
+
+  async chatCompletion(
+    messages: Message[],
+    options: ChatCompletionOptions = {}
+  ): Promise<ChatCompletionResponse> {
+    const model = options.model || this.authConfig.modelName || 'gpt-4';
+    const isMiniMax = detectMiniMaxAPI(this.authConfig.baseUrl || '');
+
+    if (isMiniMax) {
+      return this.minimaxChatCompletion(messages, options);
+    }
+
+    const isAnthropic = isAnthropicCompatible(this.authConfig.baseUrl || '');
+    if (isAnthropic) {
+      return this.anthropicNativeChatCompletion(messages, options);
+    }
+
+    // OpenAI format request
+    const requestBody: any = {
+      model,
+      messages,
+      temperature: options.temperature ?? 0.7,
+      stream: options.stream ?? false
+    };
+
+    if (options.maxTokens && options.maxTokens > 0) {
+      requestBody.max_tokens = options.maxTokens;
+    }
+
+    if (options.tools && options.tools.length > 0) {
+      requestBody.tools = options.tools;
+      requestBody.tool_choice = options.toolChoice || 'auto';
+    }
+
+    if (options.thinkingTokens && options.thinkingTokens > 0) {
+      requestBody.max_completion_tokens = options.thinkingTokens;
+    }
+
+    // Debug output（受showAIDebugInfo配置控制）
+    const showDebug = this.authConfig.showAIDebugInfo ?? false;
+    
+    if (showDebug) {
+      console.log('\n╔══════════════════════════════════════════════════════════╗');
+      console.log('║                    AI REQUEST DEBUG                      ║');
+      console.log('╚══════════════════════════════════════════════════════════╝');
+      console.log(`📦 Model: ${model}`);
+      console.log(`🌐 Base URL: ${this.authConfig.baseUrl}`);
+      console.log(`💬 Total Messages: ${messages.length} items`);
+      if (options.temperature !== undefined) console.log(`🌡️  Temperature: ${options.temperature}`);
+      if (options.maxTokens) console.log(`📏 Max Tokens: ${options.maxTokens}`);
+      if (options.tools?.length) console.log(`🔧 Tools: ${options.tools.length} items`);
+      if (options.thinkingTokens) console.log(`🧠 Thinking Tokens: ${options.thinkingTokens}`);
+      console.log('─'.repeat(60));
+      
+      // Separate system messages
+      const systemMsgs = messages.filter(m => m.role === 'system');
+      const otherMsgs = messages.filter(m => m.role !== 'system');
+      
+      if (systemMsgs.length > 0) {
+        const systemContent = typeof systemMsgs[0].content === 'string' 
+          ? systemMsgs[0].content 
+          : formatMessageContent(systemMsgs[0].content);
+        console.log('\n┌─────────────────────────────────────────────────────────────┐');
+        console.log('│ 🟫 SYSTEM                                                     │');
+        console.log('├─────────────────────────────────────────────────────────────┤');
+        console.log(renderMarkdown(systemContent).split('\n').map(l => '│ ' + l).join('\n'));
+        console.log('└─────────────────────────────────────────────────────────────┘');
+      }
+      
+      displayMessages(otherMsgs);
+      
+      console.log('\n📤 Sending request to API...\n');
+    }
+
+    try {
+      const response = await this.client.post('/chat/completions', requestBody);
+      
+      if (showDebug) {
+        console.log('\n╔══════════════════════════════════════════════════════════╗');
+        console.log('║                   AI RESPONSE DEBUG                      ║');
+        console.log('╚══════════════════════════════════════════════════════════╝');
+        console.log(`🆔 ID: ${response.data.id}`);
+        console.log(`🤖 Model: ${response.data.model}`);
+        const usage = response.data.usage;
+        if (usage) {
+          console.log(`📊 Tokens: ${usage.prompt_tokens} (prompt) + ${usage.completion_tokens} (completion) = ${usage.total_tokens} (total)`);
+        }
+        const choice = response.data.choices?.[0];
+        if (choice) {
+          console.log(`🏁 Finish Reason: ${choice.finish_reason}`);
+          
+          console.log('\n┌─────────────────────────────────────────────────────────────┐');
+          console.log('│ 🤖 ASSISTANT                                                 │');
+          console.log('├─────────────────────────────────────────────────────────────┤');
+          
+          // Display reasoning_content（如果有）
+          if (choice.message.reasoning_content) {
+            console.log('│ 🧠 REASONING:');
+            console.log('│ ───────────────────────────────────────────────────────────');
+            const reasoningLines = renderMarkdown(choice.message.reasoning_content).split('\n');
+            for (const line of reasoningLines.slice(0, 15)) {
+              console.log('│ ' + line.slice(0, 62));
+            }
+            if (choice.message.reasoning_content.length > 800) console.log('│ ... (truncated)');
+            console.log('│ ───────────────────────────────────────────────────────────');
+          }
+          
+          // Display main content
+          const content = formatResponseContent(choice.message.content);
+          const lines = content.split('\n');
+          console.log('│ 💬 CONTENT:');
+          console.log('│ ───────────────────────────────────────────────────────────');
+          for (const line of lines.slice(0, 40)) {
+            console.log('│ ' + line.slice(0, 62));
+          }
+          if (lines.length > 40) {
+            console.log(`│ ... (${lines.length - 40} more lines)`);
+          }
+          console.log('└─────────────────────────────────────────────────────────────┘');
+        }
+        console.log('╔══════════════════════════════════════════════════════════╗');
+        console.log('║                    RESPONSE ENDED                        ║');
+        console.log('╚══════════════════════════════════════════════════════════╝\n');
+      }
+      
+      return response.data;
+    } catch (error: any) {
+      // Check if error is retryable (timeout, network error, or 5xx)
+      const isRetryable = this.isRetryableError(error);
+      if (!isRetryable) {
+        if (error.response) {
+          throw new Error(
+            `API Error: ${error.response.status} - ${JSON.stringify(error.response.data)}`
+          );
+        } else if (error.request) {
+          throw new Error('Network error: No response received from server');
+        } else {
+          throw new Error(`Request error: ${error.message}`);
+        }
+      }
+
+      // Retry with exponential backoff
+      const retryResult = await withRetry(async () => {
+        const response = await this.client.post('/chat/completions', requestBody);
+        if (showDebug) {
+          console.log('\n╔══════════════════════════════════════════════════════════╗');
+          console.log('║                   AI RESPONSE DEBUG (RETRY)              ║');
+          console.log('╚══════════════════════════════════════════════════════════╝');
+          console.log(`🆔 ID: ${response.data.id}`);
+          console.log(`🤖 Model: ${response.data.model}`);
+          const usage = response.data.usage;
+          if (usage) {
+            console.log(`📊 Tokens: ${usage.prompt_tokens} (prompt) + ${usage.completion_tokens} (completion) = ${usage.total_tokens} (total)`);
+          }
+          console.log('╔══════════════════════════════════════════════════════════╗');
+          console.log('║                    RESPONSE ENDED                        ║');
+          console.log('╚══════════════════════════════════════════════════════════╝\n');
+        }
+        return response.data;
+      }, { maxRetries: 3, baseDelay: 1000, maxDelay: 10000, jitter: true });
+
+      if (!retryResult.success) {
+        throw retryResult.error || new Error('Retry failed');
+      }
+
+      if (!retryResult.data) {
+        throw new Error('Retry returned empty response');
+      }
+
+      return retryResult.data;
+    }
+  }
+
+  private isRetryableError(error: any): boolean {
+    // Timeout or network error (no response received)
+    if (error.code === 'ECONNABORTED' || !error.response) {
+      return true;
+    }
+    // 5xx server errors
+    if (error.response?.status && error.response.status >= 500) {
+      return true;
+    }
+    // 429 rate limit
+    if (error.response?.status === 429) {
+      return true;
+    }
+    return false;
+  }
+
+  // Anthropic official原生 API（使用 /v1/messages 端点）
+  private async anthropicNativeChatCompletion(
+    messages: Message[],
+    options: ChatCompletionOptions = {}
+  ): Promise<ChatCompletionResponse> {
+    const { system, messages: anthropicMessages } = this.convertToAnthropicFormat(messages);
+
+    const requestBody: any = {
+      model: options.model || this.authConfig.modelName || 'claude-sonnet-4-20250514',
+      messages: anthropicMessages,
+      temperature: options.temperature ?? 1.0,
+      stream: false,
+      max_tokens: options.maxTokens || 4096
+    };
+
+    if (system) {
+      requestBody.system = system;
+    }
+
+    // Anthropic native tool format
+    if (options.tools && options.tools.length > 0) {
+      requestBody.tools = options.tools.map(tool => ({
+        name: tool.function.name,
+        description: tool.function.description,
+        input_schema: tool.function.parameters || { type: 'object', properties: {} }
+      }));
+      
+      // Convert tool_choice 从 OpenAI 格式到 Anthropic 格式
+      const toolChoice = options.toolChoice;
+      if (toolChoice === 'none') {
+        requestBody.tool_choice = { type: 'auto' };
+      } else if (toolChoice && typeof toolChoice === 'object') {
+        if (toolChoice.type === 'function' && toolChoice.function) {
+          requestBody.tool_choice = { type: 'tool', tool: { name: toolChoice.function.name } };
+        } else {
+          requestBody.tool_choice = { type: 'auto' };
+        }
+      } else {
+        requestBody.tool_choice = { type: 'auto' };
+      }
+    }
+
+    // Anthropic thinking mode
+    if (options.thinkingTokens && options.thinkingTokens > 0) {
+      requestBody.thinking = { type: 'enabled', budget_tokens: options.thinkingTokens };
+    }
+
+    // Debug output（受showAIDebugInfo配置控制）
+    const showDebug = this.authConfig.showAIDebugInfo ?? false;
+    
+    if (showDebug) {
+      console.log('\n╔══════════════════════════════════════════════════════════╗');
+      console.log('║              AI REQUEST DEBUG (ANTHROPIC)                ║');
+      console.log('╚══════════════════════════════════════════════════════════╝');
+      console.log(`📦 Model: ${requestBody.model}`);
+      console.log(`🌐 Base URL: ${this.authConfig.baseUrl}`);
+      console.log(`💬 Total Messages: ${anthropicMessages.length} items`);
+      if (requestBody.temperature) console.log(`🌡️  Temperature: ${requestBody.temperature}`);
+      if (requestBody.max_tokens) console.log(`📏 Max Tokens: ${requestBody.max_tokens}`);
+      if (requestBody.tools) console.log(`🔧 Tools: ${requestBody.tools.length} items`);
+      if (requestBody.thinking) console.log(`🧠 Thinking Budget: ${requestBody.thinking.budget_tokens}`);
+      console.log('─'.repeat(60));
+      
+      // Display system messages
+      if (system) {
+        console.log('\n┌─────────────────────────────────────────────────────────────┐');
+        console.log('│ 🟫 SYSTEM                                                     │');
+        console.log('├─────────────────────────────────────────────────────────────┤');
+        console.log(renderMarkdown(system).split('\n').map(l => '│ ' + l).join('\n'));
+        console.log('└─────────────────────────────────────────────────────────────┘');
+      }
+      
+      // Display user and assistant messages
+      displayMessages(anthropicMessages);
+      
+      console.log('\n📤 Sending to Anthropic API (v1/messages)...\n');
+    }
+
+    try {
+      // Use Anthropic native endpoint /v1/messages
+      const response = await this.client.post('/v1/messages', requestBody);
+      
+      if (showDebug) {
+        console.log('\n╔══════════════════════════════════════════════════════════╗');
+        console.log('║             AI RESPONSE DEBUG (ANTHROPIC)                ║');
+        console.log('╚══════════════════════════════════════════════════════════╝');
+        console.log(`🆔 ID: ${response.data.id}`);
+        console.log(`🤖 Model: ${response.data.model}`);
+        const usage = response.data.usage;
+        if (usage) {
+          console.log(`📊 Tokens: ${usage.input_tokens} (input) + ${usage.output_tokens} (output) = ${usage.input_tokens + usage.output_tokens} (total)`);
+        }
+        console.log(`🏁 Stop Reason: ${response.data.stop_reason}`);
+        
+        console.log('\n┌─────────────────────────────────────────────────────────────┐');
+        console.log('│ 🤖 ASSISTANT                                                 │');
+        console.log('├─────────────────────────────────────────────────────────────┤');
+        
+        const content = response.data.content || [];
+        const reasoning = content.filter((c: any) => c.type === 'thinking').map((c: any) => c.thinking).join('');
+        const textContent = content.filter((c: any) => c.type === 'text').map((c: any) => c.text).join('');
+        
+        // Display thinking
+        if (reasoning) {
+          console.log('│ 🧠 REASONING:');
+          console.log('│ ───────────────────────────────────────────────────────────');
+          const reasoningLines = renderMarkdown(reasoning).split('\n');
+          for (const line of reasoningLines.slice(0, 15)) {
+            console.log('│ ' + line.slice(0, 62));
+          }
+          if (reasoning.length > 800) console.log('│ ... (truncated)');
+          console.log('│ ───────────────────────────────────────────────────────────');
+        }
+        
+        // Display content
+        console.log('│ 💬 CONTENT:');
+        console.log('│ ───────────────────────────────────────────────────────────');
+        const lines = renderMarkdown(textContent).split('\n');
+        for (const line of lines.slice(0, 40)) {
+          console.log('│ ' + line.slice(0, 62));
+        }
+        if (lines.length > 40) {
+          console.log(`│ ... (${lines.length - 40} more lines)`);
+        }
+        console.log('└─────────────────────────────────────────────────────────────┘');
+        
+        console.log('\n╔══════════════════════════════════════════════════════════╗');
+        console.log('║                    RESPONSE ENDED                        ║');
+        console.log('╚══════════════════════════════════════════════════════════╝\n');
+      }
+      
+      return this.convertFromAnthropicNativeResponse(response.data);
+    } catch (error: any) {
+      const isRetryable = this.isRetryableError(error);
+      if (!isRetryable) {
+        if (error.response) {
+          throw new Error(
+            `API Error: ${error.response.status} - ${JSON.stringify(error.response.data)}`
+          );
+        } else if (error.request) {
+          throw new Error('Network error: No response received from server');
+        } else {
+          throw new Error(`Request error: ${error.message}`);
+        }
+      }
+
+      const retryResult = await withRetry(async () => {
+        const response = await this.client.post('/v1/messages', requestBody);
+        if (showDebug) {
+          console.log('\n╔══════════════════════════════════════════════════════════╗');
+          console.log('║             AI RESPONSE DEBUG (ANTHROPIC RETRY)          ║');
+          console.log('╚══════════════════════════════════════════════════════════╝\n');
+        }
+        return this.convertFromAnthropicNativeResponse(response.data);
+      }, { maxRetries: 3, baseDelay: 1000, maxDelay: 10000, jitter: true });
+
+      if (!retryResult.success) {
+        throw retryResult.error || new Error('Retry failed');
+      }
+
+      if (!retryResult.data) {
+        throw new Error('Retry returned empty response');
+      }
+
+      return retryResult.data;
+    }
+  }
+
+  // MiniMax API（Automatically select based on baseUrl Anthropic 或 OpenAI 格式）
+  private async minimaxChatCompletion(
+    messages: Message[],
+    options: ChatCompletionOptions = {}
+  ): Promise<ChatCompletionResponse> {
+    const { system, messages: anthropicMessages } = this.convertToAnthropicFormat(messages);
+    const { endpoint, format } = getMiniMaxEndpoint(this.authConfig.baseUrl || '');
+
+    const requestBody: any = {
+      model: options.model || this.authConfig.modelName || 'MiniMax-M2',
+      messages: format === 'anthropic' ? anthropicMessages : messages,
+      temperature: options.temperature ?? 1.0,
+      stream: false,
+      max_tokens: options.maxTokens || 4096
+    };
+
+    if (system && format === 'anthropic') {
+      requestBody.system = system;
+    }
+
+    if (format === 'anthropic') {
+      // Anthropic format tools
+      if (options.tools && options.tools.length > 0) {
+        requestBody.tools = options.tools.map(tool => ({
+          name: tool.function.name,
+          description: tool.function.description,
+          input_schema: tool.function.parameters || { type: 'object', properties: {} }
+        }));
+        
+        const toolChoice = options.toolChoice;
+        if (toolChoice === 'none') {
+          requestBody.tool_choice = { type: 'auto' };
+        } else if (toolChoice && typeof toolChoice === 'object') {
+          if (toolChoice.type === 'function' && toolChoice.function) {
+            requestBody.tool_choice = { type: 'tool', tool: { name: toolChoice.function.name } };
+          } else {
+            requestBody.tool_choice = { type: 'auto' };
+          }
+        } else {
+          requestBody.tool_choice = { type: 'auto' };
+        }
+      }
+    } else {
+      // OpenAI format tools
+      if (options.tools && options.tools.length > 0) {
+        requestBody.tools = options.tools;
+        requestBody.tool_choice = options.toolChoice || 'auto';
+      }
+    }
+
+    // Debug output（受showAIDebugInfo配置控制）
+    const showDebug = this.authConfig.showAIDebugInfo ?? false;
+    
+    if (showDebug) {
+      console.log('\n╔══════════════════════════════════════════════════════════╗');
+      console.log('║               AI REQUEST DEBUG (MINIMAX)                 ║');
+      console.log('╚══════════════════════════════════════════════════════════╝');
+      console.log(`📦 Model: ${requestBody.model}`);
+      console.log(`🔗 Format: ${format.toUpperCase()} | Endpoint: ${endpoint}`);
+      console.log(`🌐 Base URL: ${this.authConfig.baseUrl}`);
+      console.log(`💬 Total Messages: ${requestBody.messages.length} items`);
+      if (requestBody.temperature) console.log(`🌡️  Temperature: ${requestBody.temperature}`);
+      if (requestBody.max_tokens) console.log(`📏 Max Tokens: ${requestBody.max_tokens}`);
+      if (requestBody.tools) console.log(`🔧 Tools: ${requestBody.tools.length} items`);
+      console.log('─'.repeat(60));
+      
+      // Display system messages
+      if (system && format === 'anthropic') {
+        console.log('\n┌─────────────────────────────────────────────────────────────┐');
+        console.log('│ 🟫 SYSTEM                                                     │');
+        console.log('├─────────────────────────────────────────────────────────────┤');
+        console.log(renderMarkdown(system).split('\n').map(l => '│ ' + l).join('\n'));
+        console.log('└─────────────────────────────────────────────────────────────┘');
+      }
+      
+      // Display other messages
+      displayMessages(requestBody.messages);
+      
+      console.log('\n📤 Sending to MiniMax API...\n');
+    }
+
+    try {
+      // MiniMax uses correct endpoint
+      const response = await this.client.post(endpoint, requestBody);
+      
+      if (showDebug) {
+        console.log('\n╔══════════════════════════════════════════════════════════╗');
+        console.log('║              AI RESPONSE DEBUG (MINIMAX)                 ║');
+        console.log('╚══════════════════════════════════════════════════════════╝');
+        console.log(`🆔 ID: ${response.data.id}`);
+        console.log(`🤖 Model: ${response.data.model}`);
+        const usage = response.data.usage;
+        if (usage) {
+          console.log(`📊 Tokens: ${usage.prompt_tokens} (prompt) + ${usage.completion_tokens} (completion) = ${usage.total_tokens} (total)`);
+        }
+        console.log(`🏁 Stop Reason: ${response.data.stop_reason}`);
+        
+        console.log('\n┌─────────────────────────────────────────────────────────────┐');
+        console.log('│ 🤖 ASSISTANT                                                 │');
+        console.log('├─────────────────────────────────────────────────────────────┤');
+        
+        const message = response.data.choices?.[0]?.message;
+        const content = typeof message?.content === 'string' ? message.content : JSON.stringify(message?.content);
+        
+        console.log('│ 💬 CONTENT:');
+        console.log('│ ───────────────────────────────────────────────────────────');
+        const lines = renderMarkdown(content).split('\n');
+        for (const line of lines.slice(0, 40)) {
+          console.log('│ ' + line.slice(0, 62));
+        }
+        if (lines.length > 40) {
+          console.log(`│ ... (${lines.length - 40} more lines)`);
+        }
+        console.log('└─────────────────────────────────────────────────────────────┘');
+        
+        console.log('\n╔══════════════════════════════════════════════════════════╗');
+        console.log('║                    RESPONSE ENDED                        ║');
+        console.log('╚══════════════════════════════════════════════════════════╝\n');
+      }
+      
+      if (format === 'anthropic') {
+        return this.convertFromAnthropicNativeResponse(response.data);
+      } else {
+        return this.convertFromMiniMaxResponse(response.data);
+      }
+    } catch (error: any) {
+      const isRetryable = this.isRetryableError(error);
+      if (!isRetryable) {
+        if (error.response) {
+          throw new Error(
+            `API Error: ${error.response.status} - ${JSON.stringify(error.response.data)}`
+          );
+        } else if (error.request) {
+          throw new Error('Network error: No response received from server');
+        } else {
+          throw new Error(`Request error: ${error.message}`);
+        }
+      }
+
+      const retryResult = await withRetry(async () => {
+        const response = await this.client.post(endpoint, requestBody);
+        if (showDebug) {
+          console.log('\n╔══════════════════════════════════════════════════════════╗');
+          console.log('║              AI RESPONSE DEBUG (MINIMAX RETRY)           ║');
+          console.log('╚══════════════════════════════════════════════════════════╝\n');
+        }
+        if (format === 'anthropic') {
+          return this.convertFromAnthropicNativeResponse(response.data);
+        } else {
+          return this.convertFromMiniMaxResponse(response.data);
+        }
+      }, { maxRetries: 3, baseDelay: 1000, maxDelay: 10000, jitter: true });
+
+      if (!retryResult.success) {
+        throw retryResult.error || new Error('Retry failed');
+      }
+
+      if (!retryResult.data) {
+        throw new Error('Retry returned empty response');
+      }
+
+      return retryResult.data;
+    }
+  }
+
+  // Convert Anthropic native response to unified format
+  private convertFromAnthropicNativeResponse(anthropicResponse: any): ChatCompletionResponse {
+    const content = anthropicResponse.content || [];
+    let textContent = '';
+    let reasoningContent = '';
+    const toolCalls: any[] = [];
+
+    for (const block of content) {
+      if (block.type === 'text') {
+        textContent += block.text || '';
+      } else if (block.type === 'thinking') {
+        reasoningContent += block.thinking || '';
+      } else if (block.type === 'tool_use') {
+        toolCalls.push({
+          id: block.id,
+          type: 'function',
+          function: {
+            name: block.name,
+            arguments: JSON.stringify(block.input || {})
+          }
+        });
+      }
+    }
+
+    return {
+      id: anthropicResponse.id || `anthropic-${Date.now()}`,
+      object: 'chat.completion',
+      created: Math.floor(Date.now() / 1000),
+      model: anthropicResponse.model || this.authConfig.modelName || 'claude-sonnet-4-20250514',
+      choices: [{
+        index: 0,
+        message: {
+          role: 'assistant',
+          content: textContent,
+          reasoning_content: reasoningContent || undefined,
+          tool_calls: toolCalls.length > 0 ? toolCalls : undefined
+        },
+        finish_reason: anthropicResponse.stop_reason === 'end_turn' ? 'stop' : 
+                       anthropicResponse.stop_reason === 'max_tokens' ? 'length' : 'stop'
+      }],
+      usage: anthropicResponse.usage ? {
+        prompt_tokens: anthropicResponse.usage.input_tokens || 0,
+        completion_tokens: anthropicResponse.usage.output_tokens || 0,
+        total_tokens: (anthropicResponse.usage.input_tokens || 0) + (anthropicResponse.usage.output_tokens || 0)
+      } : undefined
+    };
+  }
+
+  // Convert MiniMax response to unified format
+  private convertFromMiniMaxResponse(minimaxResponse: any): ChatCompletionResponse {
+    const message = minimaxResponse.choices?.[0]?.message;
+    const content = message?.content;
+    let textContent = '';
+    let reasoningContent = '';
+    const toolCalls: any[] = [];
+
+    if (typeof content === 'string') {
+      textContent = content.trim();
+    } else if (Array.isArray(content)) {
+      for (const block of content) {
+        if (block.type === 'text') {
+          textContent += block.text || '';
+        } else if (block.type === 'thinking') {
+          reasoningContent += block.thinking || '';
+        } else if (block.type === 'tool_use') {
+          toolCalls.push({
+            id: block.id,
+            type: 'function',
+            function: {
+              name: block.name,
+              arguments: JSON.stringify(block.input || {})
+            }
+          });
+        }
+      }
+    }
+
+    return {
+      id: minimaxResponse.id || `minimax-${Date.now()}`,
+      object: 'chat.completion',
+      created: Math.floor(Date.now() / 1000),
+      model: minimaxResponse.model || this.authConfig.modelName || 'MiniMax-M2',
+      choices: [{
+        index: 0,
+        message: {
+          role: 'assistant',
+          content: textContent,
+          reasoning_content: reasoningContent || undefined,
+          tool_calls: toolCalls.length > 0 ? toolCalls : undefined
+        },
+        finish_reason: minimaxResponse.stop_reason === 'end_turn' ? 'stop' : 
+                       minimaxResponse.stop_reason === 'max_tokens' ? 'length' : 'stop'
+      }],
+      usage: minimaxResponse.usage
+    };
+  }
+
+  async *streamChatCompletion(
+    messages: Message[],
+    options: ChatCompletionOptions = {}
+  ): AsyncGenerator<string, void, unknown> {
+    const isMiniMax = detectMiniMaxAPI(this.authConfig.baseUrl || '');
+
+    if (isMiniMax) {
+      yield* this.minimaxStreamChatCompletion(messages, options);
+      return;
+    }
+
+    const isAnthropic = isAnthropicCompatible(this.authConfig.baseUrl || '');
+    if (isAnthropic) {
+      yield* this.anthropicNativeStreamChatCompletion(messages, options);
+      return;
+    }
+
+    // OpenAI streaming response
+    const model = options.model || this.authConfig.modelName || 'gpt-4';
+
+    const requestBody: any = {
+      model,
+      messages,
+      temperature: options.temperature ?? 0.7,
+      stream: true
+    };
+
+    if (options.maxTokens && options.maxTokens > 0) {
+      requestBody.max_tokens = options.maxTokens;
+    }
+
+    if (options.tools && options.tools.length > 0) {
+      requestBody.tools = options.tools;
+      requestBody.tool_choice = options.toolChoice || 'auto';
+    }
+
+    if (options.thinkingTokens && options.thinkingTokens > 0) {
+      requestBody.max_completion_tokens = options.thinkingTokens;
+    }
+
+    // Debug output（受showAIDebugInfo配置控制）
+    const showDebug = this.authConfig.showAIDebugInfo ?? false;
+    
+    if (showDebug) {
+      console.log('\n╔══════════════════════════════════════════════════════════╗');
+      console.log('║              AI REQUEST DEBUG (STREAM)                   ║');
+      console.log('╚══════════════════════════════════════════════════════════╝');
+      console.log(`📦 Model: ${model}`);
+      console.log(`🌐 Base URL: ${this.authConfig.baseUrl}`);
+      console.log(`💬 Total Messages: ${messages.length} items`);
+      if (options.temperature) console.log(`🌡️  Temperature: ${options.temperature}`);
+      if (options.maxTokens) console.log(`📏 Max Tokens: ${options.maxTokens}`);
+      if (options.tools?.length) console.log(`🔧 Tools: ${options.tools.length} items`);
+      console.log('─'.repeat(60));
+      
+      // Separate and display messages
+      const systemMsgs = messages.filter(m => m.role === 'system');
+      const otherMsgs = messages.filter(m => m.role !== 'system');
+      
+      if (systemMsgs.length > 0) {
+        const systemContent = typeof systemMsgs[0].content === 'string' 
+          ? systemMsgs[0].content 
+          : formatMessageContent(systemMsgs[0].content);
+        console.log('\n┌─────────────────────────────────────────────────────────────┐');
+        console.log('│ 🟫 SYSTEM                                                     │');
+        console.log('├─────────────────────────────────────────────────────────────┤');
+        console.log(renderMarkdown(systemContent).split('\n').map(l => '│ ' + l).join('\n'));
+        console.log('└─────────────────────────────────────────────────────────────┘');
+      }
+      
+      displayMessages(otherMsgs);
+      
+      console.log('\n📤 Starting stream...\n');
+    }
+
+    try {
+      const response = await this.client.post('/chat/completions', requestBody, {
+        responseType: 'stream'
+      });
+
+      console.log('📥 Receiving stream chunks...\n');
+
+      let buffer = '';
+      let chunkCount = 0;
+      let outputBuffer = '';
+
+      for await (const chunk of response.data) {
+        buffer += chunk.toString();
+        const lines = buffer.split('\n');
+
+        buffer = lines.pop() || '';
+
+        for (const line of lines) {
+          const trimmedLine = line.trim();
+          if (!trimmedLine) continue;
+
+          if (trimmedLine.startsWith('data: ')) {
+            const data = trimmedLine.slice(6);
+            if (data === '[DONE]') {
+              if (showDebug) {
+                console.log('\n╔══════════════════════════════════════════════════════════╗');
+                console.log('║              STREAM COMPLETED                            ║');
+                console.log('╚══════════════════════════════════════════════════════════╝');
+                console.log(`📦 Total chunks: ${chunkCount}`);
+                console.log(`📏 Total output: ${outputBuffer.length} chars`);
+                
+                console.log('\n┌─────────────────────────────────────────────────────────────┐');
+                console.log('│ 🤖 ASSISTANT OUTPUT                                        │');
+                console.log('├─────────────────────────────────────────────────────────────┤');
+                console.log('│ 💬 CONTENT:');
+                console.log('│ ───────────────────────────────────────────────────────────');
+                const lines = renderMarkdown(outputBuffer).split('\n');
+                for (const line of lines.slice(0, 30)) {
+                  console.log('│ ' + line.slice(0, 62));
+                }
+                if (lines.length > 30) {
+                  console.log(`│ ... (${lines.length - 30} more lines)`);
+                }
+                console.log('└─────────────────────────────────────────────────────────────┘');
+                console.log('');
+              }
+              return;
+            }
+
+            try {
+              const parsed = JSON.parse(data);
+              const delta = parsed.choices?.[0]?.delta;
+              if (delta?.content) {
+                chunkCount++;
+                outputBuffer += delta.content;
+                yield delta.content;
+              } else if (delta?.reasoning_content) {
+                chunkCount++;
+                outputBuffer += delta.reasoning_content;
+                yield delta.reasoning_content;
+              }
+            } catch (e) {
+              // Silently ignore parsing errors
+            }
+          }
+        }
+      }
+
+      if (buffer.trim()) {
+        const trimmedLine = buffer.trim();
+        if (trimmedLine.startsWith('data: ')) {
+          const data = trimmedLine.slice(6);
+          if (data !== '[DONE]') {
+            try {
+              const parsed = JSON.parse(data);
+              const delta = parsed.choices?.[0]?.delta;
+              if (delta?.content) {
+                yield delta.content;
+              } else if (delta?.reasoning_content) {
+                yield delta.reasoning_content;
+              }
+            } catch (e) {
+              // Ignore final parsing errors
+            }
+          }
+        }
+      }
+      
+      console.log('\n╔══════════════════════════════════════════════════════════╗');
+      console.log('║              STREAM COMPLETED                            ║');
+      console.log('╚══════════════════════════════════════════════════════════╝\n');
+    } catch (error: any) {
+      if (error.response) {
+        throw new Error(
+          `API Error: ${error.response.status} - ${JSON.stringify(error.response.data)}`
+        );
+      } else if (error.request) {
+        throw new Error('Network error: No response received from server');
+      } else {
+        throw new Error(`Request error: ${error.message}`);
+      }
+    }
+  }
+
+  // Anthropic native streaming response（/v1/messages 端点）
+  private async *anthropicNativeStreamChatCompletion(
+    messages: Message[],
+    options: ChatCompletionOptions = {}
+  ): AsyncGenerator<string, void, unknown> {
+    const { system, messages: anthropicMessages } = this.convertToAnthropicFormat(messages);
+
+    const requestBody: any = {
+      model: options.model || this.authConfig.modelName || 'claude-sonnet-4-20250514',
+      messages: anthropicMessages,
+      temperature: options.temperature ?? 1.0,
+      stream: true,
+      max_tokens: options.maxTokens || 4096
+    };
+
+    if (system) {
+      requestBody.system = system;
+    }
+
+    // Anthropic native tool format
+    if (options.tools && options.tools.length > 0) {
+      requestBody.tools = options.tools.map(tool => ({
+        name: tool.function.name,
+        description: tool.function.description,
+        input_schema: tool.function.parameters || { type: 'object', properties: {} }
+      }));
+      
+      const toolChoice = options.toolChoice;
+      if (toolChoice === 'none') {
+        requestBody.tool_choice = { type: 'auto' };
+      } else if (toolChoice && typeof toolChoice === 'object') {
+        if (toolChoice.type === 'function' && toolChoice.function) {
+          requestBody.tool_choice = { type: 'tool', tool: { name: toolChoice.function.name } };
+        } else {
+          requestBody.tool_choice = { type: 'auto' };
+        }
+      } else {
+        requestBody.tool_choice = { type: 'auto' };
+      }
+    }
+
+    if (options.thinkingTokens && options.thinkingTokens > 0) {
+      requestBody.thinking = { type: 'enabled', budget_tokens: options.thinkingTokens };
+    }
+
+    // Debug output（受showAIDebugInfo配置控制）
+    const showDebug = this.authConfig.showAIDebugInfo ?? false;
+    
+    if (showDebug) {
+      console.log('\n╔══════════════════════════════════════════════════════════╗');
+      console.log('║         AI REQUEST DEBUG (ANTHROPIC STREAM)             ║');
+      console.log('╚══════════════════════════════════════════════════════════╝');
+      console.log(`📦 Model: ${requestBody.model}`);
+      console.log(`🌐 Base URL: ${this.authConfig.baseUrl}`);
+      console.log(`💬 Total Messages: ${anthropicMessages.length} items`);
+      if (requestBody.temperature) console.log(`🌡️  Temperature: ${requestBody.temperature}`);
+      if (requestBody.max_tokens) console.log(`📏 Max Tokens: ${requestBody.max_tokens}`);
+      if (requestBody.thinking) console.log(`🧠 Thinking Budget: ${requestBody.thinking.budget_tokens}`);
+      console.log('─'.repeat(60));
+      
+      // Display system messages
+      if (system) {
+        console.log('\n┌─────────────────────────────────────────────────────────────┐');
+        console.log('│ 🟫 SYSTEM                                                     │');
+        console.log('├─────────────────────────────────────────────────────────────┤');
+        console.log(renderMarkdown(system).split('\n').map(l => '│ ' + l).join('\n'));
+        console.log('└─────────────────────────────────────────────────────────────┘');
+      }
+      
+      displayMessages(anthropicMessages);
+      
+      console.log('\n📤 Starting Anthropic stream...\n');
+    }
+
+    try {
+      // Anthropic native streaming endpoint /v1/messages
+      const response = await this.client.post('/v1/messages', requestBody, {
+        responseType: 'stream'
+      });
+
+      console.log('📥 Receiving Anthropic stream chunks...\n');
+
+      let buffer = '';
+      let outputBuffer = '';
+
+      for await (const chunk of response.data) {
+        buffer += chunk.toString();
+        const lines = buffer.split('\n');
+        buffer = lines.pop() || '';
+
+        for (const line of lines) {
+          const trimmedLine = line.trim();
+          if (!trimmedLine) continue;
+
+          // Anthropic streaming format: data: {"type":"content_block_delta",...}
+          if (trimmedLine.startsWith('data: ')) {
+            const data = trimmedLine.slice(6);
+
+            try {
+              const parsed = JSON.parse(data);
+
+              // Anthropic event types
+              if (parsed.type === 'content_block_delta') {
+                if (parsed.delta?.type === 'text_delta' && parsed.delta.text) {
+                  outputBuffer += parsed.delta.text;
+                  yield parsed.delta.text;
+                } else if (parsed.delta?.type === 'thinking_delta' && parsed.delta.thinking) {
+                  outputBuffer += parsed.delta.thinking;
+                  yield parsed.delta.thinking;
+                }
+              } else if (parsed.type === 'message_delta') {
+                if (parsed.delta?.stop_reason) {
+                  // Message end
+                  if (showDebug) {
+                    console.log('\n╔══════════════════════════════════════════════════════════╗');
+                    console.log('║              STREAM COMPLETED                            ║');
+                    console.log('╚══════════════════════════════════════════════════════════╝');
+                    console.log(`📏 Total output: ${outputBuffer.length} chars`);
+                    
+                    console.log('\n┌─────────────────────────────────────────────────────────────┐');
+                    console.log('│ 🤖 ASSISTANT OUTPUT                                        │');
+                    console.log('├─────────────────────────────────────────────────────────────┤');
+                    console.log('│ 💬 CONTENT:');
+                    console.log('│ ───────────────────────────────────────────────────────────');
+                    const lines = renderMarkdown(outputBuffer).split('\n');
+                    for (const line of lines.slice(0, 30)) {
+                      console.log('│ ' + line.slice(0, 62));
+                    }
+                    if (lines.length > 30) {
+                      console.log(`│ ... (${lines.length - 30} more lines)`);
+                    }
+                    console.log('└─────────────────────────────────────────────────────────────┘');
+                    console.log('');
+                  }
+                  return;
+                }
+              }
+            } catch (e) {
+              // Ignore parsing errors
+            }
+          }
+        }
+      }
+      
+      console.log('\n╔══════════════════════════════════════════════════════════╗');
+      console.log('║              STREAM COMPLETED                            ║');
+      console.log('╚══════════════════════════════════════════════════════════╝\n');
+    } catch (error: any) {
+      if (error.response) {
+        throw new Error(
+          `API Error: ${error.response.status} - ${JSON.stringify(error.response.data)}`
+        );
+      } else if (error.request) {
+        throw new Error('Network error: No response received from server');
+      } else {
+        throw new Error(`Request error: ${error.message}`);
+      }
+    }
+  }
+
+  // MiniMax streaming response（Automatically select based on baseUrl格式）
+  private async *minimaxStreamChatCompletion(
+    messages: Message[],
+    options: ChatCompletionOptions = {}
+  ): AsyncGenerator<string, void, unknown> {
+    const { system, messages: anthropicMessages } = this.convertToAnthropicFormat(messages);
+    const { endpoint, format } = getMiniMaxEndpoint(this.authConfig.baseUrl || '');
+
+    const requestBody: any = {
+      model: options.model || this.authConfig.modelName || 'MiniMax-M2',
+      messages: format === 'anthropic' ? anthropicMessages : messages,
+      temperature: options.temperature ?? 1.0,
+      stream: true,
+      max_tokens: options.maxTokens || 4096
+    };
+
+    if (system && format === 'anthropic') {
+      requestBody.system = system;
+    }
+
+    if (format === 'anthropic') {
+      // Anthropic 格式的工具
+      if (options.tools && options.tools.length > 0) {
+        requestBody.tools = options.tools.map(tool => ({
+          name: tool.function.name,
+          description: tool.function.description,
+          input_schema: tool.function.parameters || { type: 'object', properties: {} }
+        }));
+        
+        const toolChoice = options.toolChoice;
+        if (toolChoice === 'none') {
+          requestBody.tool_choice = { type: 'auto' };
+        } else if (toolChoice && typeof toolChoice === 'object') {
+          if (toolChoice.type === 'function' && toolChoice.function) {
+            requestBody.tool_choice = { type: 'tool', tool: { name: toolChoice.function.name } };
+          } else {
+            requestBody.tool_choice = { type: 'auto' };
+          }
+        } else {
+          requestBody.tool_choice = { type: 'auto' };
+        }
+      }
+    } else {
+      // OpenAI format tools
+      if (options.tools && options.tools.length > 0) {
+        requestBody.tools = options.tools;
+        requestBody.tool_choice = options.toolChoice || 'auto';
+      }
+    }
+
+    console.log('\n╔══════════════════════════════════════════════════════════╗');
+    console.log('║            AI REQUEST DEBUG (MINIMAX STREAM)             ║');
+    console.log('╚══════════════════════════════════════════════════════════╝');
+    console.log(`📦 Model: ${requestBody.model}`);
+    console.log(`🔗 Format: ${format.toUpperCase()} | Endpoint: ${endpoint}`);
+    console.log(`🌐 Base URL: ${this.authConfig.baseUrl}`);
+    console.log(`💬 Total Messages: ${requestBody.messages.length} items`);
+    if (requestBody.temperature) console.log(`🌡️  Temperature: ${requestBody.temperature}`);
+    if (requestBody.max_tokens) console.log(`📏 Max Tokens: ${requestBody.max_tokens}`);
+    console.log('─'.repeat(60));
+    
+    // Display system messages
+    if (system && format === 'anthropic') {
+      console.log('\n┌─────────────────────────────────────────────────────────────┐');
+      console.log('│ 🟫 SYSTEM                                                     │');
+      console.log('├─────────────────────────────────────────────────────────────┤');
+      console.log(renderMarkdown(system).split('\n').map(l => '│ ' + l).join('\n'));
+      console.log('└─────────────────────────────────────────────────────────────┘');
+    }
+    
+    displayMessages(requestBody.messages);
+    
+    console.log('\n📤 Starting MiniMax stream...\n');
+
+    try {
+      // MiniMax uses correct endpoint
+      const response = await this.client.post(endpoint, requestBody, {
+        responseType: 'stream'
+      });
+
+      console.log('📥 Receiving MiniMax stream chunks...\n');
+
+      let buffer = '';
+      let outputBuffer = '';
+
+      for await (const chunk of response.data) {
+        buffer += chunk.toString();
+        const lines = buffer.split('\n');
+        buffer = lines.pop() || '';
+
+        for (const line of lines) {
+          const trimmedLine = line.trim();
+          if (!trimmedLine) continue;
+
+          // Parse different streaming responses based on format
+          if (format === 'anthropic') {
+            // Anthropic SSE format: data: {"type":"content_block_delta",...}
+            if (trimmedLine.startsWith('data: ')) {
+              const data = trimmedLine.slice(6);
+
+              try {
+                const parsed = JSON.parse(data);
+
+                if (parsed.type === 'content_block_delta') {
+                  if (parsed.delta?.type === 'text_delta' && parsed.delta.text) {
+                    yield parsed.delta.text;
+                  } else if (parsed.delta?.type === 'thinking_delta' && parsed.delta.thinking) {
+                    yield parsed.delta.thinking;
+                  }
+                } else if (parsed.type === 'message_delta') {
+                  if (parsed.delta?.stop_reason) {
+                    console.log('\n╔══════════════════════════════════════════════════════════╗');
+                    console.log('║              STREAM COMPLETED                            ║');
+                    console.log('╚══════════════════════════════════════════════════════════╝');
+                    console.log(`📏 Total output: ${outputBuffer.length} chars`);
+                    
+                    console.log('\n┌─────────────────────────────────────────────────────────────┐');
+                    console.log('│ 🤖 ASSISTANT OUTPUT                                        │');
+                    console.log('├─────────────────────────────────────────────────────────────┤');
+                    console.log('│ 💬 CONTENT:');
+                    console.log('│ ───────────────────────────────────────────────────────────');
+                    const lines = renderMarkdown(outputBuffer).split('\n');
+                    for (const line of lines.slice(0, 30)) {
+                      console.log('│ ' + line.slice(0, 62));
+                    }
+                    if (lines.length > 30) {
+                      console.log(`│ ... (${lines.length - 30} more lines)`);
+                    }
+                    console.log('└─────────────────────────────────────────────────────────────┘');
+                    console.log('');
+                    return;
+                  }
+                }
+              } catch (e) {
+                // Ignore parsing errors
+              }
+            }
+          } else {
+            // OpenAI SSE format: data: {...}
+            if (trimmedLine.startsWith('data: ')) {
+              const data = trimmedLine.slice(6);
+              if (data === '[DONE]') continue;
+
+              try {
+                const parsed = JSON.parse(data);
+                const delta = parsed.choices?.[0]?.delta;
+                if (delta?.content) {
+                  outputBuffer += delta.content;
+                  yield delta.content;
+                } else if (delta?.reasoning_content) {
+                  outputBuffer += delta.reasoning_content;
+                  yield delta.reasoning_content;
+                }
+              } catch (e) {
+                // Ignore parsing errors
+              }
+            }
+          }
+        }
+      }
+      
+      console.log('\n╔══════════════════════════════════════════════════════════╗');
+      console.log('║              STREAM COMPLETED                            ║');
+      console.log('╚══════════════════════════════════════════════════════════╝\n');
+    } catch (error: any) {
+      if (error.response) {
+        throw new Error(
+          `API Error: ${error.response.status} - ${JSON.stringify(error.response.data)}`
+        );
+      } else if (error.request) {
+        throw new Error('Network error: No response received from server');
+      } else {
+        throw new Error(`Request error: ${error.message}`);
+      }
+    }
+  }
+
+  async listModels(): Promise<any[]> {
+    try {
+      const response = await this.client.get('/models');
+      return response.data.data || [];
+    } catch (error: any) {
+      console.error('Failed to list models:', error);
+      return [];
+    }
+  }
+
+  updateAuthConfig(authConfig: AuthConfig): void {
+    this.authConfig = authConfig;
+    this.client.defaults.baseURL = authConfig.baseUrl;
+    
+    const isMiniMax = detectMiniMaxAPI(authConfig.baseUrl || '');
+    const isAnthropic = !isMiniMax && isAnthropicCompatible(authConfig.baseUrl || '');
+    
+    if (isMiniMax || isAnthropic) {
+      // MiniMax/Anthropic: Use x-api-key auth header
+      this.client.defaults.headers['x-api-key'] = authConfig.apiKey || '';
+      this.client.defaults.headers['anthropic-version'] = '2023-06-01';
+      // Clear Bearer header
+      delete this.client.defaults.headers['Authorization'];
+    } else {
+      // OpenAI compatible: Use Bearer token
+      this.client.defaults.headers['Authorization'] = `Bearer ${authConfig.apiKey}`;
+      // Clear x-api-key header
+      delete this.client.defaults.headers['x-api-key'];
+      delete this.client.defaults.headers['anthropic-version'];
+    }
+  }
+
+  getAuthConfig(): AuthConfig {
+    return { ...this.authConfig };
+  }
+
+  // Check if messages contain tool calls
+  hasToolCalls(messages: Message[]): boolean {
+    return messages.some(msg => {
+      if (msg.tool_calls && msg.tool_calls.length > 0) return true;
+      if (Array.isArray(msg.content)) {
+        return msg.content.some(block => 
+          block.type === 'tool_use' || 
+          (block as any).type === 'tool_result'
+        );
+      }
+      return false;
+    });
+  }
+}
+
+export function detectThinkingKeywords(text: string): 'none' | 'normal' | 'hard' | 'mega' | 'ultra' {
+  const ultraKeywords = ['super think', 'extreme think', 'deep think', 'full think', 'ultra think', 'careful think',
+    'ultrathink', 'think really super hard', 'think intensely'];
+  const megaKeywords = ['strong think', 'powerful think', 'think hard', 'try hard to think', 'think well', 'think carefully',
+    'megathink', 'think really hard', 'think a lot'];
+  const hardKeywords = ['think again', 'think more', 'think clearly', 'think thoroughly', 'consider carefully',
+    'think about it', 'think more', 'think harder'];
+  const normalKeywords = ['think', 'think', 'consider', 'think'];
+
+  const lowerText = text.toLowerCase();
+
+  if (ultraKeywords.some(keyword => lowerText.includes(keyword.toLowerCase()))) {
+    return 'ultra';
+  } else if (megaKeywords.some(keyword => lowerText.includes(keyword.toLowerCase()))) {
+    return 'mega';
+  } else if (hardKeywords.some(keyword => lowerText.includes(keyword.toLowerCase()))) {
+    return 'hard';
+  } else if (normalKeywords.some(keyword => lowerText.includes(keyword.toLowerCase()))) {
+    return 'normal';
+  }
+
+  return 'none';
+}
+
+export function getThinkingTokens(mode: 'none' | 'normal' | 'hard' | 'mega' | 'ultra'): number {
+  const tokensMap = {
+    none: 0,
+    normal: 2000,
+    hard: 4000,
+    mega: 10000,
+    ultra: 32000
+  };
+  return tokensMap[mode];
 }