clean code

AGENTS.md

@@ -1,6 +1,6 @@
 # AGENTS.md
 
-This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
+This file provides guidance to Kode automation agents (including those compatible with Claude Code's `.claude` ecosystem) when working with code in this repository.
 
 ## Development Commands
 
@@ -46,7 +46,7 @@ SKIP_BUNDLED_CHECK=true npm publish
 ## High-Level Architecture
 
 ### Core System Design
-Kode implements a **three-layer parallel architecture** inspired by Claude Code:
+Kode implements a **three-layer parallel architecture** refined for fast iteration across terminal workflows while remaining compatible with the Claude Code agent ecosystem:
 
 1. **User Interaction Layer** (`src/screens/REPL.tsx`)
    - Interactive terminal interface using Ink (React for CLI)
@@ -74,9 +74,9 @@ Kode implements a **three-layer parallel architecture** inspired by Claude Code:
 ### Agent System (`src/utils/agentLoader.ts`)
 **Dynamic Agent Configuration Loading** with 5-tier priority system:
 1. Built-in (code-embedded)
-2. `~/.claude/agents/` (Claude user)
+2. `~/.claude/agents/` (Claude Code user directory compatibility)
 3. `~/.kode/agents/` (Kode user)
-4. `./.claude/agents/` (Claude project)
+4. `./.claude/agents/` (Claude Code project directory compatibility)
 5. `./.kode/agents/` (Kode project)
 
 Agents are defined as markdown files with YAML frontmatter:
@@ -99,7 +99,7 @@ Each tool follows a consistent pattern in `src/tools/[ToolName]/`:
 - Permission-aware execution
 
 ### Service Layer
-- **Claude Service** (`src/services/claude.ts`): Primary AI model interface
+- **Anthropic Service** (`src/services/claude.ts`): Claude API integration
 - **OpenAI Service** (`src/services/openai.ts`): OpenAI-compatible models
 - **Model Adapter Factory** (`src/services/modelAdapterFactory.ts`): Unified model interface
 - **MCP Client** (`src/services/mcpClient.ts`): Model Context Protocol for tool extensions
@@ -204,4 +204,4 @@ const description = typeof tool.description === 'function'
 1. Create `.md` file with proper YAML frontmatter
 2. Place in appropriate directory based on scope
 3. Test with `/agents` command
-4. Verify tool permissions work correctly
+4. Verify tool permissions work correctly

CLAUDE.md

@@ -1,6 +1,6 @@
 # CLAUDE.md
 
-This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
+This file provides guidance to Kode automation agents (including those compatible with Claude Code's `.claude` ecosystem) when working with code in this repository.
 
 ## Development Commands
 
@@ -46,7 +46,7 @@ SKIP_BUNDLED_CHECK=true npm publish
 ## High-Level Architecture
 
 ### Core System Design
-Kode implements a **three-layer parallel architecture** inspired by Claude Code:
+Kode implements a **three-layer parallel architecture** refined for fast iteration across terminal workflows while remaining compatible with the Claude Code agent ecosystem:
 
 1. **User Interaction Layer** (`src/screens/REPL.tsx`)
    - Interactive terminal interface using Ink (React for CLI)
@@ -74,9 +74,9 @@ Kode implements a **three-layer parallel architecture** inspired by Claude Code:
 ### Agent System (`src/utils/agentLoader.ts`)
 **Dynamic Agent Configuration Loading** with 5-tier priority system:
 1. Built-in (code-embedded)
-2. `~/.claude/agents/` (Claude user)
+2. `~/.claude/agents/` (Claude Code user directory compatibility)
 3. `~/.kode/agents/` (Kode user)
-4. `./.claude/agents/` (Claude project)
+4. `./.claude/agents/` (Claude Code project directory compatibility)
 5. `./.kode/agents/` (Kode project)
 
 Agents are defined as markdown files with YAML frontmatter:
@@ -99,7 +99,7 @@ Each tool follows a consistent pattern in `src/tools/[ToolName]/`:
 - Permission-aware execution
 
 ### Service Layer
-- **Claude Service** (`src/services/claude.ts`): Primary AI model interface
+- **Anthropic Service** (`src/services/claude.ts`): Claude API integration
 - **OpenAI Service** (`src/services/openai.ts`): OpenAI-compatible models
 - **Model Adapter Factory** (`src/services/modelAdapterFactory.ts`): Unified model interface
 - **MCP Client** (`src/services/mcpClient.ts`): Model Context Protocol for tool extensions
@@ -204,4 +204,4 @@ const description = typeof tool.description === 'function'
 1. Create `.md` file with proper YAML frontmatter
 2. Place in appropriate directory based on scope
 3. Test with `/agents` command
-4. Verify tool permissions work correctly
+4. Verify tool permissions work correctly

README.md

@@ -30,17 +30,17 @@ This change reflects our belief that the future of software development is colla
 ### Full Compatibility with Multiple Standards
 
 - ✅ **AGENTS.md** - Native support for the OpenAI-initiated standard format
-- ✅ **CLAUDE.md** - Full backward compatibility with Claude Code configurations  
+- ✅ **CLAUDE.md** - Full backward compatibility with Claude Code `.claude` configurations  
 - ✅ **Subagent System** - Advanced agent delegation and task orchestration
 - ✅ **Cross-platform** - Works with 20+ AI models and providers
 
-Use `# Your documentation request` to generate and maintain your AGENTS.md file automatically, while maintaining full compatibility with existing Claude Code workflows.
+Use `# Your documentation request` to generate and maintain your AGENTS.md file automatically, while preserving compatibility with existing `.claude` workflows.
 
 ## Overview
 
 Kode is a powerful AI assistant that lives in your terminal. It can understand your codebase, edit files, run commands, and handle entire workflows for you.
 
-> **⚠️ Security Notice**: Kode runs in YOLO mode by default (equivalent to Claude's `--dangerously-skip-permissions` flag), bypassing all permission checks for maximum productivity. YOLO mode is recommended only for trusted, secure environments when working on non-critical projects. If you're working with important files or using models of questionable capability, we strongly recommend using `kode --safe` to enable permission checks and manual approval for all operations.
+> **⚠️ Security Notice**: Kode runs in YOLO mode by default (equivalent to Claude Code's `--dangerously-skip-permissions` flag), bypassing all permission checks for maximum productivity. YOLO mode is recommended only for trusted, secure environments when working on non-critical projects. If you're working with important files or using models of questionable capability, we strongly recommend using `kode --safe` to enable permission checks and manual approval for all operations.
 > 
 > **📊 Model Performance**: For optimal performance, we recommend using newer, more capable models designed for autonomous task completion. Avoid older Q&A-focused models like GPT-4o or Gemini 2.5 Pro, which are optimized for answering questions rather than sustained independent task execution. Choose models specifically trained for agentic workflows and extended reasoning capabilities.
 

README.zh-CN.md

@@ -23,7 +23,7 @@
 
 Kode 是一个强大的 AI 助手，运行在你的终端中。它能理解你的代码库、编辑文件、运行命令，并为你处理整个开发工作流。
 
-> **⚠️ 安全提示**：Kode 默认以 YOLO 模式运行（等同于 Claude 的 `--dangerously-skip-permissions` 标志），跳过所有权限检查以获得最大生产力。YOLO 模式仅建议在安全可信的环境中处理非重要项目时使用。如果您正在处理重要文件或使用能力存疑的模型，我们强烈建议使用 `kode --safe` 启用权限检查和手动审批所有操作。
+> **⚠️ 安全提示**：Kode 默认以 YOLO 模式运行（等同于 Claude Code 的 `--dangerously-skip-permissions` 标志），跳过所有权限检查以获得最大生产力。YOLO 模式仅建议在安全可信的环境中处理非重要项目时使用。如果您正在处理重要文件或使用能力存疑的模型，我们强烈建议使用 `kode --safe` 启用权限检查和手动审批所有操作。
 > 
 > **📊 模型性能建议**：为获得最佳体验，建议使用专为自主任务完成设计的新一代强大模型。避免使用 GPT-4o、Gemini 2.5 Pro 等较老的问答型模型，它们主要针对回答问题进行优化，而非持续的独立任务执行。请选择专门训练用于智能体工作流和扩展推理能力的模型。
 

docs/agents-system.md

@@ -74,8 +74,8 @@ Agents can be defined at five levels with priority order (later overrides earlie
    - Provided with Kode
    - Cannot be modified
 
-2. **Claude User** (`~/.claude/agents/`)
-   - Claude Code compatible user-level agents
+2. **.claude User (Claude Code)** (`~/.claude/agents/`)
+   - Compatibility with Claude Code user directories
    - Personal agents available across all projects
 
 3. **Kode User** (`~/.kode/agents/`)
@@ -83,8 +83,8 @@ Agents can be defined at five levels with priority order (later overrides earlie
    - Overrides Claude user agents with same name
    - Create with `/agents` command or manually
 
-4. **Claude Project** (`./.claude/agents/`)
-   - Claude Code compatible project-specific agents
+4. **.claude Project (Claude Code)** (`./.claude/agents/`)
+   - Compatibility with Claude Code project directories
    - Overrides user-level agents
 
 5. **Kode Project** (`./.kode/agents/`)
@@ -227,4 +227,4 @@ Planned improvements:
 - Performance metrics per agent
 - Agent composition (agents using other agents)
 - Cloud-based agent sharing
-- Agent versioning and rollback
+- Agent versioning and rollback

src/commands/agents.tsx

@@ -20,7 +20,7 @@ import { randomUUID } from 'crypto'
 
 const execAsync = promisify(exec)
 
-// Core constants aligned with Claude Code architecture
+// Core constants aligned with the Claude Code agent architecture
 const AGENT_LOCATIONS = {
   USER: "user",
   PROJECT: "project", 
@@ -324,7 +324,7 @@ function validateAgentConfig(config: Partial<CreateState>, existingAgents: Agent
   }
 }
 
-// File system operations with Claude Code alignment
+// File system operations retained for Claude Code parity
 function getAgentDirectory(location: AgentLocation): string {
   if (location === AGENT_LOCATIONS.BUILT_IN || location === AGENT_LOCATIONS.ALL) {
     throw new Error(`Cannot get directory path for ${location} agents`)
@@ -545,7 +545,7 @@ async function updateAgent(
   writeFileSync(filePath, content, { encoding: 'utf-8', flag: 'w' })
 }
 
-// Enhanced UI Components with Claude Code alignment
+// Enhanced UI components retained for Claude Code parity
 
 interface HeaderProps {
   title: string
@@ -1574,7 +1574,7 @@ function AgentListView({
             <Box marginBottom={1}>
               <Text bold color={theme.primary}>💭 What are agents?</Text>
             </Box>
-            <Text>Specialized AI assistants that Claude can delegate to for specific tasks.</Text>
+            <Text>Specialized AI assistants that Kode can delegate to for specific tasks, compatible with Claude Code `.claude` agent packs.</Text>
             <Text>Each agent has its own context, prompt, and tools.</Text>
             
             <Box marginTop={1} marginBottom={1}>

src/commands/init.ts

@@ -1,5 +1,5 @@
 import type { Command } from '../commands'
-import { markProjectOnboardingComplete } from '../ProjectOnboarding'
+import { markProjectOnboardingComplete } from '../components/ProjectOnboarding'
 import { PROJECT_FILE } from '../constants/product'
 const command = {
   type: 'prompt',

src/commands/terminalSetup.ts

@@ -5,7 +5,7 @@ import chalk from 'chalk'
 import { getTheme } from '../utils/theme'
 import { env } from '../utils/env'
 import { getGlobalConfig, saveGlobalConfig } from '../utils/config'
-import { markProjectOnboardingComplete } from '../ProjectOnboarding'
+import { markProjectOnboardingComplete } from '../components/ProjectOnboarding'
 import { readFileSync, writeFileSync } from 'fs'
 import { join } from 'path'
 import { safeParseJSON } from '../utils/json'

src/components/Help.tsx

@@ -175,10 +175,10 @@ export function Help({
                 Custom commands loaded from:
               </Text>
               <Text color={theme.secondaryText}>
-                • {getCustomCommandDirectories().userClaude} (user: prefix)
+                • {getCustomCommandDirectories().userClaude} (Claude `.claude` user scope)
               </Text>
               <Text color={theme.secondaryText}>
-                • {getCustomCommandDirectories().projectClaude} (project: prefix)
+                • {getCustomCommandDirectories().projectClaude} (Claude `.claude` project scope)
               </Text>
               <Text color={theme.secondaryText}>
                 Use /refresh-commands to reload after changes
@@ -190,10 +190,10 @@ export function Help({
                 Create custom commands by adding .md files to:
               </Text>
               <Text color={theme.secondaryText}>
-                • {getCustomCommandDirectories().userClaude} (user: prefix)
+                • {getCustomCommandDirectories().userClaude} (Claude `.claude` user scope)
               </Text>
               <Text color={theme.secondaryText}>
-                • {getCustomCommandDirectories().projectClaude} (project: prefix)
+                • {getCustomCommandDirectories().projectClaude} (Claude `.claude` project scope)
               </Text>
               <Text color={theme.secondaryText}>
                 Use /refresh-commands to reload after creation

src/components/InvalidConfigDialog.tsx

@@ -18,7 +18,7 @@ interface InvalidConfigDialogProps {
 }
 
 /**
- * Dialog shown when the Claude config file contains invalid JSON
+ * Dialog shown when the Kode config file contains invalid JSON
  */
 function InvalidConfigDialog({
   filePath,

src/components/ProjectOnboarding.tsx

@@ -6,17 +6,17 @@ import {
   getGlobalConfig,
   saveCurrentProjectConfig,
   saveGlobalConfig,
-} from './utils/config.js'
+} from '../utils/config.js'
 import { existsSync } from 'fs'
 import { join } from 'path'
 import { homedir } from 'os'
-import terminalSetup from './commands/terminalSetup'
-import { getTheme } from './utils/theme'
-import { RELEASE_NOTES } from './constants/releaseNotes'
+import terminalSetup from '../commands/terminalSetup'
+import { getTheme } from '../utils/theme'
+import { RELEASE_NOTES } from '../constants/releaseNotes'
 import { gt } from 'semver'
-import { isDirEmpty } from './utils/file'
-import { MACRO } from './constants/macros'
-import { PROJECT_FILE, PRODUCT_NAME } from './constants/product'
+import { isDirEmpty } from '../utils/file'
+import { MACRO } from '../constants/macros'
+import { PROJECT_FILE, PRODUCT_NAME } from '../constants/product'
 
 // Function to mark onboarding as complete
 export function markProjectOnboardingComplete(): void {
@@ -74,9 +74,9 @@ export default function ProjectOnboarding({
 
   // Load what we need for onboarding
   // NOTE: This whole component is statically rendered Once
-  const hasClaudeMd = existsSync(join(workspaceDir, PROJECT_FILE))
+  const workspaceHasProjectGuide = existsSync(join(workspaceDir, PROJECT_FILE))
   const isWorkspaceDirEmpty = isDirEmpty(workspaceDir)
-  const needsClaudeMd = !hasClaudeMd && !isWorkspaceDirEmpty
+  const shouldRecommendProjectGuide = !workspaceHasProjectGuide && !isWorkspaceDirEmpty
   const showTerminalTip =
     terminalSetup.isEnabled && !getGlobalConfig().shiftEnterKeyBindingInstalled
 
@@ -106,9 +106,9 @@ export default function ProjectOnboarding({
                   </React.Fragment>,
                 )
               }
-              if (needsClaudeMd) {
+              if (shouldRecommendProjectGuide) {
                 items.push(
-                  <React.Fragment key="claudemd">
+                  <React.Fragment key="projectGuide">
                     {/* @ts-expect-error - OrderedList.Item children prop issue */}
                     <OrderedList.Item>
                       <Text color={theme.secondaryText}>

src/components/PromptInput.tsx

@@ -308,7 +308,7 @@ function PromptInput({
         addToHistory(mode === 'koding' ? `#${input}` : input)
         onInputChange('')
 
-        // Create additional context to inform Claude this is for KODING.md
+        // Create additional context to inform the assistant this is for KODING.md
         const kodingContext =
           'The user is using Koding mode. Format your response as a comprehensive, well-structured document suitable for adding to AGENTS.md. Use proper markdown formatting with headings, lists, code blocks, etc. The response should be complete and ready to add to AGENTS.md documentation.'
 
@@ -354,8 +354,8 @@ function PromptInput({
         if (messages.length) {
           await onQuery(messages)
 
-          // After query completes, the last message should be Claude's response
-          // We'll set up a one-time listener to capture and save Claude's response
+        // After query completes, the last message should be the assistant's response
+        // We'll set up a one-time listener to capture and save that response
           // This will be handled by the REPL component or message handler
         }
 
@@ -524,7 +524,7 @@ function PromptInput({
       onShowMessageSelector()
     }
 
-    // Shift+Tab for mode cycling (matching original Claude Code implementation)
+    // Shift+Tab for mode cycling (retains legacy keyboard behavior)
     if (key.shift && key.tab) {
       cycleMode()
       return true // Explicitly handled

src/components/messages/AssistantTextMessage.tsx

@@ -74,7 +74,7 @@ export function AssistantTextMessage({
   }
 
   switch (text) {
-    // Local JSX commands don't need a response, but we still want Claude to see them
+    // Local JSX commands don't need a response, but we still want the assistant to see them
     // Tool results render their own interrupt messages
     case NO_RESPONSE_REQUESTED:
     case INTERRUPT_MESSAGE_FOR_TOOL_USE:

src/constants/macros.ts

@@ -5,7 +5,7 @@ const pkg = require('../../package.json')
 
 export const MACRO = {
   VERSION: pkg.version,
-  README_URL: 'https://docs.anthropic.com/s/claude-code',
+  README_URL: 'https://github.com/shareAI-lab/kode#readme',
   PACKAGE_URL: '@shareai-lab/kode',
   ISSUES_EXPLAINER: 'report the issue at https://github.com/shareAI-lab/kode/issues',
 }

src/context.ts

@@ -19,7 +19,8 @@ import { lastX } from './utils/generators'
 import { getGitEmail } from './utils/user'
 import { PROJECT_FILE } from './constants/product'
 /**
- * Find all AGENTS.md and CLAUDE.md files in the current working directory
+ * Locate AGENTS.md and CLAUDE.md files for backward compatibility with
+ * existing documentation workflows.
  */
 export async function getClaudeFiles(): Promise<string | null> {
   const abortController = new AbortController()

src/entrypoints/cli.tsx

@@ -377,7 +377,6 @@ ${commandList}`,
           ),
           getClients(),
         ])
-        // logStartup()
         const inputPrompt = [prompt, stdinContent].filter(Boolean).join('\n')
         if (print) {
           if (!inputPrompt) {
@@ -1291,7 +1290,6 @@ ${commandList}`,
         loadLogList(CACHE_PATHS.messages()),
         getClients(),
       ])
-      // logStartup()
 
       // If a specific conversation is requested, load and resume it directly
       if (identifier !== undefined) {
@@ -1393,7 +1391,7 @@ ${commandList}`,
       })()
     })
 
-  // claude context (TODO: deprecate)
+  // legacy context (TODO: deprecate)
   const context = program
     .command('context')
     .description(

src/screens/REPL.tsx

@@ -2,7 +2,7 @@ import { ToolUseBlockParam } from '@anthropic-ai/sdk/resources/index.mjs'
 import { Box, Newline, Static, Text } from 'ink'
 import ProjectOnboarding, {
   markProjectOnboardingComplete,
-} from '../ProjectOnboarding.js'
+} from '../components/ProjectOnboarding.js'
 import { CostThresholdDialog } from '../components/CostThresholdDialog'
 import * as React from 'react'
 import { useEffect, useMemo, useRef, useState, useCallback } from 'react'
@@ -337,7 +337,7 @@ export function REPL({
 
     setMessages(oldMessages => [...oldMessages, ...newMessages])
 
-    // Mark onboarding as complete when any user message is sent to Claude
+    // Mark onboarding as complete when any user message is sent to the assistant
     markProjectOnboardingComplete()
 
     // The last message is an assistant message if the user input was a bash command,

src/services/claude.ts

@@ -148,9 +148,9 @@ class KodeContextManager {
 
       // 在调试模式下记录加载结果
       if (process.env.NODE_ENV === 'development') {
-        console.log(
-          `[KodeContext] Loaded ${this.projectDocsCache.length} characters from project docs`,
-        )
+        debugLogger.info('KODE_CONTEXT_LOADED', {
+          characters: this.projectDocsCache.length,
+        })
       }
     } catch (error) {
       console.warn('[KodeContext] Failed to load project docs:', error)
@@ -1212,7 +1212,7 @@ export function formatSystemPromptWithContext(
   agentId?: string,
   skipContextReminders = false, // Parameter kept for API compatibility but not used anymore
 ): { systemPrompt: string[]; reminders: string } {
-  // 构建增强的系统提示 - 对齐官方 Claude Code 直接注入方式
+  // 构建增强的系统提示，保持与原先直接注入方式的兼容
   const enhancedPrompt = [...systemPrompt]
   let reminders = ''
 
@@ -1759,7 +1759,10 @@ function getAssistantMessageFromError(error: unknown): AssistantMessage {
   }
   if (error instanceof Error) {
     if (process.env.NODE_ENV === 'development') {
-      console.log(error)
+      debugLogger.error('ANTHROPIC_API_ERROR', {
+        message: error.message,
+        stack: error.stack,
+      })
     }
     return createAssistantAPIErrorMessage(
       `${API_ERROR_MESSAGE_PREFIX}: ${error.message}`,

src/services/customCommands.ts

@@ -136,7 +136,6 @@ export async function resolveFileReferences(content: string): Promise<string> {
 function validateAllowedTools(allowedTools: string[] | undefined): boolean {
   // Log allowed tools for debugging and future integration
   if (allowedTools && allowedTools.length > 0) {
-    console.log('Command allowed tools:', allowedTools)
     // TODO: Integrate with src/permissions.ts tool permission system
     // TODO: Connect to Tool.tsx needsPermissions() mechanism
   }
@@ -147,8 +146,8 @@ function validateAllowedTools(allowedTools: string[] | undefined): boolean {
  * Frontmatter configuration for custom commands
  *
  * This interface defines the YAML frontmatter structure that can be used
- * to configure custom commands. It follows the same pattern as Claude Desktop's
- * custom command system but with additional fields for enhanced functionality.
+ * to configure custom commands. It mirrors the Claude Desktop custom command
+ * system for compatibility while adding Kode-specific enhancements.
  */
 export interface CustomCommandFrontmatter {
   /** Display name for the command (overrides filename-based naming) */
@@ -436,10 +435,10 @@ function createCustomCommand(
     async getPromptForCommand(args: string): Promise<MessageParam[]> {
       let prompt = content.trim()
 
-      // Process argument substitution following Claude Code conventions
+  // Process argument substitution following legacy conventions
       // This supports both the official $ARGUMENTS format and legacy {arg} format
 
-      // Step 1: Handle $ARGUMENTS placeholder (official Claude Code format)
+      // Step 1: Handle $ARGUMENTS placeholder (legacy command format)
       if (prompt.includes('$ARGUMENTS')) {
         prompt = prompt.replace(/\$ARGUMENTS/g, args || '')
       }
@@ -650,9 +649,6 @@ export const loadCustomCommands = memoize(
  */
 export const reloadCustomCommands = (): void => {
   loadCustomCommands.cache.clear()
-  console.log(
-    'Custom commands cache cleared. Commands will be reloaded on next use.',
-  )
 }
 
 /**

src/services/mentionProcessor.ts

@@ -9,6 +9,7 @@ import { getAvailableAgentTypes } from '../utils/agentLoader'
 import { existsSync } from 'fs'
 import { resolve } from 'path'
 import { getCwd } from '../utils/state'
+import { debug as debugLogger } from '../utils/debugLogger'
 
 export interface MentionContext {
   type: 'agent' | 'file'
@@ -159,10 +160,10 @@ class MentionProcessorService {
       
       // Log cache refresh for debugging mention resolution issues
       if (agents.length !== previousCacheSize) {
-        console.log('[MentionProcessor] Agent cache refreshed:', {
+        debugLogger.info('MENTION_PROCESSOR_CACHE_REFRESHED', {
           agentCount: agents.length,
           previousCacheSize,
-          cacheAge: now - this.lastAgentCheck
+          cacheAge: now - this.lastAgentCheck,
         })
       }
     } catch (error) {
@@ -232,17 +233,17 @@ class MentionProcessorService {
       }
       
       // Debug log for mention event emission tracking
-      console.log('[MentionProcessor] Emitted mention event:', {
+      debugLogger.info('MENTION_PROCESSOR_EVENT_EMITTED', {
         type: isAskModel ? 'ask-model' : 'agent',
         mention,
-        agentType: isAskModel ? undefined : agentType
+        agentType: isAskModel ? undefined : agentType,
       })
     } catch (error) {
-      console.error('[MentionProcessor] Failed to emit mention event:', {
+      debugLogger.error('MENTION_PROCESSOR_EVENT_FAILED', {
         mention,
         agentType,
         isAskModel,
-        error: error instanceof Error ? error.message : error
+        error: error instanceof Error ? error.message : error,
       })
     }
   }
@@ -270,4 +271,4 @@ export const processMentions = (input: string) =>
  * Clear mention processor caches
  */
 export const clearMentionCache = () =>
-  mentionProcessor.clearCache()
+  mentionProcessor.clearCache()

src/tools/TaskTool/TaskTool.tsx

@@ -55,12 +55,12 @@ const inputSchema = z.object({
 
 export const TaskTool = {
   async prompt({ safeMode }) {
-    // Match original Claude Code - prompt returns full agent descriptions
+    // Ensure agent prompts remain compatible with Claude Code `.claude` agent packs
     return await getPrompt(safeMode)
   },
   name: TOOL_NAME,
   async description() {
-    // Match original Claude Code exactly - simple description
+    // Ensure metadata stays compatible with Claude Code `.claude` agent packs
     return "Launch a new task"
   },
   inputSchema,

src/tools/TaskTool/prompt.ts

@@ -18,7 +18,7 @@ export async function getTaskTools(safeMode: boolean): Promise<Tool[]> {
 }
 
 export async function getPrompt(safeMode: boolean): Promise<string> {
-  // Extracted directly from original Claude Code obfuscated source
+  // Maintain compatibility with Claude Code `.claude` agent descriptions
   const agents = await getActiveAgents()
   
   // Format exactly as in original: (Tools: tool1, tool2)
@@ -29,7 +29,7 @@ export async function getPrompt(safeMode: boolean): Promise<string> {
     return `- ${agent.agentType}: ${agent.whenToUse} (Tools: ${toolsStr})`
   }).join('\n')
   
-  // 100% exact copy from original Claude Code source
+  // Keep the wording aligned so shared `.claude` agent packs behave identically
   return `Launch a new agent to handle complex, multi-step tasks autonomously. 
 
 Available agent types and the tools they have access to:

src/types/PermissionMode.ts

@@ -1,4 +1,4 @@
-// Permission mode types based on original Claude Code implementation
+// Permission mode types retained for compatibility with earlier agent implementations
 export type PermissionMode =
   | 'default'
   | 'acceptEdits'
@@ -35,7 +35,7 @@ export interface ModeConfig {
   }
 }
 
-// Mode configuration based on original Claude Code
+// Mode configuration preserved for Claude Code parity
 export const MODE_CONFIGS: Record<PermissionMode, ModeConfig> = {
   default: {
     name: 'default',
@@ -100,7 +100,7 @@ export const MODE_CONFIGS: Record<PermissionMode, ModeConfig> = {
   },
 }
 
-// Mode cycling function (based on original yg2 function)
+// Mode cycling function preserved from the Claude Code workflow
 export function getNextPermissionMode(
   currentMode: PermissionMode,
   isBypassAvailable: boolean = true,

src/utils/Cursor.ts

@@ -1,4 +1,5 @@
 import wrapAnsi from 'wrap-ansi'
+import { debug as debugLogger } from './debugLogger'
 
 type WrappedText = string[]
 type Position = {
@@ -320,11 +321,12 @@ export class MeasuredText {
         // For non-blank lines
         const startOffset = this.text.indexOf(text, searchOffset)
         if (startOffset === -1) {
-          console.log('Debug: Failed to find wrapped line in original text')
-          console.log('Debug: Current text:', text)
-          console.log('Debug: Full original text:', this.text)
-          console.log('Debug: Search offset:', searchOffset)
-          console.log('Debug: Wrapped text:', wrappedText)
+          debugLogger.error('CURSOR_WRAP_MISMATCH', {
+            currentText: text,
+            originalText: this.text,
+            searchOffset,
+            wrappedText,
+          })
           throw new Error('Failed to find wrapped line in original text')
         }
 

src/utils/agentLoader.ts

@@ -1,7 +1,8 @@
 /**
  * Agent configuration loader
- * Loads agent configurations from markdown files with YAML frontmatter
- * Following Claude Code's agent system architecture
+ * Loads agent configurations from markdown files with YAML frontmatter.
+ * Maintains compatibility with Claude Code `.claude` agent directories while
+ * prioritizing Kode-specific overrides.
  */
 
 import { existsSync, readFileSync, readdirSync, statSync, watch, FSWatcher } from 'fs'
@@ -231,7 +232,7 @@ let watchers: FSWatcher[] = []
 export async function startAgentWatcher(onChange?: () => void): Promise<void> {
   await stopAgentWatcher() // Clean up any existing watchers
   
-  // Watch both .claude and .kode directories
+  // Watch both Claude (.claude) and native (.kode) directories
   const userClaudeDir = join(homedir(), '.claude', 'agents')
   const userKodeDir = join(homedir(), '.kode', 'agents')
   const projectClaudeDir = join(getCwd(), '.claude', 'agents')

src/utils/ask.tsx

@@ -22,7 +22,7 @@ type Props = {
   verbose?: boolean
 }
 
-// Sends a single prompt to the Claude API and returns the response.
+// Sends a single prompt to the Anthropic Messages API and returns the response.
 // Assumes that claude is being used non-interactively -- will not
 // ask the user for permissions or further input.
 export async function ask({