codingbuddy-rules 4.5.0 → 5.1.0

package/.ai-rules/adapters/aider.md

@@ -0,0 +1,374 @@
+# Aider Integration Guide
+
+Guide for using codingbuddy with Aider (CLI AI pair programming tool).
+
+## Overview
+
+codingbuddy integrates with Aider through:
+
+1. **AGENTS.md** - Industry standard format compatible with all AI tools
+2. **CONVENTIONS.md** - Aider's native convention file format
+3. **`.aider.conf.yml`** - Aider configuration file
+
+## Two Usage Contexts
+
+### End Users (Your Project)
+
+End users provide codingbuddy rules to Aider via read-only convention files:
+
+```bash
+# Load codingbuddy rules as read-only context
+aider --read CONVENTIONS.md
+```
+
+Or configure in `.aider.conf.yml`:
+
+```yaml
+read:
+  - CONVENTIONS.md
+```
+
+### Monorepo Contributors
+
+Contributors to the codingbuddy repository can reference `.ai-rules/` files directly:
+
+```bash
+# Load multiple rule files as read-only context
+aider --read packages/rules/.ai-rules/rules/core.md \
+      --read packages/rules/.ai-rules/rules/augmented-coding.md \
+      --read packages/rules/.ai-rules/rules/project.md \
+      --read AGENTS.md
+```
+
+Or in `.aider.conf.yml`:
+
+```yaml
+read:
+  - packages/rules/.ai-rules/rules/core.md
+  - packages/rules/.ai-rules/rules/augmented-coding.md
+  - packages/rules/.ai-rules/rules/project.md
+  - AGENTS.md
+```
+
+## DRY Principle
+
+**Single Source of Truth**: `packages/rules/.ai-rules/`
+
+- All Agent definitions, rules, skills managed only in `.ai-rules/`
+- `CONVENTIONS.md` acts as a **pointer or summary** referencing `.ai-rules/`
+- No duplication; reference source files via `--read`
+
+## Aider Configuration System
+
+### `.aider.conf.yml`
+
+Aider's primary configuration file using YAML format. Loaded from these locations (in order, later files take priority):
+
+1. **Home directory** (`~/.aider.conf.yml`) - Global personal settings
+2. **Git root** (`.aider.conf.yml`) - Project-level settings
+3. **Current directory** (`.aider.conf.yml`) - Directory-specific settings
+
+### Configuration Example
+
+```yaml
+# .aider.conf.yml - codingbuddy integration
+
+# Load codingbuddy rules as read-only context
+read:
+  - CONVENTIONS.md
+  - AGENTS.md
+
+# Recommended settings for codingbuddy workflow
+auto-commits: true
+suggest-shell-commands: true
+```
+
+### List Format Options
+
+Lists can be specified as bullet lists or inline arrays:
+
+```yaml
+# Bullet list
+read:
+  - CONVENTIONS.md
+  - AGENTS.md
+
+# Inline array
+read: [CONVENTIONS.md, AGENTS.md]
+```
+
+### CONVENTIONS.md
+
+Aider's native mechanism for specifying coding guidelines. Create a `CONVENTIONS.md` in your project root:
+
+```markdown
+# Project Conventions
+
+## Workflow
+Follow PLAN/ACT/EVAL workflow modes.
+- Start in PLAN mode for analysis and planning
+- Move to ACT mode for implementation (TDD: Red -> Green -> Refactor)
+- Use EVAL mode for code quality evaluation
+
+## Code Quality
+- TypeScript strict mode (no `any`)
+- 90%+ test coverage goal
+- SOLID principles
+- No mocking - test real behavior
+
+## TDD Cycle
+1. Write a failing test (Red)
+2. Implement minimum code to pass (Green)
+3. Refactor for clarity (Refactor)
+
+## Project Structure
+See `packages/rules/.ai-rules/rules/project.md` for layered architecture details.
+
+## Specialist Agents
+See `packages/rules/.ai-rules/agents/README.md` for available specialists.
+```
+
+Load it with:
+
+```bash
+aider --read CONVENTIONS.md
+```
+
+> **Tip:** Use `--read` (read-only) instead of `--file` (editable) for convention files. This prevents accidental modifications and enables prompt caching.
+
+### `.aiderignore`
+
+Aider supports `.aiderignore` for excluding files from its context, using `.gitignore`-style patterns:
+
+```gitignore
+# .aiderignore
+node_modules/
+dist/
+*.min.js
+*.lock
+```
+
+Place in the git root directory. For custom ignore files, use `--aiderignore <filename>`.
+
+## Integration Methods
+
+### Method 1: CONVENTIONS.md (Recommended for End Users)
+
+Create a `CONVENTIONS.md` summarizing your project's coding standards, then load it as read-only context:
+
+```bash
+aider --read CONVENTIONS.md
+```
+
+### Method 2: Direct `.ai-rules/` References (Recommended for Contributors)
+
+Load specific rule files directly for comprehensive coverage:
+
+```yaml
+# .aider.conf.yml
+read:
+  - packages/rules/.ai-rules/rules/core.md
+  - packages/rules/.ai-rules/rules/augmented-coding.md
+  - packages/rules/.ai-rules/rules/project.md
+  - AGENTS.md
+```
+
+### Method 3: Combined Approach
+
+Use `CONVENTIONS.md` for quick reference and load detailed rules separately:
+
+```yaml
+# .aider.conf.yml
+read:
+  - CONVENTIONS.md
+  - packages/rules/.ai-rules/rules/core.md
+```
+
+## Directory Structure
+
+```
+project/
+├── .aider.conf.yml              # Aider configuration
+├── .aiderignore                 # File exclusion patterns
+├── CONVENTIONS.md               # Coding conventions (for Aider)
+├── AGENTS.md                    # Cross-platform agent definitions
+├── packages/rules/.ai-rules/
+│   ├── rules/
+│   │   ├── core.md
+│   │   ├── project.md
+│   │   └── augmented-coding.md
+│   ├── agents/
+│   │   └── *.json
+│   └── adapters/
+│       └── aider.md             # This guide
+└── ~/.aider.conf.yml            # Global user settings
+```
+
+## Usage
+
+### Basic Workflow
+
+```bash
+# Start Aider with codingbuddy conventions
+aider --read CONVENTIONS.md
+
+# In the Aider chat, reference workflow modes
+> Please follow PLAN mode to design this feature first
+
+# Add specific files for editing
+> /add src/auth/login.ts
+
+# Add read-only context
+> /read packages/rules/.ai-rules/agents/security-specialist.json
+```
+
+### In-Chat Commands
+
+| Command | Purpose |
+|---------|---------|
+| `/add <file>` | Add file for editing |
+| `/read <file>` | Add file as read-only context |
+| `/drop <file>` | Remove file from context |
+| `/ask <question>` | Ask without making changes |
+| `/architect <request>` | Use architect mode for planning |
+
+### Architect Mode
+
+Aider's `/architect` command maps naturally to codingbuddy's PLAN mode:
+
+```bash
+# PLAN mode equivalent
+> /architect Design a user authentication system with JWT
+
+# ACT mode equivalent (regular chat)
+> Implement the authentication system following TDD
+```
+
+### Using Specialist Context
+
+Load specialist agent definitions as read-only context for focused reviews:
+
+```bash
+# Security review
+> /read packages/rules/.ai-rules/agents/security-specialist.json
+> Review the auth implementation from a security perspective
+
+# Performance review
+> /read packages/rules/.ai-rules/agents/performance-specialist.json
+> Analyze performance bottlenecks in this module
+```
+
+## Community Conventions
+
+Aider maintains a community conventions repository at [github.com/Aider-AI/conventions](https://github.com/Aider-AI/conventions) with pre-built convention files for various tech stacks. These can complement codingbuddy rules.
+
+## MCP Support
+
+> **Note:** As of early 2026, Aider does not natively support MCP servers. Rules integration relies on the `--read` flag and `CONVENTIONS.md` approach described above. If MCP support is added in the future, the same codingbuddy MCP server configuration used by other tools can be applied.
+
+### Workaround: Using Aider with MCP-Provided Rules
+
+Export rules from the MCP server to files, then load them in Aider:
+
+```bash
+# Export rules content to a file (manual step)
+# Then load in Aider
+aider --read exported-rules.md
+```
+
+## Skills
+
+Without MCP support, Aider accesses codingbuddy skills through file references:
+
+### Using Skills
+
+```bash
+# Load a skill as read-only context
+> /read packages/rules/.ai-rules/skills/systematic-debugging/SKILL.md
+> Debug this authentication failure using the systematic debugging approach
+
+> /read packages/rules/.ai-rules/skills/test-driven-development/SKILL.md
+> Implement this feature following TDD workflow
+```
+
+### Available Skills
+
+| Skill | File |
+|-------|------|
+| Systematic Debugging | `skills/systematic-debugging/SKILL.md` |
+| Test-Driven Development | `skills/test-driven-development/SKILL.md` |
+| Brainstorming | `skills/brainstorming/SKILL.md` |
+| Writing Plans | `skills/writing-plans/SKILL.md` |
+| Executing Plans | `skills/executing-plans/SKILL.md` |
+| Frontend Design | `skills/frontend-design/SKILL.md` |
+
+## Known Limitations
+
+Aider is a CLI tool with a different architecture than IDE-based AI assistants:
+
+| Feature | Status | Workaround |
+|---------|--------|------------|
+| **MCP Server support** | ❌ Not available | Use `--read` flag with convention/rule files |
+| **Mode keywords (PLAN/ACT/EVAL/AUTO)** | ❌ Not native | Manually instruct Aider to follow mode workflows |
+| **Specialist agent dispatch** | ❌ Not available | Load specialist JSON as read-only context with `/read` |
+| **Background subagents** | ❌ Not available | Sequential analysis only |
+| **Session hooks** | ❌ Not available | Rely on `.aider.conf.yml` for always-on settings |
+| **Context document persistence** | ❌ Not available | Manually manage context between sessions |
+| **Skills auto-detection** | ❌ Not available | Manually load skills via `/read` |
+| **Autonomous loop (AUTO mode)** | ❌ Not available | Use Aider's interactive chat loop manually |
+
+### Aider Strengths
+
+Despite these limitations, Aider offers unique advantages:
+
+- **CLI-native**: Works in any terminal, SSH sessions, CI/CD pipelines
+- **Git-aware**: Automatic commits, diff-based editing
+- **Repo-map**: Automatic codebase understanding via tree-sitter
+- **Multi-model**: Supports OpenAI, Anthropic, local models, and more
+- **Prompt caching**: Read-only files are cached for efficient token usage
+- **Architect mode**: Built-in plan-then-implement workflow
+
+## AGENTS.md
+
+Industry standard format compatible with all AI tools (Aider, Cursor, Claude Code, Codex, etc.):
+
+```markdown
+# AGENTS.md
+
+This project uses codingbuddy MCP server to manage AI Agents.
+
+## Quick Start
+...
+```
+
+See `AGENTS.md` in project root for details. Aider can read this file:
+
+```bash
+aider --read AGENTS.md
+```
+
+## Verification Status
+
+> Documentation based on Aider public documentation and community resources. Runtime verification has been performed for core features.
+
+| Pattern | Status | Notes |
+|---------|--------|-------|
+| `CONVENTIONS.md` loading via `--read` | ✅ Verified | Documented in official Aider docs |
+| `.aider.conf.yml` configuration | ✅ Verified | YAML format with multi-location loading |
+| `.aiderignore` file support | ✅ Verified | `.gitignore`-style patterns |
+| Read-only file context (`/read`) | ✅ Verified | Files cached with prompt caching enabled |
+| Architect mode for planning | ✅ Verified | Maps to PLAN mode workflow |
+| MCP server integration | ❌ Not supported | No native MCP support as of early 2026 |
+| AGENTS.md compatibility | ✅ Verified | Standard markdown, loadable via `--read` |
+| Community conventions repo | ✅ Verified | Available at github.com/Aider-AI/conventions |
+
+## Reference
+
+- [Aider Documentation](https://aider.chat/docs/)
+- [Aider Conventions Guide](https://aider.chat/docs/usage/conventions.html)
+- [Aider Configuration](https://aider.chat/docs/config/aider_conf.html)
+- [Aider Options Reference](https://aider.chat/docs/config/options.html)
+- [Aider In-Chat Commands](https://aider.chat/docs/usage/commands.html)
+- [Community Conventions Repository](https://github.com/Aider-AI/conventions)
+- [AGENTS.md Official Spec](https://agents.md)
+- [codingbuddy Documentation](../../docs/)

package/.ai-rules/adapters/antigravity.md

@@ -30,9 +30,9 @@ End users access rules **only through MCP tools**. No local rule files needed.
 }
 ```
 
-> **Important:** Antigravity의 `roots/list` MCP capability 지원 여부는 미확인입니다.
-> `CODINGBUDDY_PROJECT_ROOT` 없이는 서버가 프로젝트의 `codingbuddy.config.json`을 찾지 못하여
-> `language` 등 설정이 기본값으로 동작합니다. 항상 이 환경변수를 프로젝트의 절대 경로로 설정하세요.
+> **Important:** Whether Antigravity supports the `roots/list` MCP capability has not been confirmed.
+> Without `CODINGBUDDY_PROJECT_ROOT`, the server cannot locate your project's `codingbuddy.config.json`,
+> causing settings such as `language` to use default values. Always set this environment variable to your project's absolute path.
 
 Optional: Create `.antigravity/rules/instructions.md` for basic integration:
 
@@ -567,7 +567,7 @@ Use the `AUTO` keyword (or localized versions) at the start of your message:
 | Language | Keyword |
 |----------|---------|
 | English | `AUTO` |
-| Korean | `자동` |
+| Korean | `AUTO` |
 | Japanese | `自動` |
 | Chinese | `自动` |
 | Spanish | `AUTOMÁTICO` |
@@ -579,7 +579,7 @@ AUTO implement user authentication feature
 ```
 
 ```
-자동 사용자 인증 기능 구현해줘
+AUTO implement user authentication feature
 ```
 
 When AUTO keyword is detected, Antigravity calls `parse_mode` MCP tool which returns AUTO mode instructions.
@@ -626,7 +626,7 @@ task_boundary(
 )
 ```
 
-> **Antigravity limitation:** AUTO mode에는 강제 루프 메커니즘이 없습니다. 자세한 내용은 [Known Limitations](#known-limitations)를 참조하세요.
+> **Antigravity limitation:** AUTO mode has no enforcement loop mechanism. See [Known Limitations](#known-limitations) for details.
 
 ## Context Document Management
 

package/.ai-rules/adapters/claude-code.md

@@ -191,8 +191,8 @@ Use `recommend_skills` MCP tool to get skill recommendations based on user promp
 recommend_skills({ prompt: "There is a bug in the login" })
 // => recommends: systematic-debugging
 
-recommend_skills({ prompt: "로그인에 버그가 있어" })
-// => recommends: systematic-debugging (Korean support)
+recommend_skills({ prompt: "There is a bug in the login" })
+// => recommends: systematic-debugging (multi-language support)
 
 recommend_skills({ prompt: "Build a dashboard component" })
 // => recommends: frontend-design
@@ -523,6 +523,70 @@ Each workflow mode activates different specialist agents:
 
 **Important:** Specialists from one mode do NOT carry over to the next mode. Each mode has its own recommended specialist set.
 
+### Auto-Dispatch Enforcement
+
+When `parse_mode` returns `dispatch="auto"` or `dispatchReady` with specialist agents, dispatching is **mandatory** — not optional.
+
+**Rule:** Every listed specialist MUST be dispatched. Skipping any specialist is a protocol violation.
+
+#### Teams-Based Specialist Dispatch (Preferred)
+
+Teams are preferred over the Agent tool for specialist dispatch because they enable structured coordination and message-based reporting:
+
+```
+1. TeamCreate({ team_name: "<task>-specialists" })
+2. Spawn specialists as teammates:
+   Agent({ team_name, name: "security-specialist", subagent_type: "general-purpose", prompt: ... })
+   Agent({ team_name, name: "code-quality-specialist", subagent_type: "general-purpose", prompt: ... })
+3. Create and assign tasks:
+   TaskCreate({ subject: "Security review of auth module" })
+   TaskUpdate({ taskId, owner: "security-specialist" })
+4. Specialists work autonomously, report via SendMessage:
+   SendMessage({ to: "team-lead", message: "## Security Findings\n- ...", summary: "Security review done" })
+5. Team lead collects all findings
+6. Shutdown: SendMessage({ to: "security-specialist", message: { type: "shutdown_request" } })
+```
+
+#### SendMessage-Based Reporting
+
+Specialists report findings through `SendMessage` to the team lead. This enables:
+- Structured collection of all specialist outputs
+- Consolidated summary for the user
+- Clear audit trail of what each specialist found
+
+**Report format:**
+```markdown
+## [Specialist Name] Findings
+
+### Critical
+- [finding]
+
+### High
+- [finding]
+
+### Medium
+- [finding]
+
+### Recommendations
+- [recommendation]
+```
+
+#### Fallback: Agent Tool
+
+If Teams-based dispatch fails, fall back to the Agent tool:
+- Use `run_in_background: true` for each specialist
+- Collect results via `TaskOutput`
+- Document the fallback reason in your response
+
+#### Red Flags
+
+| Thought | Reality |
+|---------|---------|
+| "I can handle this analysis myself" | Specialists have domain expertise. Dispatch them. |
+| "It's just a small change" | dispatch="auto" means the system determined specialists are needed. |
+| "I'll save time by skipping" | Skipping causes missed issues that cost more later. |
+| "I'll dispatch later" | Dispatch IMMEDIATELY when dispatch="auto" is returned. |
+
 ### Execution Strategy Selection (MANDATORY)
 
 When `parse_mode` returns `availableStrategies`:
@@ -621,7 +685,7 @@ Use the `AUTO` keyword (or localized versions) at the start of your message:
 | Language | Keyword |
 |----------|---------|
 | English | `AUTO` |
-| Korean | `자동` |
+| Korean | `AUTO` |
 | Japanese | `自動` |
 | Chinese | `自动` |
 | Spanish | `AUTOMATICO` |
@@ -633,7 +697,7 @@ AUTO implement user authentication with JWT tokens
 ```
 
 ```
-자동 사용자 인증 기능을 JWT로 구현해줘
+AUTO implement user authentication with JWT
 ```
 
 ### Expected Behavior

package/.ai-rules/adapters/codex.md

@@ -30,10 +30,10 @@ End users access rules **only through MCP tools**. No local rule files needed.
 }
 ```
 
-> **Important:** Codex(GitHub Copilot)의 `roots/list` MCP capability 지원 여부는 미확인입니다.
-> `CODINGBUDDY_PROJECT_ROOT` 없이는 서버가 프로젝트의 `codingbuddy.config.json`을 찾지 못하여
-> `language` 등 설정이 기본값으로 동작합니다. 항상 이 환경변수를 프로젝트의 절대 경로로 설정하세요.
-> Codex가 `${workspaceFolder}` 변수 확장을 지원하는 경우, 절대 경로 대신 사용할 수 있습니다.
+> **Important:** Whether Codex (GitHub Copilot) supports the `roots/list` MCP capability has not been confirmed.
+> Without `CODINGBUDDY_PROJECT_ROOT`, the server cannot locate your project's `codingbuddy.config.json`,
+> causing settings such as `language` to use default values. Always set this environment variable to your project's absolute path.
+> If Codex supports `${workspaceFolder}` variable expansion, you can use it instead of an absolute path.
 
 ### Monorepo Contributors
 
@@ -599,7 +599,7 @@ Use the `AUTO` keyword (or localized versions) at the start of your message:
 | Language | Keyword |
 |----------|---------|
 | English | `AUTO` |
-| Korean | `자동` |
+| Korean | `AUTO` |
 | Japanese | `自動` |
 | Chinese | `自动` |
 | Spanish | `AUTOMATICO` |

package/.ai-rules/adapters/cursor.md

@@ -403,7 +403,7 @@ Use the `AUTO` keyword (or localized versions) at the start of your message:
 | Language | Keyword |
 |----------|---------|
 | English | `AUTO` |
-| Korean | `자동` |
+| Korean | `AUTO` |
 | Japanese | `自動` |
 | Chinese | `自动` |
 | Spanish | `AUTOMATICO` |
@@ -415,7 +415,7 @@ AUTO implement user authentication feature
 ```
 
 ```
-자동 사용자 인증 기능 구현해줘
+AUTO implement user authentication feature
 ```
 
 When AUTO keyword is detected, Cursor calls `parse_mode` MCP tool which returns AUTO mode instructions.

package/.ai-rules/adapters/kiro.md

@@ -30,9 +30,9 @@ End users access rules **only through MCP tools**. No local rule files needed.
 }
 ```
 
-> **Important:** Kiro의 `roots/list` MCP capability 지원 여부는 미확인입니다.
-> `CODINGBUDDY_PROJECT_ROOT` 없이는 서버가 프로젝트의 `codingbuddy.config.json`을 찾지 못하여
-> `language` 등 설정이 기본값으로 동작합니다. 항상 이 환경변수를 프로젝트의 절대 경로로 설정하세요.
+> **Important:** Whether Kiro supports the `roots/list` MCP capability has not been confirmed.
+> Without `CODINGBUDDY_PROJECT_ROOT`, the server cannot locate your project's `codingbuddy.config.json`,
+> causing settings such as `language` to use default values. Always set this environment variable to your project's absolute path.
 
 Optional: Create `.kiro/rules/guidelines.md` for basic integration:
 
@@ -85,8 +85,8 @@ MCP server configuration for codingbuddy tools:
 }
 ```
 
-> **Note:** Kiro는 `${VARIABLE}` 문법으로 환경변수 확장을 지원합니다.
-> `${workspaceFolder}` 가 지원되는 경우 절대 경로 대신 사용할 수 있습니다.
+> **Note:** Kiro supports environment variable expansion using the `${VARIABLE}` syntax.
+> If `${workspaceFolder}` is supported, you can use it instead of an absolute path.
 
 **MCP configuration paths:**
 - **Project-level**: `.kiro/settings/mcp.json`
@@ -526,7 +526,7 @@ Use the `AUTO` keyword (or localized versions) at the start of your message:
 | Language | Keyword |
 |----------|---------|
 | English | `AUTO` |
-| Korean | `자동` |
+| Korean | `AUTO` |
 | Japanese | `自動` |
 | Chinese | `自动` |
 | Spanish | `AUTOMÁTICO` |
@@ -538,7 +538,7 @@ AUTO implement user authentication feature
 ```
 
 ```
-자동 사용자 인증 기능 구현해줘
+AUTO implement user authentication feature
 ```
 
 When AUTO keyword is detected, Kiro calls `parse_mode` MCP tool which returns AUTO mode instructions.
@@ -571,7 +571,7 @@ module.exports = {
 - Bug fixes needing comprehensive testing
 - Code quality improvements with measurable criteria
 
-> **Kiro limitation:** AUTO mode에는 강제 루프 메커니즘이 없습니다. 자세한 내용은 [Known Limitations](#known-limitations)를 참조하세요.
+> **Kiro limitation:** AUTO mode has no enforcement loop mechanism. See [Known Limitations](#known-limitations) for details.
 
 ## Context Document Management
 

package/.ai-rules/adapters/opencode.md

@@ -184,9 +184,9 @@ Add to your MCP configuration:
 }
 ```
 
-> **Important:** OpenCode/Crush의 `roots/list` MCP capability 지원 여부는 미확인입니다.
-> `CODINGBUDDY_PROJECT_ROOT` 없이는 서버가 프로젝트의 `codingbuddy.config.json`을 찾지 못하여
-> `language` 등 설정이 기본값으로 동작합니다. 항상 이 환경변수를 프로젝트의 절대 경로로 설정하세요.
+> **Important:** Whether OpenCode/Crush supports the `roots/list` MCP capability has not been confirmed.
+> Without `CODINGBUDDY_PROJECT_ROOT`, the server cannot locate your project's `codingbuddy.config.json`,
+> causing settings such as `language` to use default values. Always set this environment variable to your project's absolute path.
 
 #### Available MCP Tools
 
@@ -526,9 +526,9 @@ npx codingbuddy@latest mcp
 
 **4. Project Config Not Detected**
 ```bash
-# CODINGBUDDY_PROJECT_ROOT가 MCP env에 설정되어 있는지 확인
-# 이 환경변수 없이는 codingbuddy.config.json을 찾지 못합니다
-# .opencode.json 또는 crush.json의 mcp 섹션에 추가:
+# Verify that CODINGBUDDY_PROJECT_ROOT is set in MCP env
+# Without this environment variable, codingbuddy.config.json cannot be found
+# Add to the mcp section of .opencode.json or crush.json:
 "env": {
   "CODINGBUDDY_PROJECT_ROOT": "/absolute/path/to/your/project"
 }
@@ -752,7 +752,7 @@ Use the `AUTO` keyword (or localized versions) at the start of your message:
 | Language | Keyword |
 |----------|---------|
 | English | `AUTO` |
-| Korean | `자동` |
+| Korean | `AUTO` |
 | Japanese | `自動` |
 | Chinese | `自动` |
 | Spanish | `AUTOMÁTICO` |

package/.ai-rules/adapters/q.md

@@ -178,7 +178,7 @@ Use the `AUTO` keyword (or localized versions) at the start of your message:
 | Language | Keyword |
 |----------|---------|
 | English | `AUTO` |
-| Korean | `자동` |
+| Korean | `AUTO` |
 | Japanese | `自動` |
 | Chinese | `自动` |
 | Spanish | `AUTOMATICO` |
@@ -186,7 +186,7 @@ Use the `AUTO` keyword (or localized versions) at the start of your message:
 ### Example Usage
 
 ```
-AUTO 새로운 Lambda 함수 만들어줘
+AUTO create a new Lambda function
 ```
 
 ### Workflow