diff --git a/.env.example b/.env.example
index c426100..e29bec3 100644
--- a/.env.example
+++ b/.env.example
@@ -7,3 +7,15 @@
 # - false: Browser opens a visible window (useful for debugging)
 # Defaults to 'false' if not specified
 # PLAYWRIGHT_HEADLESS=false
+
+# GLM/Alternative API Configuration (Optional)
+# To use Zhipu AI's GLM models instead of Claude, uncomment and set these variables.
+# This only affects AutoCoder - your global Claude Code settings remain unchanged.
+# Get an API key at: https://z.ai/subscribe
+#
+# ANTHROPIC_BASE_URL=https://api.z.ai/api/anthropic
+# ANTHROPIC_AUTH_TOKEN=your-zhipu-api-key
+# API_TIMEOUT_MS=3000000
+# ANTHROPIC_DEFAULT_SONNET_MODEL=glm-4.7
+# ANTHROPIC_DEFAULT_OPUS_MODEL=glm-4.7
+# ANTHROPIC_DEFAULT_HAIKU_MODEL=glm-4.5-air
diff --git a/README.md b/README.md
index 9af8172..3ed7f15 100644
--- a/README.md
+++ b/README.md
@@ -290,18 +290,18 @@ When test progress increases, the agent sends:
 
 ### Using GLM Models (Alternative to Claude)
 
-To use Zhipu AI's GLM models instead of Claude, create a settings file at `~/.claude/settings.json`:
+To use Zhipu AI's GLM models instead of Claude, add these variables to your `.env` file in the AutoCoder directory:
 
-```json
-{
-  "env": {
-    "ANTHROPIC_BASE_URL": "https://api.z.ai/api/anthropic",
-    "ANTHROPIC_AUTH_TOKEN": "your-zhipu-api-key"
-  }
-}
+```bash
+ANTHROPIC_BASE_URL=https://api.z.ai/api/anthropic
+ANTHROPIC_AUTH_TOKEN=your-zhipu-api-key
+API_TIMEOUT_MS=3000000
+ANTHROPIC_DEFAULT_SONNET_MODEL=glm-4.7
+ANTHROPIC_DEFAULT_OPUS_MODEL=glm-4.7
+ANTHROPIC_DEFAULT_HAIKU_MODEL=glm-4.5-air
 ```
 
-This routes Claude Code requests through Zhipu's Claude-compatible API, allowing you to use GLM-4.7 and other models while keeping all Claude Code features (MCP servers, hooks, permissions).
+This routes AutoCoder's API requests through Zhipu's Claude-compatible API, allowing you to use GLM-4.7 and other models. **This only affects AutoCoder** - your global Claude Code settings remain unchanged.
 
 Get an API key at: https://z.ai/subscribe
 
diff --git a/client.py b/client.py
index f232a0a..fdf6d54 100644
--- a/client.py
+++ b/client.py
@@ -25,6 +25,18 @@ load_dotenv()
 # When False, browser window is visible (default - useful for monitoring agent progress)
 DEFAULT_PLAYWRIGHT_HEADLESS = False
 
+# Environment variables to pass through to Claude CLI for API configuration
+# These allow using alternative API endpoints (e.g., GLM via z.ai) without
+# affecting the user's global Claude Code settings
+API_ENV_VARS = [
+    "ANTHROPIC_BASE_URL",              # Custom API endpoint (e.g., https://api.z.ai/api/anthropic)
+    "ANTHROPIC_AUTH_TOKEN",            # API authentication token
+    "API_TIMEOUT_MS",                  # Request timeout in milliseconds
+    "ANTHROPIC_DEFAULT_SONNET_MODEL",  # Model override for Sonnet
+    "ANTHROPIC_DEFAULT_OPUS_MODEL",    # Model override for Opus
+    "ANTHROPIC_DEFAULT_HAIKU_MODEL",   # Model override for Haiku
+]
+
 
 def get_playwright_headless() -> bool:
     """
@@ -205,6 +217,21 @@ def create_client(project_dir: Path, model: str, yolo_mode: bool = False):
             "args": playwright_args,
         }
 
+    # Build environment overrides for API endpoint configuration
+    # These override system env vars for the Claude CLI subprocess,
+    # allowing AutoCoder to use alternative APIs (e.g., GLM) without
+    # affecting the user's global Claude Code settings
+    sdk_env = {}
+    for var in API_ENV_VARS:
+        value = os.getenv(var)
+        if value:
+            sdk_env[var] = value
+
+    if sdk_env:
+        print(f"   - API overrides: {', '.join(sdk_env.keys())}")
+        if "ANTHROPIC_BASE_URL" in sdk_env:
+            print(f"   - GLM Mode: Using {sdk_env['ANTHROPIC_BASE_URL']}")
+
     return ClaudeSDKClient(
         options=ClaudeAgentOptions(
             model=model,
@@ -222,5 +249,6 @@ def create_client(project_dir: Path, model: str, yolo_mode: bool = False):
             max_turns=1000,
             cwd=str(project_dir.resolve()),
             settings=str(settings_file.resolve()),  # Use absolute path
+            env=sdk_env,  # Pass API configuration overrides to CLI subprocess
         )
     )
diff --git a/server/main.py b/server/main.py
index 8be2a50..9340315 100644
--- a/server/main.py
+++ b/server/main.py
@@ -6,6 +6,7 @@ Main entry point for the Autonomous Coding UI server.
 Provides REST API, WebSocket, and static file serving.
 """
 
+import os
 import shutil
 from contextlib import asynccontextmanager
 from pathlib import Path
@@ -148,7 +149,11 @@ async def setup_status():
     # Note: CLI no longer stores credentials in ~/.claude/.credentials.json
     # The existence of ~/.claude indicates the CLI has been configured
     claude_dir = Path.home() / ".claude"
-    credentials = claude_dir.exists() and claude_dir.is_dir()
+    has_claude_config = claude_dir.exists() and claude_dir.is_dir()
+
+    # If GLM mode is configured via .env, we have alternative credentials
+    glm_configured = bool(os.getenv("ANTHROPIC_BASE_URL") and os.getenv("ANTHROPIC_AUTH_TOKEN"))
+    credentials = has_claude_config or glm_configured
 
     # Check for Node.js and npm
     node = shutil.which("node") is not None
diff --git a/server/routers/settings.py b/server/routers/settings.py
index 18362ee..78d6ff8 100644
--- a/server/routers/settings.py
+++ b/server/routers/settings.py
@@ -6,6 +6,7 @@ API endpoints for global settings management.
 Settings are stored in the registry database and shared across all projects.
 """
 
+import os
 import sys
 from pathlib import Path
 
@@ -33,6 +34,11 @@ def _parse_yolo_mode(value: str | None) -> bool:
     return (value or "false").lower() == "true"
 
 
+def _is_glm_mode() -> bool:
+    """Check if GLM API is configured via environment variables."""
+    return bool(os.getenv("ANTHROPIC_BASE_URL"))
+
+
 @router.get("/models", response_model=ModelsResponse)
 async def get_available_models():
     """Get list of available models.
@@ -54,6 +60,7 @@ async def get_settings():
     return SettingsResponse(
         yolo_mode=_parse_yolo_mode(all_settings.get("yolo_mode")),
         model=all_settings.get("model", DEFAULT_MODEL),
+        glm_mode=_is_glm_mode(),
     )
 
 
@@ -71,4 +78,5 @@ async def update_settings(update: SettingsUpdate):
     return SettingsResponse(
         yolo_mode=_parse_yolo_mode(all_settings.get("yolo_mode")),
         model=all_settings.get("model", DEFAULT_MODEL),
+        glm_mode=_is_glm_mode(),
     )
diff --git a/server/schemas.py b/server/schemas.py
index 72d6bf4..e9b9c31 100644
--- a/server/schemas.py
+++ b/server/schemas.py
@@ -289,6 +289,7 @@ class SettingsResponse(BaseModel):
     """Response schema for global settings."""
     yolo_mode: bool = False
     model: str = DEFAULT_MODEL
+    glm_mode: bool = False  # True if GLM API is configured via .env
 
 
 class ModelsResponse(BaseModel):
diff --git a/server/services/assistant_chat_session.py b/server/services/assistant_chat_session.py
index 9e067f1..cf18241 100755
--- a/server/services/assistant_chat_session.py
+++ b/server/services/assistant_chat_session.py
@@ -33,6 +33,16 @@ logger = logging.getLogger(__name__)
 # Root directory of the project
 ROOT_DIR = Path(__file__).parent.parent.parent
 
+# Environment variables to pass through to Claude CLI for API configuration
+API_ENV_VARS = [
+    "ANTHROPIC_BASE_URL",
+    "ANTHROPIC_AUTH_TOKEN",
+    "API_TIMEOUT_MS",
+    "ANTHROPIC_DEFAULT_SONNET_MODEL",
+    "ANTHROPIC_DEFAULT_OPUS_MODEL",
+    "ANTHROPIC_DEFAULT_HAIKU_MODEL",
+]
+
 # Read-only feature MCP tools
 READONLY_FEATURE_MCP_TOOLS = [
     "mcp__features__feature_get_stats",
@@ -234,6 +244,9 @@ class AssistantChatSession:
         # Use system Claude CLI
         system_cli = shutil.which("claude")
 
+        # Build environment overrides for API configuration
+        sdk_env = {var: os.getenv(var) for var in API_ENV_VARS if os.getenv(var)}
+
         try:
             self.client = ClaudeSDKClient(
                 options=ClaudeAgentOptions(
@@ -246,6 +259,7 @@ class AssistantChatSession:
                     max_turns=100,
                     cwd=str(self.project_dir.resolve()),
                     settings=str(settings_file.resolve()),
+                    env=sdk_env,
                 )
             )
             await self.client.__aenter__()
diff --git a/server/services/expand_chat_session.py b/server/services/expand_chat_session.py
index b187804..71e56bb 100644
--- a/server/services/expand_chat_session.py
+++ b/server/services/expand_chat_session.py
@@ -9,6 +9,7 @@ Uses the expand-project.md skill to help users add features to existing projects
 import asyncio
 import json
 import logging
+import os
 import re
 import shutil
 import threading
@@ -27,6 +28,16 @@ load_dotenv()
 
 logger = logging.getLogger(__name__)
 
+# Environment variables to pass through to Claude CLI for API configuration
+API_ENV_VARS = [
+    "ANTHROPIC_BASE_URL",
+    "ANTHROPIC_AUTH_TOKEN",
+    "API_TIMEOUT_MS",
+    "ANTHROPIC_DEFAULT_SONNET_MODEL",
+    "ANTHROPIC_DEFAULT_OPUS_MODEL",
+    "ANTHROPIC_DEFAULT_HAIKU_MODEL",
+]
+
 
 async def _make_multimodal_message(content_blocks: list[dict]) -> AsyncGenerator[dict, None]:
     """
@@ -153,6 +164,9 @@ class ExpandChatSession:
         project_path = str(self.project_dir.resolve())
         system_prompt = skill_content.replace("$ARGUMENTS", project_path)
 
+        # Build environment overrides for API configuration
+        sdk_env = {var: os.getenv(var) for var in API_ENV_VARS if os.getenv(var)}
+
         # Create Claude SDK client
         try:
             self.client = ClaudeSDKClient(
@@ -168,6 +182,7 @@ class ExpandChatSession:
                     max_turns=100,
                     cwd=str(self.project_dir.resolve()),
                     settings=str(settings_file.resolve()),
+                    env=sdk_env,
                 )
             )
             await self.client.__aenter__()
diff --git a/server/services/spec_chat_session.py b/server/services/spec_chat_session.py
index b3b4e1c..e073a4e 100644
--- a/server/services/spec_chat_session.py
+++ b/server/services/spec_chat_session.py
@@ -8,6 +8,7 @@ Uses the create-spec.md skill to guide users through app spec creation.
 
 import json
 import logging
+import os
 import shutil
 import threading
 from datetime import datetime
@@ -24,6 +25,16 @@ load_dotenv()
 
 logger = logging.getLogger(__name__)
 
+# Environment variables to pass through to Claude CLI for API configuration
+API_ENV_VARS = [
+    "ANTHROPIC_BASE_URL",
+    "ANTHROPIC_AUTH_TOKEN",
+    "API_TIMEOUT_MS",
+    "ANTHROPIC_DEFAULT_SONNET_MODEL",
+    "ANTHROPIC_DEFAULT_OPUS_MODEL",
+    "ANTHROPIC_DEFAULT_HAIKU_MODEL",
+]
+
 
 async def _make_multimodal_message(content_blocks: list[dict]) -> AsyncGenerator[dict, None]:
     """
@@ -147,6 +158,10 @@ class SpecChatSession:
         # Use Opus for best quality spec generation
         # Use system Claude CLI to avoid bundled Bun runtime crash (exit code 3) on Windows
         system_cli = shutil.which("claude")
+
+        # Build environment overrides for API configuration
+        sdk_env = {var: os.getenv(var) for var in API_ENV_VARS if os.getenv(var)}
+
         try:
             self.client = ClaudeSDKClient(
                 options=ClaudeAgentOptions(
@@ -163,6 +178,7 @@ class SpecChatSession:
                     max_turns=100,
                     cwd=str(self.project_dir.resolve()),
                     settings=str(settings_file.resolve()),
+                    env=sdk_env,
                 )
             )
             # Enter the async context and track it
diff --git a/ui/src/App.tsx b/ui/src/App.tsx
index 4a33b9e..50b0297 100644
--- a/ui/src/App.tsx
+++ b/ui/src/App.tsx
@@ -1,6 +1,6 @@
 import { useState, useEffect, useCallback } from 'react'
 import { useQueryClient } from '@tanstack/react-query'
-import { useProjects, useFeatures, useAgentStatus } from './hooks/useProjects'
+import { useProjects, useFeatures, useAgentStatus, useSettings } from './hooks/useProjects'
 import { useProjectWebSocket } from './hooks/useWebSocket'
 import { useFeatureSound } from './hooks/useFeatureSound'
 import { useCelebration } from './hooks/useCelebration'
@@ -46,6 +46,7 @@ function App() {
   const queryClient = useQueryClient()
   const { data: projects, isLoading: projectsLoading } = useProjects()
   const { data: features } = useFeatures(selectedProject)
+  const { data: settings } = useSettings()
   useAgentStatus(selectedProject) // Keep polling for status updates
   const wsState = useProjectWebSocket(selectedProject)
 
@@ -210,6 +211,16 @@ function App() {
                   >
                     <Settings size={18} />
                   </button>
+
+                  {/* GLM Mode Badge */}
+                  {settings?.glm_mode && (
+                    <span
+                      className="px-2 py-1 text-xs font-bold bg-purple-500 text-white rounded border-2 border-black shadow-neo-sm"
+                      title="Using GLM API (configured via .env)"
+                    >
+                      GLM
+                    </span>
+                  )}
                 </>
               )}
             </div>
diff --git a/ui/src/hooks/useProjects.ts b/ui/src/hooks/useProjects.ts
index 6a1098f..d6081a7 100644
--- a/ui/src/hooks/useProjects.ts
+++ b/ui/src/hooks/useProjects.ts
@@ -217,6 +217,7 @@ const DEFAULT_MODELS: ModelsResponse = {
 const DEFAULT_SETTINGS: Settings = {
   yolo_mode: false,
   model: 'claude-opus-4-5-20251101',
+  glm_mode: false,
 }
 
 export function useAvailableModels() {
diff --git a/ui/src/lib/types.ts b/ui/src/lib/types.ts
index cceb704..0851617 100644
--- a/ui/src/lib/types.ts
+++ b/ui/src/lib/types.ts
@@ -387,6 +387,7 @@ export interface ModelsResponse {
 export interface Settings {
   yolo_mode: boolean
   model: string
+  glm_mode: boolean
 }
 
 export interface SettingsUpdate {