commit 202731df7487c081645ea6637ad8ceb7bc1ec555
Author: h88782481 <54714341+h88782481@users.noreply.github.com>
Date:   Mon Mar 9 14:18:42 2026 +0800

    初始化提交

diff --git a/.dockerignore b/.dockerignore
new file mode 100644
index 0000000..e6d35eb
--- /dev/null
+++ b/.dockerignore
@@ -0,0 +1,31 @@
+# 版本管理
+.git/
+.gitignore
+
+# 环境变量
+.env
+.env.*
+!.env.example
+
+# Python 缓存
+__pycache__/
+*.pyc
+*.pyo
+*.pyd
+
+# 虚拟环境
+.venv/
+venv/
+
+# IDE
+.vscode/
+.idea/
+.cursor/
+
+# Docker 自身
+Dockerfile
+compose.yml
+.dockerignore
+
+# 文档
+README.md
diff --git a/.env.example b/.env.example
new file mode 100644
index 0000000..dd6bc1e
--- /dev/null
+++ b/.env.example
@@ -0,0 +1,6 @@
+PROXY_TARGET_URL=https://your-relay-station.com
+PROXY_API_KEY=sk-xxx
+PROXY_PORT=3029
+API_TIMEOUT=300
+ACCESS_API_KEY=your-access-key
+DEBUG=false
diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..d951f32
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,20 @@
+# Python 缓存
+__pycache__/
+*.pyc
+*.pyo
+*.pyd
+
+# 环境变量
+.env
+
+# 虚拟环境
+.venv/
+venv/
+
+# 运行时数据
+data/
+
+# IDE
+.vscode/
+.idea/
+.cursor/
diff --git a/Dockerfile b/Dockerfile
new file mode 100644
index 0000000..4dc3f9c
--- /dev/null
+++ b/Dockerfile
@@ -0,0 +1,25 @@
+# ---- 阶段 1: 安装依赖 ----
+FROM python:3.13-alpine AS builder
+
+WORKDIR /build
+COPY requirements.txt .
+RUN pip install --no-cache-dir --prefix=/install -r requirements.txt
+
+# ---- 阶段 2: 运行环境 ----
+FROM python:3.13-alpine
+
+WORKDIR /app
+
+COPY --from=builder /install /usr/local
+
+COPY *.py ./
+COPY routes/ routes/
+COPY adapters/ adapters/
+COPY utils/ utils/
+COPY static/ static/
+
+RUN mkdir -p data
+
+EXPOSE 3029
+
+CMD ["python", "start.py"]
diff --git a/LICENSE b/LICENSE
new file mode 100644
index 0000000..b77bf2a
--- /dev/null
+++ b/LICENSE
@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
diff --git a/README.md b/README.md
new file mode 100644
index 0000000..6cd334c
--- /dev/null
+++ b/README.md
@@ -0,0 +1,129 @@
+# API 2 Cursor
+
+让 Cursor 通过第三方中转站使用任意 LLM 模型的 API 代理服务。
+
+## 它解决什么问题
+
+Cursor 根据模型名发送不同格式的请求：
+
+| Cursor 模型名风格 | 请求格式 |
+|---|---|
+| `claude-sonnet-*`、`glm-*` | `/v1/chat/completions` (OpenAI CC) |
+| `gpt-*`、`claude-opus-*` | `/v1/responses` (OpenAI Responses) |
+
+而中转站通常只支持 `/v1/chat/completions` 或 `/v1/messages`。
+
+本项目在中间做协议转换，**不管 Cursor 发什么格式，都能正确转发到中转站；不管中转站返回什么格式，都让 Cursor 能正确接收**。
+
+## 架构
+
+```
+Cursor                      API 2 Cursor                      中转站
+  │                              │                              │
+  ├─ /v1/chat/completions ──→ chat.py ─┬─ openai 后端 ────────→ /v1/chat/completions
+  │                                     └─ anthropic 后端 ────→ /v1/messages
+  │                              │
+  ├─ /v1/responses ──────→ responses.py → 转为 CC → 同上 → 转回 Responses
+  │                              │
+  └─ /v1/messages ───────→ messages.py → 直接透传 ────────────→ /v1/messages
+```
+
+## 快速开始
+
+### 直接运行
+
+```bash
+cd api2cursor
+pip install -r requirements.txt
+cp .env.example .env
+# 编辑 .env 填入中转站地址和密钥
+python start.py
+```
+
+### Docker 部署
+
+```bash
+cd api2cursor
+cp .env.example .env
+# 编辑 .env
+docker compose up -d
+```
+
+服务启动后访问 `http://localhost:3029/admin` 进入管理面板。
+
+## 配置
+
+### 环境变量
+
+| 变量 | 说明 | 默认值 |
+|---|---|---|
+| `PROXY_TARGET_URL` | 上游中转站地址 | `https://api.anthropic.com` |
+| `PROXY_API_KEY` | 上游 API 密钥 | |
+| `PROXY_PORT` | 服务监听端口 | `3029` |
+| `API_TIMEOUT` | 请求超时（秒） | `300` |
+| `ACCESS_API_KEY` | 访问鉴权密钥，留空不启用 | |
+| `DEBUG` | 调试模式，输出详细请求/响应日志 | `false` |
+
+### 模型映射
+
+在管理面板 (`/admin`) 中配置模型映射：
+
+- **Cursor 模型名** — 在 Cursor 自定义模型中填入的名称
+- **上游模型名** — 发送到中转站的实际模型名
+- **后端类型** — `openai` (CC 格式) / `anthropic` (Messages 格式) / `auto` (自动检测)
+- **自定义地址/密钥** — 可选，覆盖全局设置，实现分流到不同中转站
+
+**示例**：在 Cursor 中添加 `claude-sonnet-4-5-20250929`，映射到上游 `gpt-5.3-codex`，后端选 `openai`。Cursor 会用 CC 格式发送请求，代理直接转发到中转站的 `/v1/chat/completions`。
+
+> **提示**：使用 Claude 风格的模型名（如 `claude-sonnet-4-5-20250929`）可以让 Cursor 显示思考过程（thinking）。
+
+### 在 Cursor 中配置
+
+1. 打开 Cursor 设置 → Models
+2. 添加自定义模型，名称填映射中配置的 Cursor 模型名
+3. Override OpenAI Base URL 填 `http://localhost:3029`
+4. API Key 填 `ACCESS_API_KEY` 的值（未配置则随意填）
+
+## 项目结构
+
+```
+api2cursor/
+├── start.py               # 启动入口
+├── app.py                 # Flask 应用工厂
+├── config.py              # 环境变量配置
+├── settings.py            # 持久化配置管理
+├── routes/                # 路由层
+│   ├── chat.py            #   /v1/chat/completions
+│   ├── responses.py       #   /v1/responses
+│   ├── messages.py        #   /v1/messages (透传)
+│   └── admin.py           #   管理面板 + API
+├── adapters/              # 适配层（格式转换）
+│   ├── openai_anthropic.py#   CC ↔ Messages 双向转换
+│   ├── openai_fixer.py    #   OpenAI 请求/响应修复
+│   └── responses_adapter.py#  Responses ↔ CC 双向转换
+├── utils/                 # 工具层
+│   ├── http.py            #   请求转发、SSE 解析
+│   ├── tool_fixer.py      #   工具参数修复
+│   └── think_tag.py       #   <think> 标签提取
+└── static/                # 管理面板前端
+    ├── admin.html
+    ├── admin.css
+    └── admin.js
+```
+
+## 兼容性修复
+
+代理自动处理以下兼容性问题：
+
+- Cursor 扁平格式 tools → 标准 OpenAI 嵌套格式
+- `reasoningContent` → `reasoning_content`
+- `<think>` 标签 → `reasoning_content`
+- 旧版 `function_call` → 新版 `tool_calls`
+- `tool_calls` 缺失 `id` / `index` / `type` 字段补全
+- 智能引号 → 普通引号（StrReplace 工具精确匹配修复）
+- `file_path` → `path` 字段映射
+- `finish_reason` 修正
+
+## 许可证
+
+[MIT](LICENSE)
diff --git a/adapters/__init__.py b/adapters/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/adapters/openai_anthropic.py b/adapters/openai_anthropic.py
new file mode 100644
index 0000000..50c508d
--- /dev/null
+++ b/adapters/openai_anthropic.py
@@ -0,0 +1,350 @@
+"""OpenAI Chat Completions ↔ Anthropic Messages 格式转换
+
+请求方向: CC → Messages（Cursor 的 CC 请求转为 Anthropic 格式发给上游）
+响应方向: Messages → CC（上游 Anthropic 响应转为 CC 格式返回给 Cursor）
+包含非流式和流式两种转换。
+"""
+
+import json
+import uuid
+import logging
+
+from utils.tool_fixer import normalize_args, repair_str_replace_args, fix_anthropic_tool_use
+from utils.http import gen_id
+
+logger = logging.getLogger(__name__)
+
+# Anthropic stop_reason → OpenAI finish_reason
+_STOP_REASON_MAP = {
+    'end_turn': 'stop',
+    'max_tokens': 'length',
+    'tool_use': 'tool_calls',
+    'stop_sequence': 'stop',
+}
+
+
+# ═══════════════════════════════════════════════════════════
+#  请求转换: CC → Messages
+# ═══════════════════════════════════════════════════════════
+
+
+def cc_to_messages_request(payload):
+    """将 OpenAI CC 格式请求转换为 Anthropic Messages 格式"""
+    messages = payload.get('messages', [])
+    anthropic_msgs = []
+    system_parts = []
+
+    for msg in messages:
+        role = msg.get('role', '')
+        content = msg.get('content', '')
+
+        # system 消息提取到顶层
+        if role == 'system':
+            system_parts.append(_flatten_text(content))
+            continue
+
+        anthropic_role = 'assistant' if role == 'assistant' else 'user'
+        anthropic_content = _convert_content(msg)
+
+        # assistant 的 tool_calls → tool_use content blocks
+        if role == 'assistant' and 'tool_calls' in msg:
+            blocks = _to_blocks(anthropic_content)
+            for tc in msg['tool_calls']:
+                func = tc.get('function', {})
+                arguments = func.get('arguments', '{}')
+                if isinstance(arguments, str):
+                    try:
+                        arguments = json.loads(arguments)
+                    except json.JSONDecodeError:
+                        arguments = {}
+                blocks.append({
+                    'type': 'tool_use',
+                    'id': tc.get('id', f'toolu_{uuid.uuid4().hex[:24]}'),
+                    'name': func.get('name', ''),
+                    'input': arguments,
+                })
+            anthropic_content = blocks
+
+        # tool 角色 → user + tool_result
+        if role == 'tool':
+            text = content if isinstance(content, str) else json.dumps(content)
+            anthropic_content = [{
+                'type': 'tool_result',
+                'tool_use_id': msg.get('tool_call_id', ''),
+                'content': text,
+            }]
+            anthropic_role = 'user'
+
+        if not anthropic_content and anthropic_content != 0:
+            continue
+
+        anthropic_msgs.append({'role': anthropic_role, 'content': anthropic_content})
+
+    # Anthropic 要求角色必须交替
+    anthropic_msgs = _merge_same_role(anthropic_msgs)
+
+    result = {
+        'model': payload.get('model', 'claude-sonnet-4-20250514'),
+        'messages': anthropic_msgs,
+        'max_tokens': max(payload.get('max_tokens') or 8192, 8192),
+    }
+
+    if system_parts:
+        result['system'] = '\n\n'.join(system_parts)
+    if 'tools' in payload:
+        result['tools'] = _convert_tools(payload['tools'])
+    for key in ('temperature', 'top_p', 'stream'):
+        if key in payload:
+            result[key] = payload[key]
+
+    return result
+
+
+# ═══════════════════════════════════════════════════════════
+#  非流式响应转换: Messages → CC
+# ═══════════════════════════════════════════════════════════
+
+
+def messages_to_cc_response(data, request_id=None):
+    """将 Anthropic Messages 响应转换为 OpenAI CC 格式"""
+    request_id = request_id or gen_id('chatcmpl-')
+    data = fix_anthropic_tool_use(data)
+
+    content_text = ''
+    reasoning = ''
+    tool_calls = []
+
+    for block in data.get('content', []):
+        if not isinstance(block, dict):
+            continue
+        btype = block.get('type', '')
+        if btype == 'text':
+            content_text += block.get('text', '')
+        elif btype == 'thinking':
+            reasoning += block.get('thinking', '')
+        elif btype == 'tool_use':
+            args = block.get('input', {})
+            if isinstance(args, dict):
+                args = normalize_args(args)
+                args = repair_str_replace_args(block.get('name', ''), args)
+            tool_calls.append({
+                'index': len(tool_calls),
+                'id': block.get('id', f'toolu_{uuid.uuid4().hex[:24]}'),
+                'type': 'function',
+                'function': {
+                    'name': block.get('name', ''),
+                    'arguments': json.dumps(args, ensure_ascii=False) if isinstance(args, dict) else str(args),
+                },
+            })
+
+    stop_reason = data.get('stop_reason', 'end_turn')
+    message = {'role': 'assistant', 'content': content_text or None}
+    if reasoning:
+        message['reasoning_content'] = reasoning
+    if tool_calls:
+        message['tool_calls'] = tool_calls
+
+    usage = data.get('usage', {})
+    return {
+        'id': request_id,
+        'object': 'chat.completion',
+        'model': data.get('model', 'claude'),
+        'choices': [{
+            'index': 0,
+            'message': message,
+            'finish_reason': _STOP_REASON_MAP.get(stop_reason, 'stop'),
+        }],
+        'usage': {
+            'prompt_tokens': usage.get('input_tokens', 0),
+            'completion_tokens': usage.get('output_tokens', 0),
+            'total_tokens': usage.get('input_tokens', 0) + usage.get('output_tokens', 0),
+        },
+    }
+
+
+# ═══════════════════════════════════════════════════════════
+#  流式响应转换: Anthropic SSE → CC chunks
+# ═══════════════════════════════════════════════════════════
+
+
+class AnthropicStreamConverter:
+    """将 Anthropic SSE 事件逐个转换为 OpenAI CC 流式 chunk"""
+
+    def __init__(self, request_id=None):
+        self._id = request_id or gen_id('chatcmpl-')
+        self._tool_index = -1
+        self._input_tokens = 0
+        self._output_tokens = 0
+
+    def process_event(self, event_type, event_data):
+        """处理一个 Anthropic SSE 事件，返回 CC chunk JSON 字符串列表"""
+        chunks = []
+
+        if event_type == 'message_start':
+            msg = event_data.get('message', {})
+            self._input_tokens = msg.get('usage', {}).get('input_tokens', 0)
+            chunk = self._make_chunk(delta={'role': 'assistant', 'content': ''})
+            if msg.get('model'):
+                chunk['model'] = msg['model']
+            chunks.append(json.dumps(chunk))
+
+        elif event_type == 'content_block_start':
+            block = event_data.get('content_block', {})
+            if block.get('type') == 'tool_use':
+                self._tool_index += 1
+                chunks.append(json.dumps(self._make_chunk(delta={
+                    'tool_calls': [{
+                        'index': self._tool_index,
+                        'id': block.get('id', f'toolu_{uuid.uuid4().hex[:24]}'),
+                        'type': 'function',
+                        'function': {'name': block.get('name', ''), 'arguments': ''},
+                    }]
+                })))
+
+        elif event_type == 'content_block_delta':
+            delta = event_data.get('delta', {})
+            dtype = delta.get('type', '')
+            if dtype == 'text_delta' and delta.get('text'):
+                chunks.append(json.dumps(self._make_chunk(
+                    delta={'content': delta['text']})))
+            elif dtype == 'thinking_delta' and delta.get('thinking'):
+                chunks.append(json.dumps(self._make_chunk(
+                    delta={'reasoning_content': delta['thinking']})))
+            elif dtype == 'input_json_delta' and delta.get('partial_json'):
+                chunks.append(json.dumps(self._make_chunk(delta={
+                    'tool_calls': [{
+                        'index': self._tool_index,
+                        'function': {'arguments': delta['partial_json']},
+                    }]
+                })))
+
+        elif event_type == 'message_delta':
+            delta = event_data.get('delta', {})
+            usage = event_data.get('usage', {})
+            self._output_tokens = usage.get('output_tokens', 0)
+            finish = _STOP_REASON_MAP.get(delta.get('stop_reason', ''), 'stop')
+            chunk = self._make_chunk(delta={}, finish_reason=finish)
+            chunk['usage'] = {
+                'prompt_tokens': self._input_tokens,
+                'completion_tokens': self._output_tokens,
+                'total_tokens': self._input_tokens + self._output_tokens,
+            }
+            chunks.append(json.dumps(chunk))
+
+        return chunks
+
+    def _make_chunk(self, delta, finish_reason=None):
+        choice = {'index': 0, 'delta': delta}
+        if finish_reason:
+            choice['finish_reason'] = finish_reason
+        return {
+            'id': self._id,
+            'object': 'chat.completion.chunk',
+            'model': 'claude',
+            'choices': [choice],
+        }
+
+
+# ═══════════════════════════════════════════════════════════
+#  内部辅助函数
+# ═══════════════════════════════════════════════════════════
+
+
+def _flatten_text(content):
+    """将 content 扁平化为纯文本"""
+    if isinstance(content, str):
+        return content
+    if isinstance(content, list):
+        parts = []
+        for p in content:
+            if isinstance(p, str):
+                parts.append(p)
+            elif isinstance(p, dict) and p.get('type') == 'text':
+                parts.append(p.get('text', ''))
+        return '\n'.join(parts)
+    return str(content)
+
+
+def _convert_content(msg):
+    """将 OpenAI 消息的 content 字段转为 Anthropic 格式"""
+    content = msg.get('content', '')
+    if content is None:
+        return ''
+    if isinstance(content, str):
+        return content
+    if isinstance(content, list):
+        blocks = []
+        for part in content:
+            if isinstance(part, str):
+                blocks.append({'type': 'text', 'text': part})
+            elif isinstance(part, dict):
+                ptype = part.get('type', '')
+                if ptype == 'text':
+                    blocks.append({'type': 'text', 'text': part.get('text', '')})
+                elif ptype == 'image_url':
+                    blocks.append(_convert_image(part))
+                elif ptype in ('tool_use', 'tool_result'):
+                    blocks.append(part)
+        return blocks
+    return str(content)
+
+
+def _convert_image(part):
+    """将 OpenAI image_url 格式转为 Anthropic image 格式"""
+    url_data = part.get('image_url', {})
+    url = url_data.get('url', '') if isinstance(url_data, dict) else str(url_data)
+    if url.startswith('data:'):
+        media_type, _, b64 = url.partition(';base64,')
+        return {
+            'type': 'image',
+            'source': {
+                'type': 'base64',
+                'media_type': media_type.replace('data:', '') or 'image/png',
+                'data': b64,
+            },
+        }
+    return {'type': 'image', 'source': {'type': 'url', 'url': url}}
+
+
+def _convert_tools(tools):
+    """将 OpenAI tools 转为 Anthropic tools 格式（兼容 Cursor 扁平格式）"""
+    result = []
+    for tool in tools:
+        if tool.get('type') == 'function' and 'function' in tool:
+            func = tool['function']
+            result.append({
+                'name': func.get('name', ''),
+                'description': func.get('description', ''),
+                'input_schema': func.get('parameters', {'type': 'object', 'properties': {}}),
+            })
+        elif 'name' in tool and 'input_schema' in tool:
+            result.append({
+                'name': tool.get('name', ''),
+                'description': tool.get('description', ''),
+                'input_schema': tool.get('input_schema', {'type': 'object', 'properties': {}}),
+            })
+    return result
+
+
+def _to_blocks(content):
+    """将 content 统一转为 blocks 列表"""
+    if isinstance(content, str):
+        return [{'type': 'text', 'text': content}] if content else []
+    if isinstance(content, list):
+        return list(content)
+    return [{'type': 'text', 'text': str(content)}] if content else []
+
+
+def _merge_same_role(messages):
+    """合并相邻同角色消息（Anthropic 要求角色必须交替）"""
+    if not messages:
+        return messages
+    merged = [messages[0]]
+    for msg in messages[1:]:
+        if msg['role'] == merged[-1]['role']:
+            prev = _to_blocks(merged[-1]['content'])
+            curr = _to_blocks(msg['content'])
+            merged[-1]['content'] = prev + curr
+        else:
+            merged.append(msg)
+    return merged
diff --git a/adapters/openai_fixer.py b/adapters/openai_fixer.py
new file mode 100644
index 0000000..971def8
--- /dev/null
+++ b/adapters/openai_fixer.py
@@ -0,0 +1,267 @@
+"""OpenAI 格式修复
+
+修复 Cursor 发出的 OpenAI 格式请求和上游返回的响应中的各种兼容性问题：
+  请求修复: Cursor 扁平格式 tools → 标准嵌套格式, tool_choice 规范化
+  响应修复: reasoningContent → reasoning_content, <think> 标签提取,
+           function_call → tool_calls, tool_calls 字段补全, 参数修复
+"""
+
+import json
+import logging
+
+from utils.http import gen_id
+from utils.tool_fixer import normalize_args, repair_str_replace_args
+from utils.think_tag import extract_from_text
+
+logger = logging.getLogger(__name__)
+
+
+# ─── 请求预处理 ───────────────────────────────────
+
+
+def normalize_request(payload, upstream_model=None):
+    """预处理 Cursor 发来的 OpenAI 格式请求"""
+    if upstream_model:
+        payload['model'] = upstream_model
+
+    # Cursor 可能在 CC 端点发送 Anthropic 格式的 tool_use/tool_result 消息
+    if 'messages' in payload:
+        payload['messages'] = _convert_anthropic_messages(payload['messages'])
+
+    if 'tools' not in payload:
+        return payload
+
+    # 修复 Cursor 可能发出的扁平格式 tools
+    normalized = []
+    for tool in payload['tools']:
+        if tool.get('type') == 'function' and 'function' in tool:
+            normalized.append(tool)
+        elif 'name' in tool:
+            normalized.append({
+                'type': 'function',
+                'function': {
+                    'name': tool.get('name', ''),
+                    'description': tool.get('description', ''),
+                    'parameters': tool.get('input_schema')
+                                  or tool.get('parameters')
+                                  or {'type': 'object', 'properties': {}},
+                },
+            })
+        else:
+            normalized.append(tool)
+    payload['tools'] = normalized
+
+    # tool_choice 规范化
+    tc = payload.get('tool_choice')
+    if isinstance(tc, dict):
+        if tc.get('type') == 'auto':
+            payload['tool_choice'] = 'auto'
+        elif tc.get('type') == 'any':
+            payload['tool_choice'] = 'required'
+
+    return payload
+
+
+def _convert_anthropic_messages(messages):
+    """将消息中的 Anthropic 格式 tool_use/tool_result 转为 OpenAI 格式
+
+    Cursor 有时在 CC 端点中发送 Anthropic 风格的内容块：
+      assistant: [{"type":"tool_use", "id":"...", "name":"Read", "input":{...}}]
+      user:      [{"type":"tool_result", "tool_use_id":"...", "content":[...]}]
+    OpenAI 格式应为：
+      assistant: {"tool_calls":[{"id":"...", "function":{"name":"Read","arguments":"..."}}]}
+      tool:      {"tool_call_id":"...", "content":"..."}
+    """
+    converted = []
+    for msg in messages:
+        content = msg.get('content')
+        if not isinstance(content, list):
+            converted.append(msg)
+            continue
+
+        has_tool_use = any(
+            isinstance(b, dict) and b.get('type') == 'tool_use' for b in content
+        )
+        has_tool_result = any(
+            isinstance(b, dict) and b.get('type') == 'tool_result' for b in content
+        )
+
+        if not has_tool_use and not has_tool_result:
+            converted.append(msg)
+            continue
+
+        role = msg.get('role', '')
+
+        if role == 'assistant' and has_tool_use:
+            text_parts = []
+            tool_calls = []
+            for block in content:
+                if not isinstance(block, dict):
+                    continue
+                if block.get('type') == 'text':
+                    text_parts.append(block.get('text', ''))
+                elif block.get('type') == 'tool_use':
+                    tool_calls.append({
+                        'id': block.get('id', gen_id('call_')),
+                        'type': 'function',
+                        'function': {
+                            'name': block.get('name', ''),
+                            'arguments': json.dumps(
+                                block.get('input', {}), ensure_ascii=False
+                            ),
+                        },
+                    })
+            new_msg = {'role': 'assistant'}
+            new_msg['content'] = '\n'.join(text_parts) if text_parts else None
+            if tool_calls:
+                new_msg['tool_calls'] = tool_calls
+            converted.append(new_msg)
+
+        elif has_tool_result:
+            other_parts = []
+            for block in content:
+                if not isinstance(block, dict):
+                    continue
+                if block.get('type') == 'tool_result':
+                    rc = block.get('content', '')
+                    if isinstance(rc, list):
+                        rc = '\n'.join(
+                            b.get('text', '') for b in rc
+                            if isinstance(b, dict) and b.get('type') == 'text'
+                        )
+                    elif not isinstance(rc, str):
+                        rc = str(rc)
+                    converted.append({
+                        'role': 'tool',
+                        'tool_call_id': block.get('tool_use_id', ''),
+                        'content': rc,
+                    })
+                else:
+                    other_parts.append(block)
+            if other_parts:
+                converted.append({'role': role, 'content': other_parts})
+        else:
+            converted.append(msg)
+
+    return converted
+
+
+# ─── 非流式响应修复 ───────────────────────────────
+
+
+def fix_response(data):
+    """修复上游返回的非流式 OpenAI 响应"""
+    if not isinstance(data, dict):
+        return data
+
+    for choice in (data.get('choices') or []):
+        msg = choice.get('message') or {}
+
+        # reasoningContent → reasoning_content
+        if 'reasoningContent' in msg and 'reasoning_content' not in msg:
+            msg['reasoning_content'] = msg.pop('reasoningContent')
+
+        # <think> 标签 → reasoning_content
+        content = msg.get('content') or ''
+        if isinstance(content, str) and '<think>' in content and not msg.get('reasoning_content'):
+            cleaned, reasoning = extract_from_text(content)
+            if reasoning:
+                msg['reasoning_content'] = reasoning
+                msg['content'] = cleaned
+                logger.info(f'提取 <think> 标签 → reasoning_content ({len(reasoning)} 字符)')
+
+        # 旧版 function_call → 新版 tool_calls
+        if 'function_call' in msg and 'tool_calls' not in msg:
+            fc = msg.pop('function_call')
+            msg['tool_calls'] = [{
+                'id': gen_id('call_'),
+                'type': 'function',
+                'function': {
+                    'name': fc.get('name', ''),
+                    'arguments': fc.get('arguments', '{}'),
+                },
+            }]
+            if choice.get('finish_reason') == 'function_call':
+                choice['finish_reason'] = 'tool_calls'
+
+        # 修复 tool_calls 字段
+        _fix_tool_calls(msg, choice)
+
+    return data
+
+
+# ─── 流式 chunk 修复 ──────────────────────────────
+
+
+def fix_stream_chunk(data):
+    """修复上游返回的流式 OpenAI chunk"""
+    if not isinstance(data, dict):
+        return data
+
+    for choice in (data.get('choices') or []):
+        delta = choice.get('delta') or {}
+
+        # reasoningContent → reasoning_content
+        if 'reasoningContent' in delta and 'reasoning_content' not in delta:
+            delta['reasoning_content'] = delta.pop('reasoningContent')
+
+        # 旧版 function_call → tool_calls
+        if 'function_call' in delta and 'tool_calls' not in delta:
+            fc = delta.pop('function_call')
+            tc = {'index': 0, 'type': 'function', 'function': {}}
+            if 'name' in fc:
+                tc['id'] = gen_id('call_')
+                tc['function']['name'] = fc['name']
+            if 'arguments' in fc:
+                tc['function']['arguments'] = fc['arguments']
+            delta['tool_calls'] = [tc]
+            if choice.get('finish_reason') == 'function_call':
+                choice['finish_reason'] = 'tool_calls'
+
+        # 补全 tool_calls 字段
+        for tc in (delta.get('tool_calls') or []):
+            if 'index' not in tc:
+                tc['index'] = 0
+            func = tc.get('function') or {}
+            if 'id' in tc or 'name' in func:
+                if not tc.get('id'):
+                    tc['id'] = gen_id('call_')
+                if 'type' not in tc:
+                    tc['type'] = 'function'
+
+        if choice.get('finish_reason') == 'function_call':
+            choice['finish_reason'] = 'tool_calls'
+
+    return data
+
+
+# ─── 内部辅助 ─────────────────────────────────────
+
+
+def _fix_tool_calls(msg, choice):
+    """修复消息中的 tool_calls 字段"""
+    tool_calls = msg.get('tool_calls')
+    if not tool_calls:
+        return
+
+    for i, tc in enumerate(tool_calls):
+        if not tc.get('id'):
+            tc['id'] = gen_id('call_')
+        if 'index' not in tc:
+            tc['index'] = i
+        if tc.get('type') != 'function':
+            tc['type'] = 'function'
+
+        func = tc.get('function', {})
+        args_raw = func.get('arguments', '{}')
+        try:
+            args = json.loads(args_raw) if isinstance(args_raw, str) else (args_raw or {})
+        except json.JSONDecodeError:
+            args = {}
+
+        args = normalize_args(args)
+        args = repair_str_replace_args(func.get('name', ''), args)
+        func['arguments'] = json.dumps(args, ensure_ascii=False)
+
+    if choice.get('finish_reason') not in ('tool_calls', 'function_call'):
+        choice['finish_reason'] = 'tool_calls'
diff --git a/adapters/responses_adapter.py b/adapters/responses_adapter.py
new file mode 100644
index 0000000..18cd2c8
--- /dev/null
+++ b/adapters/responses_adapter.py
@@ -0,0 +1,533 @@
+"""Responses API 适配
+
+Cursor 对 GPT/Claude-Opus 等模型使用 /v1/responses 格式。
+本模块将 Responses 格式与 Chat Completions 格式互相转换：
+  请求: Responses → CC
+  响应: CC → Responses（非流式 + 流式）
+  流式: 支持从 CC chunks 或 Anthropic SSE 事件直接转换
+"""
+
+import json
+import logging
+
+from utils.http import gen_id
+
+logger = logging.getLogger(__name__)
+
+
+# ═══════════════════════════════════════════════════════════
+#  请求转换: Responses → CC
+# ═══════════════════════════════════════════════════════════
+
+
+def responses_to_cc(payload):
+    """将 /v1/responses 请求转换为 /v1/chat/completions 格式"""
+    messages = []
+
+    if payload.get('instructions'):
+        messages.append({'role': 'system', 'content': payload['instructions']})
+
+    input_data = payload.get('input', [])
+    if isinstance(input_data, str):
+        messages.append({'role': 'user', 'content': input_data})
+    elif isinstance(input_data, list):
+        _convert_input_items(input_data, messages)
+
+    result = {
+        'model': payload.get('model', ''),
+        'messages': messages,
+        'stream': payload.get('stream', False),
+    }
+
+    if 'tools' in payload:
+        result['tools'] = _convert_tools(payload['tools'])
+    for key in ('temperature', 'top_p'):
+        if key in payload:
+            result[key] = payload[key]
+    if 'max_output_tokens' in payload:
+        result['max_tokens'] = payload['max_output_tokens']
+    if 'tool_choice' in payload:
+        result['tool_choice'] = payload['tool_choice']
+
+    return result
+
+
+# ═══════════════════════════════════════════════════════════
+#  非流式响应转换: CC → Responses
+# ═══════════════════════════════════════════════════════════
+
+
+def cc_to_responses(cc_resp, model=''):
+    """将 CC 响应转换为 Responses 格式"""
+    choice = (cc_resp.get('choices') or [{}])[0]
+    msg = choice.get('message') or {}
+    finish = choice.get('finish_reason', 'stop')
+
+    output = []
+
+    if msg.get('reasoning_content'):
+        output.append({
+            'type': 'reasoning',
+            'id': gen_id('rs_'),
+            'summary': [{'type': 'summary_text', 'text': msg['reasoning_content']}],
+        })
+
+    if msg.get('content'):
+        output.append({
+            'type': 'message',
+            'id': gen_id('msg_'),
+            'status': 'completed',
+            'role': 'assistant',
+            'content': [{'type': 'output_text', 'text': msg['content']}],
+        })
+
+    for tc in (msg.get('tool_calls') or []):
+        func = tc.get('function') or {}
+        output.append({
+            'type': 'function_call',
+            'id': gen_id('fc_'),
+            'status': 'completed',
+            'call_id': tc.get('id', gen_id('call_')),
+            'name': func.get('name', ''),
+            'arguments': func.get('arguments', '{}'),
+        })
+
+    usage = cc_resp.get('usage', {})
+    return {
+        'id': cc_resp.get('id', gen_id('resp_')),
+        'object': 'response',
+        'status': 'incomplete' if finish == 'length' else 'completed',
+        'model': model or cc_resp.get('model', ''),
+        'output': output,
+        'usage': {
+            'input_tokens': usage.get('prompt_tokens', 0),
+            'output_tokens': usage.get('completion_tokens', 0),
+            'total_tokens': usage.get('total_tokens', 0),
+        },
+    }
+
+
+# ═══════════════════════════════════════════════════════════
+#  流式转换器: CC chunks / Anthropic SSE → Responses SSE
+# ═══════════════════════════════════════════════════════════
+
+
+class ResponsesStreamConverter:
+    """有状态转换器：将 CC 流式 chunk 或 Anthropic SSE 事件转为 Responses SSE 事件"""
+
+    def __init__(self, response_id=None, model=''):
+        self.resp_id = response_id or gen_id('resp_')
+        self.model = model
+
+        # 思考内容缓冲
+        self._rs_buf = ''
+        self._rs_started = False
+        self._rs_closed = False
+        self._rs_id = gen_id('rs_')
+
+        # 文本内容缓冲
+        self._text_buf = ''
+        self._text_started = False
+        self._text_closed = False
+        self._msg_id = gen_id('msg_')
+
+        # 工具调用缓冲 {index: {name, args, call_id, fc_id}}
+        self._tools = {}
+        self._output_items = []
+        self._finished = False
+        self._input_tokens = 0
+
+    # ─── 公开接口 ─────────────────────────────────
+
+    def start_events(self):
+        """生成流开始事件"""
+        return [self._sse('response.created', {
+            'id': self.resp_id, 'object': 'response',
+            'status': 'in_progress', 'model': self.model, 'output': [],
+        })]
+
+    def process_cc_chunk(self, chunk):
+        """处理 CC 格式的流式 chunk，返回 Responses SSE 事件列表"""
+        events = []
+        for choice in (chunk.get('choices') or []):
+            delta = choice.get('delta') or {}
+            finish = choice.get('finish_reason')
+
+            if delta.get('reasoning_content'):
+                events.extend(self._on_reasoning(delta['reasoning_content']))
+            if delta.get('content') is not None and delta['content'] != '':
+                events.extend(self._on_text(delta['content']))
+            for tc in (delta.get('tool_calls') or []):
+                events.extend(self._on_tool_call(tc))
+            if finish and not self._finished:
+                self._finished = True
+                events.extend(self._do_finish(finish, chunk.get('usage')))
+
+        return events
+
+    def process_anthropic_event(self, event_type, event_data):
+        """直接处理 Anthropic SSE 事件（跳过 CC 中间转换，更高效）"""
+        events = []
+
+        if event_type == 'message_start':
+            usage = event_data.get('message', {}).get('usage', {})
+            self._input_tokens = usage.get('input_tokens', 0)
+
+        elif event_type == 'content_block_start':
+            block = event_data.get('content_block', {})
+            btype = block.get('type', '')
+            if btype == 'thinking' and not self._rs_started:
+                self._rs_started = True
+                events.append(self._sse('response.output_item.added', {
+                    'type': 'reasoning', 'id': self._rs_id, 'summary': [],
+                }))
+            elif btype == 'text':
+                events.extend(self._ensure_text_started())
+            elif btype == 'tool_use':
+                events.extend(self._start_tool_from_block(block))
+
+        elif event_type == 'content_block_delta':
+            delta = event_data.get('delta', {})
+            dtype = delta.get('type', '')
+            if dtype == 'thinking_delta' and delta.get('thinking'):
+                self._rs_buf += delta['thinking']
+                events.append(self._sse('response.reasoning_summary_text.delta', {
+                    'type': 'summary_text', 'delta': delta['thinking'],
+                }))
+            elif dtype == 'text_delta' and delta.get('text'):
+                self._text_buf += delta['text']
+                events.append(self._sse('response.output_text.delta', {
+                    'type': 'output_text', 'delta': delta['text'],
+                }))
+            elif dtype == 'input_json_delta' and delta.get('partial_json') and self._tools:
+                idx = max(self._tools.keys())
+                self._tools[idx]['args'] += delta['partial_json']
+                events.append(self._sse('response.function_call_arguments.delta', {
+                    'type': 'function_call', 'delta': delta['partial_json'],
+                }))
+
+        elif event_type == 'message_delta':
+            delta = event_data.get('delta', {})
+            stop = delta.get('stop_reason', 'end_turn')
+            usage = event_data.get('usage', {})
+            finish = {'tool_use': 'tool_calls', 'max_tokens': 'length'}.get(stop, 'stop')
+            if not self._finished:
+                self._finished = True
+                u = {
+                    'input_tokens': self._input_tokens,
+                    'output_tokens': usage.get('output_tokens', 0),
+                    'total_tokens': self._input_tokens + usage.get('output_tokens', 0),
+                }
+                events.extend(self._do_finish(finish, u))
+
+        return events
+
+    def finalize(self):
+        """流结束时补发未关闭的事件"""
+        if self._finished:
+            return []
+        self._finished = True
+        return self._do_finish('stop', None)
+
+    # ─── 内部事件处理 ─────────────────────────────
+
+    def _on_reasoning(self, text):
+        """处理思考内容 delta"""
+        events = []
+        if not self._rs_started:
+            self._rs_started = True
+            events.append(self._sse('response.output_item.added', {
+                'type': 'reasoning', 'id': self._rs_id, 'summary': [],
+            }))
+        self._rs_buf += text
+        events.append(self._sse('response.reasoning_summary_text.delta', {
+            'type': 'summary_text', 'delta': text,
+        }))
+        return events
+
+    def _on_text(self, text):
+        """处理文本内容 delta"""
+        events = self._ensure_text_started()
+        self._text_buf += text
+        events.append(self._sse('response.output_text.delta', {
+            'type': 'output_text', 'delta': text,
+        }))
+        return events
+
+    def _on_tool_call(self, tc):
+        """处理工具调用 delta"""
+        events = []
+        idx = tc.get('index', 0)
+        func = tc.get('function') or {}
+
+        if idx not in self._tools:
+            if self._rs_started and not self._rs_closed:
+                events.extend(self._close_reasoning())
+            if self._text_started and not self._text_closed:
+                events.extend(self._close_text())
+            call_id = tc.get('id', gen_id('call_'))
+            name = func.get('name', '')
+            fc_id = gen_id('fc_')
+            self._tools[idx] = {'name': name, 'args': '', 'call_id': call_id, 'fc_id': fc_id}
+            events.append(self._sse('response.output_item.added', {
+                'type': 'function_call', 'id': fc_id,
+                'status': 'in_progress', 'call_id': call_id,
+                'name': name, 'arguments': '',
+            }))
+
+        if func.get('name'):
+            self._tools[idx]['name'] = func['name']
+        if func.get('arguments', ''):
+            self._tools[idx]['args'] += func['arguments']
+            events.append(self._sse('response.function_call_arguments.delta', {
+                'type': 'function_call', 'delta': func['arguments'],
+            }))
+        return events
+
+    def _ensure_text_started(self):
+        """确保文本输出项已开始"""
+        events = []
+        if self._rs_started and not self._rs_closed:
+            events.extend(self._close_reasoning())
+        if not self._text_started:
+            self._text_started = True
+            events.append(self._sse('response.output_item.added', {
+                'type': 'message', 'id': self._msg_id,
+                'status': 'in_progress', 'role': 'assistant', 'content': [],
+            }))
+            events.append(self._sse('response.content_part.added', {
+                'type': 'output_text', 'text': '',
+            }))
+        return events
+
+    def _start_tool_from_block(self, block):
+        """从 Anthropic tool_use block 开始新的工具调用"""
+        events = []
+        if self._rs_started and not self._rs_closed:
+            events.extend(self._close_reasoning())
+        if self._text_started and not self._text_closed:
+            events.extend(self._close_text())
+        idx = len(self._tools)
+        tool_id = block.get('id', gen_id('toolu_'))
+        name = block.get('name', '')
+        fc_id = gen_id('fc_')
+        self._tools[idx] = {'name': name, 'args': '', 'call_id': tool_id, 'fc_id': fc_id}
+        events.append(self._sse('response.output_item.added', {
+            'type': 'function_call', 'id': fc_id,
+            'status': 'in_progress', 'call_id': tool_id,
+            'name': name, 'arguments': '',
+        }))
+        return events
+
+    # ─── 关闭/结束事件 ────────────────────────────
+
+    def _close_reasoning(self):
+        if self._rs_closed:
+            return []
+        self._rs_closed = True
+        rs = {
+            'type': 'reasoning', 'id': self._rs_id,
+            'summary': [{'type': 'summary_text', 'text': self._rs_buf}],
+        }
+        self._output_items.append(rs)
+        return [
+            self._sse('response.reasoning_summary_text.done', {
+                'type': 'summary_text', 'text': self._rs_buf,
+            }),
+            self._sse('response.output_item.done', rs),
+        ]
+
+    def _close_text(self):
+        if self._text_closed:
+            return []
+        self._text_closed = True
+        msg = {
+            'type': 'message', 'id': self._msg_id,
+            'status': 'completed', 'role': 'assistant',
+            'content': [{'type': 'output_text', 'text': self._text_buf}],
+        }
+        self._output_items.append(msg)
+        return [
+            self._sse('response.output_text.done', {'type': 'output_text', 'text': self._text_buf}),
+            self._sse('response.output_item.done', msg),
+        ]
+
+    def _do_finish(self, finish_reason, usage):
+        """生成流结束的所有关闭事件"""
+        events = []
+        if self._rs_started and not self._rs_closed:
+            events.extend(self._close_reasoning())
+        if self._text_started and not self._text_closed:
+            events.extend(self._close_text())
+
+        for idx in sorted(self._tools.keys()):
+            buf = self._tools[idx]
+            events.append(self._sse('response.function_call_arguments.done', {
+                'type': 'function_call', 'arguments': buf['args'],
+            }))
+            fc = {
+                'type': 'function_call', 'id': buf['fc_id'],
+                'status': 'completed', 'call_id': buf['call_id'],
+                'name': buf['name'], 'arguments': buf['args'],
+            }
+            events.append(self._sse('response.output_item.done', fc))
+            self._output_items.append(fc)
+
+        usage_data = usage if isinstance(usage, dict) else {}
+        events.append(self._sse('response.completed', {
+            'id': self.resp_id, 'object': 'response',
+            'status': 'incomplete' if finish_reason == 'length' else 'completed',
+            'model': self.model, 'output': self._output_items, 'usage': usage_data,
+        }))
+        return events
+
+    def _sse(self, event_type, data):
+        """构建 SSE 事件字符串"""
+        return f'event: {event_type}\ndata: {json.dumps(data, ensure_ascii=False)}\n\n'
+
+
+# ═══════════════════════════════════════════════════════════
+#  内部辅助函数
+# ═══════════════════════════════════════════════════════════
+
+
+def _convert_input_items(items, messages):
+    """将 Responses input 数组转换为 CC messages"""
+    i = 0
+    while i < len(items):
+        item = items[i]
+
+        if isinstance(item, str):
+            messages.append({'role': 'user', 'content': item})
+            i += 1
+            continue
+
+        if not isinstance(item, dict):
+            i += 1
+            continue
+
+        item_type = item.get('type', '')
+        role = item.get('role', '')
+
+        # 简单角色消息（无 type 字段）
+        if role and not item_type:
+            content = item.get('content', '')
+            if isinstance(content, list):
+                content = _extract_text(content)
+            messages.append({'role': role, 'content': content or ''})
+            i += 1
+            continue
+
+        # Responses message 对象
+        if item_type == 'message' or (role and not item_type):
+            role = item.get('role', 'assistant')
+            content = _extract_text(item.get('content', []))
+            msg = {'role': role, 'content': content or ''}
+            if role == 'assistant':
+                tool_calls, consumed = _collect_function_calls(items, i + 1)
+                if tool_calls:
+                    msg['tool_calls'] = tool_calls
+                    if not msg['content']:
+                        msg['content'] = None
+                    messages.append(msg)
+                    i += 1 + consumed
+                    continue
+            messages.append(msg)
+            i += 1
+            continue
+
+        # function_call（工具调用）
+        if item_type == 'function_call':
+            tc = {
+                'id': item.get('call_id') or gen_id('call_'),
+                'type': 'function',
+                'function': {
+                    'name': item.get('name', ''),
+                    'arguments': item.get('arguments', '{}'),
+                },
+            }
+            if messages and messages[-1]['role'] == 'assistant':
+                messages[-1].setdefault('tool_calls', []).append(tc)
+                if not messages[-1].get('content'):
+                    messages[-1]['content'] = None
+            else:
+                messages.append({'role': 'assistant', 'content': None, 'tool_calls': [tc]})
+            i += 1
+            continue
+
+        # function_call_output（工具结果）
+        if item_type == 'function_call_output':
+            output = item.get('output', '')
+            if not isinstance(output, str):
+                output = json.dumps(output, ensure_ascii=False)
+            messages.append({
+                'role': 'tool',
+                'tool_call_id': item.get('call_id', ''),
+                'content': output,
+            })
+            i += 1
+            continue
+
+        if role:
+            messages.append({'role': role, 'content': str(item.get('content', ''))})
+        i += 1
+
+
+def _collect_function_calls(items, start):
+    """收集紧随 assistant message 之后的连续 function_call 项"""
+    tool_calls = []
+    j = start
+    while j < len(items):
+        nxt = items[j]
+        if isinstance(nxt, dict) and nxt.get('type') == 'function_call':
+            tool_calls.append({
+                'id': nxt.get('call_id') or gen_id('call_'),
+                'type': 'function',
+                'function': {
+                    'name': nxt.get('name', ''),
+                    'arguments': nxt.get('arguments', '{}'),
+                },
+            })
+            j += 1
+        else:
+            break
+    return tool_calls, j - start
+
+
+def _extract_text(content):
+    """从 content 中提取纯文本"""
+    if isinstance(content, str):
+        return content
+    if not isinstance(content, list):
+        return str(content) if content else ''
+    texts = []
+    for part in content:
+        if isinstance(part, str):
+            texts.append(part)
+        elif isinstance(part, dict):
+            t = part.get('type', '')
+            if t in ('output_text', 'input_text', 'text'):
+                texts.append(part.get('text', ''))
+            elif t == 'refusal':
+                texts.append(part.get('refusal', ''))
+    return '\n'.join(texts) if texts else ''
+
+
+def _convert_tools(tools):
+    """将 Responses tools 转为 CC tools 格式"""
+    result = []
+    for t in tools:
+        if t.get('type') != 'function':
+            continue
+        if 'function' in t:
+            result.append(t)
+        else:
+            result.append({
+                'type': 'function',
+                'function': {
+                    'name': t.get('name', ''),
+                    'description': t.get('description', ''),
+                    'parameters': t.get('parameters', {'type': 'object', 'properties': {}}),
+                },
+            })
+    return result
diff --git a/app.py b/app.py
new file mode 100644
index 0000000..5a5137d
--- /dev/null
+++ b/app.py
@@ -0,0 +1,70 @@
+"""Flask 应用工厂
+
+创建并配置 Flask 应用：
+  - 注册所有路由蓝图
+  - 设置 JSON 错误处理器（避免返回 HTML）
+  - 配置全局鉴权中间件
+"""
+
+import logging
+
+from flask import Flask, jsonify, request
+from flask_cors import CORS
+
+import settings
+from config import Config
+from routes import register_routes
+
+logger = logging.getLogger(__name__)
+
+
+def create_app():
+    app = Flask(__name__)
+    CORS(app)
+    settings.load()
+
+    # ─── JSON 错误处理器 ──────────────────────────
+
+    @app.errorhandler(404)
+    def not_found(e):
+        return jsonify({'error': {'message': '未找到', 'type': 'not_found'}}), 404
+
+    @app.errorhandler(405)
+    def method_not_allowed(e):
+        return jsonify({'error': {'message': '方法不允许', 'type': 'method_not_allowed'}}), 405
+
+    @app.errorhandler(500)
+    def internal_error(e):
+        return jsonify({'error': {'message': '服务器内部错误', 'type': 'server_error'}}), 500
+
+    # ─── 全局鉴权中间件 ──────────────────────────
+
+    @app.before_request
+    def check_access():
+        if not Config.ACCESS_API_KEY:
+            return
+
+        # 无需鉴权的路径
+        skip = ('/health', '/admin', '/static/', '/api/admin')
+        if any(request.path == p or request.path.startswith(p) for p in skip):
+            return
+
+        auth = request.headers.get('Authorization', '')
+        token = auth[7:] if auth.startswith('Bearer ') else request.headers.get('x-api-key', '')
+        if token != Config.ACCESS_API_KEY:
+            logger.warning(f'鉴权拒绝: {request.path}')
+            return jsonify({
+                'error': {'message': 'API 密钥无效', 'type': 'authentication_error'}
+            }), 401
+
+    # ─── 健康检查 ────────────────────────────────
+
+    @app.route('/health', methods=['GET'])
+    def health():
+        return jsonify({'status': 'ok', 'target': settings.get_url()})
+
+    # ─── 注册路由蓝图 ────────────────────────────
+
+    register_routes(app)
+
+    return app
diff --git a/compose.yml b/compose.yml
new file mode 100644
index 0000000..944b836
--- /dev/null
+++ b/compose.yml
@@ -0,0 +1,16 @@
+services:
+  api2cursor:
+    build: .
+    ports:
+      - "${PROXY_PORT:-3029}:${PROXY_PORT:-3029}"
+    env_file:
+      - .env
+    volumes:
+      - ./data:/app/data
+    restart: unless-stopped
+    healthcheck:
+      test: ["CMD", "wget", "--quiet", "--tries=1", "--spider", "http://localhost:${PROXY_PORT:-3029}/health"]
+      interval: 30s
+      timeout: 5s
+      retries: 3
+      start_period: 10s
diff --git a/config.py b/config.py
new file mode 100644
index 0000000..d84b134
--- /dev/null
+++ b/config.py
@@ -0,0 +1,18 @@
+"""环境变量配置"""
+
+import os
+
+
+class Config:
+    # 上游 API 地址
+    PROXY_TARGET_URL = os.getenv('PROXY_TARGET_URL', 'https://api.anthropic.com')
+    # 上游 API 密钥
+    PROXY_API_KEY = os.getenv('PROXY_API_KEY', '')
+    # 服务监听端口
+    PROXY_PORT = int(os.getenv('PROXY_PORT', '3029'))
+    # 请求超时时间（秒）
+    API_TIMEOUT = int(os.getenv('API_TIMEOUT', '300'))
+    # 访问鉴权密钥，留空则不启用鉴权
+    ACCESS_API_KEY = os.getenv('ACCESS_API_KEY', '')
+    # 调试模式：开启后输出详细的请求/响应日志
+    DEBUG = os.getenv('DEBUG', '').lower() in ('1', 'true', 'yes', 'on')
diff --git a/requirements.txt b/requirements.txt
new file mode 100644
index 0000000..376e481
--- /dev/null
+++ b/requirements.txt
@@ -0,0 +1,5 @@
+flask
+flask-cors
+requests
+python-dotenv
+waitress
diff --git a/routes/__init__.py b/routes/__init__.py
new file mode 100644
index 0000000..e7f2b3c
--- /dev/null
+++ b/routes/__init__.py
@@ -0,0 +1,14 @@
+"""路由注册"""
+
+
+def register_routes(app):
+    """将所有路由蓝图注册到 Flask 应用"""
+    from routes.chat import bp as chat_bp
+    from routes.responses import bp as responses_bp
+    from routes.messages import bp as messages_bp
+    from routes.admin import bp as admin_bp
+
+    app.register_blueprint(chat_bp)
+    app.register_blueprint(responses_bp)
+    app.register_blueprint(messages_bp)
+    app.register_blueprint(admin_bp)
diff --git a/routes/admin.py b/routes/admin.py
new file mode 100644
index 0000000..eadd51e
--- /dev/null
+++ b/routes/admin.py
@@ -0,0 +1,195 @@
+"""路由: 管理面板
+
+提供 Web 管理界面和 API：
+  - /admin         — 管理面板页面
+  - /v1/models     — 模型列表（供 Cursor 查询）
+  - /api/admin/*   — 登录验证、全局设置 CRUD、模型映射 CRUD
+"""
+
+import os
+import logging
+
+from flask import Blueprint, request, jsonify, send_from_directory
+
+import settings
+from config import Config
+
+logger = logging.getLogger(__name__)
+
+_STATIC_DIR = os.path.join(os.path.dirname(os.path.dirname(os.path.abspath(__file__))), 'static')
+
+bp = Blueprint('admin', __name__)
+
+
+# ─── 静态页面 ─────────────────────────────────────
+
+
+@bp.route('/admin')
+@bp.route('/admin/')
+def admin_page():
+    return send_from_directory(_STATIC_DIR, 'admin.html')
+
+
+@bp.route('/static/<path:filename>')
+def static_files(filename):
+    return send_from_directory(_STATIC_DIR, filename)
+
+
+# ─── 模型列表 ─────────────────────────────────────
+
+
+@bp.route('/v1/models', methods=['GET'])
+def list_models():
+    mappings = settings.get().get('model_mappings', {})
+    models = [{
+        'id': name,
+        'object': 'model',
+        'owned_by': info.get('backend', 'custom'),
+    } for name, info in mappings.items()]
+
+    if not models:
+        models.append({
+            'id': 'claude-sonnet-4-5-20250929',
+            'object': 'model',
+            'owned_by': 'anthropic',
+        })
+    return jsonify({'object': 'list', 'data': models})
+
+
+# ─── 登录验证 ─────────────────────────────────────
+
+
+@bp.route('/api/admin/login', methods=['POST'])
+def admin_login():
+    data = request.get_json(force=True)
+    if not Config.ACCESS_API_KEY:
+        return jsonify({'ok': True, 'message': '未配置鉴权'})
+    if data.get('key', '') == Config.ACCESS_API_KEY:
+        return jsonify({'ok': True})
+    return jsonify({'ok': False, 'message': '密钥错误'}), 401
+
+
+# ─── 全局设置 ─────────────────────────────────────
+
+
+@bp.route('/api/admin/settings', methods=['GET'])
+def get_settings():
+    err = _check_auth()
+    if err:
+        return err
+    s = settings.get()
+    return jsonify({
+        'proxy_target_url': s.get('proxy_target_url', ''),
+        'proxy_api_key': s.get('proxy_api_key', ''),
+        'env_target_url': Config.PROXY_TARGET_URL,
+        'env_api_key': '***' if Config.PROXY_API_KEY else '',
+    })
+
+
+@bp.route('/api/admin/settings', methods=['PUT'])
+def update_settings():
+    err = _check_auth()
+    if err:
+        return err
+    data = request.get_json(force=True)
+    s = settings.get()
+    for key in ('proxy_target_url', 'proxy_api_key'):
+        if key in data:
+            s[key] = data[key]
+    return _save_and_respond(s, '全局设置已更新')
+
+
+# ─── 模型映射 CRUD ────────────────────────────────
+
+
+@bp.route('/api/admin/mappings', methods=['GET'])
+def list_mappings():
+    err = _check_auth()
+    if err:
+        return err
+    return jsonify(settings.get().get('model_mappings', {}))
+
+
+@bp.route('/api/admin/mappings', methods=['POST'])
+def add_mapping():
+    err = _check_auth()
+    if err:
+        return err
+    data = request.get_json(force=True)
+    name = data.get('name', '').strip()
+    if not name:
+        return jsonify({'error': '名称不能为空'}), 400
+
+    s = settings.get()
+    mappings = s.setdefault('model_mappings', {})
+    mappings[name] = {
+        'upstream_model': data.get('upstream_model', name),
+        'backend': data.get('backend', 'auto'),
+        'target_url': data.get('target_url', ''),
+        'api_key': data.get('api_key', ''),
+    }
+    return _save_and_respond(s, f'映射已添加: {name}')
+
+
+@bp.route('/api/admin/mappings/<path:name>', methods=['PUT'])
+def update_mapping(name):
+    err = _check_auth()
+    if err:
+        return err
+    data = request.get_json(force=True)
+    s = settings.get()
+    mappings = s.get('model_mappings', {})
+    if name not in mappings:
+        return jsonify({'error': '映射不存在'}), 404
+
+    new_name = data.get('name', name).strip()
+    entry = {
+        'upstream_model': data.get('upstream_model', name),
+        'backend': data.get('backend', 'auto'),
+        'target_url': data.get('target_url', ''),
+        'api_key': data.get('api_key', ''),
+    }
+    if new_name != name:
+        del mappings[name]
+    mappings[new_name] = entry
+    s['model_mappings'] = mappings
+    return _save_and_respond(s, f'映射已更新: {name} → {new_name}')
+
+
+@bp.route('/api/admin/mappings/<path:name>', methods=['DELETE'])
+def delete_mapping(name):
+    err = _check_auth()
+    if err:
+        return err
+    s = settings.get()
+    mappings = s.get('model_mappings', {})
+    if name in mappings:
+        del mappings[name]
+        s['model_mappings'] = mappings
+        return _save_and_respond(s, f'映射已删除: {name}')
+    return jsonify({'ok': True})
+
+
+# ─── 内部辅助 ─────────────────────────────────────
+
+
+def _check_auth():
+    """Admin API 鉴权，返回 None 表示通过"""
+    if not Config.ACCESS_API_KEY:
+        return None
+    auth = request.headers.get('Authorization', '')
+    token = auth[7:] if auth.startswith('Bearer ') else request.headers.get('x-api-key', '')
+    if token != Config.ACCESS_API_KEY:
+        return jsonify({'error': '未授权'}), 401
+    return None
+
+
+def _save_and_respond(data, log_msg):
+    """保存配置并返回响应"""
+    try:
+        settings.save(data)
+    except OSError as e:
+        logger.error(f'保存失败: {e}')
+        return jsonify({'error': {'message': f'保存失败: {e}', 'type': 'save_error'}}), 500
+    logger.info(log_msg)
+    return jsonify({'ok': True})
diff --git a/routes/chat.py b/routes/chat.py
new file mode 100644
index 0000000..8e106da
--- /dev/null
+++ b/routes/chat.py
@@ -0,0 +1,216 @@
+"""路由: /v1/chat/completions
+
+处理 Cursor 发来的 OpenAI Chat Completions 格式请求。
+根据模型映射的 backend 字段分发到 OpenAI 或 Anthropic 后端。
+"""
+
+import json
+import logging
+
+from flask import Blueprint, request, jsonify
+
+import settings
+from config import Config
+from adapters.openai_fixer import normalize_request, fix_response, fix_stream_chunk
+from adapters.openai_anthropic import (
+    cc_to_messages_request, messages_to_cc_response, AnthropicStreamConverter,
+)
+from adapters.responses_adapter import responses_to_cc
+from utils.http import (
+    build_openai_headers, build_anthropic_headers,
+    forward_request, sse_response,
+    iter_openai_sse, iter_anthropic_sse,
+)
+from utils.think_tag import ThinkTagExtractor
+logger = logging.getLogger(__name__)
+
+
+def _dbg(msg):
+    """DEBUG 模式下输出详细日志"""
+    if Config.DEBUG:
+        logger.info(f'[调试] {msg}')
+
+bp = Blueprint('chat', __name__)
+
+
+@bp.route('/v1/chat/completions', methods=['POST'])
+def chat_completions():
+    payload = request.get_json(force=True)
+    is_stream = payload.get('stream', False)
+    # 保留 Cursor 发送的原始模型名，响应时需要回填
+    cursor_model = payload.get('model', 'unknown')
+    msg_count = len(payload.get('messages', []))
+
+    # 容错：Responses 格式误入 CC 端点
+    if msg_count == 0 and 'input' in payload:
+        logger.info('检测到 Responses 格式（有 input 无 messages），自动转换')
+        payload = responses_to_cc(payload)
+        msg_count = len(payload.get('messages', []))
+    elif msg_count == 0:
+        logger.warning(f'messages 为空, payload keys: {list(payload.keys())}')
+
+    mapping = settings.resolve_model(cursor_model)
+    backend = mapping['backend']
+    upstream = mapping['upstream_model']
+    url_base = mapping['target_url']
+    api_key = mapping['api_key']
+
+    logger.info(
+        f'[CC] {cursor_model} → {upstream} '
+        f'后端={backend} 流式={is_stream} 消息数={msg_count}'
+    )
+    _log_messages(payload)
+
+    if backend == 'openai':
+        return _via_openai(payload, upstream, url_base, api_key, is_stream, cursor_model)
+    else:
+        return _via_anthropic(payload, upstream, url_base, api_key, is_stream, cursor_model)
+
+
+# ─── OpenAI 后端 ──────────────────────────────────
+
+
+def _via_openai(payload, upstream, url_base, api_key, is_stream, cursor_model):
+    """通过 OpenAI 兼容后端转发"""
+    _dbg(f'Cursor 原始请求 keys={list(payload.keys())} '
+         f'其他字段={json.dumps({k: v for k, v in payload.items() if k != "messages"}, ensure_ascii=False, default=str)[:500]}')
+
+    payload = normalize_request(payload, upstream)
+    _dbg(f'normalize 后 model={payload.get("model")} tools数={len(payload.get("tools", []))}')
+
+    headers = build_openai_headers(api_key)
+    url = f'{url_base.rstrip("/")}/v1/chat/completions'
+
+    if not is_stream:
+        payload['stream'] = False
+        resp, err = forward_request(url, headers, payload)
+        if err:
+            return err
+        raw = resp.json()
+        _dbg(f'上游原始响应={json.dumps(raw, ensure_ascii=False, default=str)[:1000]}')
+        data = fix_response(raw)
+        data['model'] = cursor_model
+        _dbg(f'修复后响应={json.dumps(data, ensure_ascii=False, default=str)[:1000]}')
+        usage = data.get('usage', {})
+        logger.info(
+            f'[CC] 完成 prompt={usage.get("prompt_tokens", 0)} '
+            f'completion={usage.get("completion_tokens", 0)}'
+        )
+        return jsonify(data)
+
+    # 流式处理
+    payload['stream'] = True
+    _n = [0]
+
+    def generate():
+        resp, err = forward_request(url, headers, payload, stream=True)
+        if err:
+            yield f'data: {json.dumps({"error": {"message": err, "type": "upstream_error"}})}\n\n'
+            return
+
+        think_ext = ThinkTagExtractor()
+
+        for chunk in iter_openai_sse(resp):
+            if chunk is None:  # [DONE]
+                _dbg(f'流结束，共 {_n[0]} 个 chunk')
+                yield 'data: [DONE]\n\n'
+                return
+
+            if _n[0] < 10:
+                _dbg(f'上游原始 chunk#{_n[0]}={json.dumps(chunk, ensure_ascii=False, default=str)[:500]}')
+
+            chunk = fix_stream_chunk(chunk)
+            chunk['model'] = cursor_model
+
+            for out in think_ext.process_chunk(chunk):
+                if _n[0] < 10:
+                    _dbg(f'发给Cursor chunk#{_n[0]}={json.dumps(out, ensure_ascii=False, default=str)[:500]}')
+                yield f'data: {json.dumps(out)}\n\n'
+
+            _n[0] += 1
+
+    return sse_response(generate())
+
+
+# ─── Anthropic 后端 ───────────────────────────────
+
+
+def _via_anthropic(payload, upstream, url_base, api_key, is_stream, cursor_model):
+    """通过 Anthropic 后端转发（CC → Messages → CC）"""
+    payload['model'] = upstream
+    anthropic_payload = cc_to_messages_request(payload)
+    _dbg(f'CC→Messages 转换后 keys={list(anthropic_payload.keys())} '
+         f'messages数={len(anthropic_payload.get("messages", []))}')
+
+    headers = build_anthropic_headers(api_key)
+    url = f'{url_base.rstrip("/")}/v1/messages'
+
+    if not is_stream:
+        anthropic_payload['stream'] = False
+        resp, err = forward_request(url, headers, anthropic_payload)
+        if err:
+            return err
+        raw = resp.json()
+        _dbg(f'上游原始响应={json.dumps(raw, ensure_ascii=False, default=str)[:1000]}')
+        data = messages_to_cc_response(raw)
+        data['model'] = cursor_model
+        _dbg(f'Messages→CC 转换后={json.dumps(data, ensure_ascii=False, default=str)[:1000]}')
+        usage = data.get('usage', {})
+        logger.info(
+            f'[CC] 完成 prompt={usage.get("prompt_tokens", 0)} '
+            f'completion={usage.get("completion_tokens", 0)}'
+        )
+        return jsonify(data)
+
+    # 流式处理
+    anthropic_payload['stream'] = True
+    converter = AnthropicStreamConverter()
+    _n = [0]
+
+    def generate():
+        resp, err = forward_request(url, headers, anthropic_payload, stream=True)
+        if err:
+            yield f'data: {json.dumps({"error": {"message": err, "type": "upstream_error"}})}\n\n'
+            return
+
+        for event_type, event_data in iter_anthropic_sse(resp):
+            if _n[0] < 10:
+                _dbg(f'上游事件#{_n[0]} {event_type}={json.dumps(event_data, ensure_ascii=False, default=str)[:500]}')
+
+            for chunk_str in converter.process_event(event_type, event_data):
+                try:
+                    chunk_obj = json.loads(chunk_str)
+                    chunk_obj['model'] = cursor_model
+                    chunk_str = json.dumps(chunk_obj)
+                except (json.JSONDecodeError, TypeError):
+                    pass
+                if _n[0] < 10:
+                    _dbg(f'发给Cursor chunk#{_n[0]}={chunk_str[:500]}')
+                yield f'data: {chunk_str}\n\n'
+
+            _n[0] += 1
+
+        _dbg(f'流结束，共 {_n[0]} 个事件')
+        yield 'data: [DONE]\n\n'
+
+    return sse_response(generate())
+
+
+def _log_messages(payload):
+    """记录请求中的消息摘要"""
+    for i, msg in enumerate(payload.get('messages', [])):
+        role = msg.get('role', '?')
+        content = msg.get('content')
+        extra = ''
+        if 'tool_calls' in msg:
+            extra += f' tool_calls={len(msg["tool_calls"])}'
+        if msg.get('tool_call_id'):
+            extra += f' tool_call_id={msg["tool_call_id"]}'
+
+        if isinstance(content, list):
+            info = f'list[{len(content)}]'
+        elif isinstance(content, str):
+            info = f'str[{len(content)}]'
+        else:
+            info = type(content).__name__
+        logger.info(f'  消息[{i}] {role} {info}{extra}')
diff --git a/routes/messages.py b/routes/messages.py
new file mode 100644
index 0000000..0a4d89e
--- /dev/null
+++ b/routes/messages.py
@@ -0,0 +1,147 @@
+"""路由: /v1/messages
+
+Anthropic Messages API 透传。当 Cursor 直接发送 Anthropic 格式请求时，
+直接转发到上游并原样返回。处理非标准的 reasoning_content 字段，
+将其注入为标准的 thinking blocks。
+"""
+
+import json
+import logging
+
+import requests as req_lib
+from flask import Blueprint, request, jsonify
+
+import settings
+from config import Config
+from utils.http import build_anthropic_headers, forward_request, sse_response
+
+logger = logging.getLogger(__name__)
+
+bp = Blueprint('messages', __name__)
+
+
+@bp.route('/v1/messages', methods=['POST'])
+def messages_passthrough():
+    payload = request.get_json(force=True)
+    model = payload.get('model', 'unknown')
+    is_stream = payload.get('stream', False)
+
+    logger.info(f'[透传] model={model} 流式={is_stream}')
+
+    url_base = settings.get_url()
+    api_key = settings.get_key()
+    headers = build_anthropic_headers(api_key)
+    url = f'{url_base.rstrip("/")}/v1/messages'
+
+    if not is_stream:
+        resp, err = forward_request(url, headers, payload)
+        if err:
+            return err
+        data = resp.json()
+        _inject_thinking(data)
+        return jsonify(data)
+
+    # 流式透传
+    def generate():
+        try:
+            resp = req_lib.post(
+                url, headers=headers, json=payload,
+                timeout=Config.API_TIMEOUT, stream=True,
+            )
+            if resp.status_code != 200:
+                body = resp.content.decode('utf-8', errors='replace')
+                logger.warning(f'上游返回 {resp.status_code}: {body[:300]}')
+                yield f'data: {json.dumps({"error": {"message": body, "type": "upstream_error"}})}\n\n'
+                return
+
+            yield from _process_stream(resp)
+        except req_lib.RequestException as e:
+            logger.error(f'请求上游失败: {e}')
+            yield f'data: {json.dumps({"error": {"message": str(e), "type": "proxy_error"}})}\n\n'
+
+    return sse_response(generate())
+
+
+# ─── 内部辅助 ─────────────────────────────────────
+
+
+def _inject_thinking(data):
+    """将非标准 reasoning_content 字段注入为 Anthropic thinking block"""
+    rc = data.pop('reasoning_content', None) or data.pop('reasoningContent', None)
+    if not rc:
+        return
+
+    content = data.get('content')
+    if not isinstance(content, list):
+        content = []
+
+    # 避免重复注入
+    if any(isinstance(b, dict) and b.get('type') == 'thinking' for b in content):
+        return
+
+    content.insert(0, {'type': 'thinking', 'thinking': rc})
+    data['content'] = content
+    logger.info(f'已注入 thinking block ({len(rc)} 字符)')
+
+
+def _process_stream(resp):
+    """处理 /v1/messages 流式响应，检测并注入 thinking 事件"""
+    reasoning_buf = ''
+    injected = False
+
+    for line in resp.iter_lines():
+        if not line:
+            continue
+        decoded = line.decode('utf-8', errors='replace')
+
+        if not decoded.startswith('data:'):
+            yield decoded + '\n\n'
+            continue
+
+        data_str = decoded[5:].strip()
+        if not data_str:
+            yield decoded + '\n\n'
+            continue
+
+        try:
+            event_data = json.loads(data_str)
+        except json.JSONDecodeError:
+            yield decoded + '\n\n'
+            continue
+
+        modified = False
+
+        # 提取 reasoning_content
+        for container_key in ('message', 'delta'):
+            container = event_data.get(container_key)
+            if not container:
+                continue
+            rc = container.pop('reasoning_content', None) or container.pop('reasoningContent', None)
+            if rc:
+                reasoning_buf += rc
+                modified = True
+
+        # 在首个 text_delta 前注入 thinking blocks
+        if reasoning_buf and not injected:
+            if event_data.get('delta', {}).get('type') == 'text_delta':
+                injected = True
+                yield from _emit_thinking_blocks(reasoning_buf)
+                reasoning_buf = ''
+
+        yield f'data: {json.dumps(event_data)}\n\n' if modified else decoded + '\n\n'
+
+
+def _emit_thinking_blocks(text):
+    """生成 thinking block 的 SSE 事件序列"""
+    yield (
+        f'event: content_block_start\n'
+        f'data: {json.dumps({"type": "content_block_start", "index": 0, "content_block": {"type": "thinking", "thinking": ""}})}\n\n'
+    )
+    yield (
+        f'event: content_block_delta\n'
+        f'data: {json.dumps({"type": "content_block_delta", "index": 0, "delta": {"type": "thinking_delta", "thinking": text}})}\n\n'
+    )
+    yield (
+        f'event: content_block_stop\n'
+        f'data: {json.dumps({"type": "content_block_stop", "index": 0})}\n\n'
+    )
diff --git a/routes/responses.py b/routes/responses.py
new file mode 100644
index 0000000..12cbd2a
--- /dev/null
+++ b/routes/responses.py
@@ -0,0 +1,126 @@
+"""路由: /v1/responses
+
+处理 Cursor 对 GPT/Claude-Opus 等模型发出的 Responses API 格式请求。
+转换为 CC 格式后分发到对应后端，响应再转回 Responses 格式。
+"""
+
+import json
+import logging
+
+from flask import Blueprint, request, jsonify
+
+import settings
+from adapters.responses_adapter import responses_to_cc, cc_to_responses, ResponsesStreamConverter
+from adapters.openai_fixer import normalize_request, fix_response, fix_stream_chunk
+from adapters.openai_anthropic import cc_to_messages_request, messages_to_cc_response
+from utils.http import (
+    build_openai_headers, build_anthropic_headers,
+    forward_request, sse_response,
+    iter_openai_sse, iter_anthropic_sse,
+)
+from utils.think_tag import ThinkTagExtractor
+
+logger = logging.getLogger(__name__)
+
+bp = Blueprint('responses', __name__)
+
+
+@bp.route('/v1/responses', methods=['POST'])
+def responses_endpoint():
+    payload = request.get_json(force=True)
+    model = payload.get('model', 'unknown')
+    is_stream = payload.get('stream', False)
+
+    mapping = settings.resolve_model(model)
+    backend = mapping['backend']
+    upstream = mapping['upstream_model']
+    url_base = mapping['target_url']
+    api_key = mapping['api_key']
+
+    logger.info(f'[Responses] {model} → {upstream} 后端={backend} 流式={is_stream}')
+
+    # Responses → CC
+    cc_payload = responses_to_cc(payload)
+    cc_payload['model'] = upstream
+
+    if backend == 'openai':
+        return _via_openai(cc_payload, url_base, api_key, is_stream, model)
+    else:
+        return _via_anthropic(cc_payload, url_base, api_key, is_stream, model)
+
+
+# ─── OpenAI 后端 ──────────────────────────────────
+
+
+def _via_openai(cc_payload, url_base, api_key, is_stream, display_model):
+    """通过 OpenAI 后端处理"""
+    cc_payload = normalize_request(cc_payload)
+    headers = build_openai_headers(api_key)
+    url = f'{url_base.rstrip("/")}/v1/chat/completions'
+
+    if not is_stream:
+        cc_payload['stream'] = False
+        resp, err = forward_request(url, headers, cc_payload)
+        if err:
+            return err
+        return jsonify(cc_to_responses(fix_response(resp.json()), display_model))
+
+    # 流式处理
+    cc_payload['stream'] = True
+    converter = ResponsesStreamConverter(model=display_model)
+
+    def generate():
+        yield from converter.start_events()
+
+        resp, err = forward_request(url, headers, cc_payload, stream=True)
+        if err:
+            yield f'event: error\ndata: {json.dumps({"error": err})}\n\n'
+            return
+
+        think_ext = ThinkTagExtractor()
+        for chunk in iter_openai_sse(resp):
+            if chunk is None:
+                yield from converter.finalize()
+                return
+            chunk = fix_stream_chunk(chunk)
+            for out in think_ext.process_chunk(chunk):
+                yield from converter.process_cc_chunk(out)
+
+    return sse_response(generate())
+
+
+# ─── Anthropic 后端 ───────────────────────────────
+
+
+def _via_anthropic(cc_payload, url_base, api_key, is_stream, display_model):
+    """通过 Anthropic 后端处理"""
+    anthropic_payload = cc_to_messages_request(cc_payload)
+    headers = build_anthropic_headers(api_key)
+    url = f'{url_base.rstrip("/")}/v1/messages'
+
+    if not is_stream:
+        anthropic_payload['stream'] = False
+        resp, err = forward_request(url, headers, anthropic_payload)
+        if err:
+            return err
+        cc_data = messages_to_cc_response(resp.json())
+        return jsonify(cc_to_responses(cc_data, display_model))
+
+    # 流式处理：Anthropic SSE → Responses SSE（跳过 CC 中间态）
+    anthropic_payload['stream'] = True
+    converter = ResponsesStreamConverter(model=display_model)
+
+    def generate():
+        yield from converter.start_events()
+
+        resp, err = forward_request(url, headers, anthropic_payload, stream=True)
+        if err:
+            yield f'event: error\ndata: {json.dumps({"error": err})}\n\n'
+            return
+
+        for event_type, event_data in iter_anthropic_sse(resp):
+            yield from converter.process_anthropic_event(event_type, event_data)
+
+        yield from converter.finalize()
+
+    return sse_response(generate())
diff --git a/settings.py b/settings.py
new file mode 100644
index 0000000..d7d6d61
--- /dev/null
+++ b/settings.py
@@ -0,0 +1,97 @@
+"""持久化配置管理
+
+使用 data/settings.json 存储可通过管理面板修改的设置：
+  - proxy_target_url / proxy_api_key: 可覆盖环境变量的全局配置
+  - model_mappings: Cursor 模型名 → {upstream_model, backend, target_url, api_key}
+"""
+
+import json
+import os
+import threading
+
+from config import Config
+
+# 数据目录放在项目根目录下，便于 Docker 卷挂载
+_ROOT_DIR = os.path.dirname(os.path.abspath(__file__))
+DATA_DIR = os.path.join(_ROOT_DIR, 'data')
+SETTINGS_FILE = os.path.join(DATA_DIR, 'settings.json')
+
+_lock = threading.Lock()
+_cache = None
+
+_DEFAULTS = {
+    'proxy_target_url': '',
+    'proxy_api_key': '',
+    'model_mappings': {},
+}
+
+
+def load():
+    """从文件加载配置"""
+    global _cache
+    with _lock:
+        if os.path.exists(SETTINGS_FILE):
+            try:
+                with open(SETTINGS_FILE, 'r', encoding='utf-8') as f:
+                    _cache = {**_DEFAULTS, **json.load(f)}
+            except (json.JSONDecodeError, OSError):
+                _cache = dict(_DEFAULTS)
+        else:
+            _cache = dict(_DEFAULTS)
+    return dict(_cache)
+
+
+def save(data):
+    """保存配置到文件"""
+    global _cache
+    with _lock:
+        os.makedirs(DATA_DIR, exist_ok=True)
+        _cache = {**_DEFAULTS, **data}
+        with open(SETTINGS_FILE, 'w', encoding='utf-8') as f:
+            json.dump(_cache, f, ensure_ascii=False, indent=2)
+
+
+def get():
+    """获取当前配置（优先使用缓存）"""
+    if _cache is None:
+        return load()
+    return dict(_cache)
+
+
+def get_url():
+    """获取生效的上游 URL：配置文件优先，环境变量兜底"""
+    return get().get('proxy_target_url') or Config.PROXY_TARGET_URL
+
+
+def get_key():
+    """获取生效的 API 密钥：配置文件优先，环境变量兜底"""
+    return get().get('proxy_api_key') or Config.PROXY_API_KEY
+
+
+def resolve_model(model_name):
+    """解析模型映射，返回 {upstream_model, backend, target_url, api_key}"""
+    settings = get()
+    mappings = settings.get('model_mappings', {})
+    base_url, base_key = get_url(), get_key()
+
+    if model_name in mappings:
+        m = mappings[model_name]
+        return {
+            'upstream_model': m.get('upstream_model') or model_name,
+            'backend': m.get('backend') or _auto_detect(model_name),
+            'target_url': m.get('target_url') or base_url,
+            'api_key': m.get('api_key') or base_key,
+        }
+
+    return {
+        'upstream_model': model_name,
+        'backend': _auto_detect(model_name),
+        'target_url': base_url,
+        'api_key': base_key,
+    }
+
+
+def _auto_detect(name):
+    """根据模型名自动判断后端类型"""
+    lower = (name or '').lower()
+    return 'anthropic' if ('claude' in lower or 'anthropic' in lower) else 'openai'
diff --git a/start.py b/start.py
new file mode 100644
index 0000000..96e621e
--- /dev/null
+++ b/start.py
@@ -0,0 +1,38 @@
+"""启动入口
+
+用法: python start.py
+"""
+
+import logging
+
+from dotenv import load_dotenv
+
+load_dotenv()
+
+logging.basicConfig(
+    level=logging.INFO,
+    format='%(asctime)s [%(name)s] %(levelname)s: %(message)s',
+)
+
+from config import Config
+from app import create_app
+
+
+def main():
+    app = create_app()
+    print(f'代理服务启动于 0.0.0.0:{Config.PROXY_PORT}')
+    print(f'上游地址: {Config.PROXY_TARGET_URL}')
+    print(f'管理面板: http://localhost:{Config.PROXY_PORT}/admin')
+
+    from waitress import serve
+    serve(
+        app,
+        host='0.0.0.0',
+        port=Config.PROXY_PORT,
+        channel_timeout=Config.API_TIMEOUT,
+        send_bytes=1,
+    )
+
+
+if __name__ == '__main__':
+    main()
diff --git a/static/admin.css b/static/admin.css
new file mode 100644
index 0000000..1844c47
--- /dev/null
+++ b/static/admin.css
@@ -0,0 +1,76 @@
+*,*::before,*::after{box-sizing:border-box;margin:0;padding:0}
+:root{
+  --bg:#0b1120;--surface:#151d2e;--card:#1a2332;--input:#212d3f;
+  --border:#2a3a50;--text:#e2e8f0;--muted:#8899ab;--primary:#3b82f6;
+  --primary-hover:#2563eb;--green:#22c55e;--red:#ef4444;--yellow:#eab308;
+  --radius:10px;
+}
+body{font-family:-apple-system,BlinkMacSystemFont,'Segoe UI','PingFang SC','Microsoft YaHei',sans-serif;background:var(--bg);color:var(--text);min-height:100vh;line-height:1.6}
+input,select,button,textarea{font-family:inherit;font-size:inherit}
+a{color:var(--primary);text-decoration:none}
+code{background:var(--input);padding:1px 5px;border-radius:4px;font-size:12px;font-family:Consolas,Monaco,monospace}
+.container{max-width:960px;margin:0 auto;padding:0 20px}
+
+#login{display:flex;align-items:center;justify-content:center;min-height:100vh;background:linear-gradient(145deg,#0b1120 0%,#121a2e 50%,#0b1120 100%)}
+.login-card{background:var(--card);border:1px solid var(--border);border-radius:16px;padding:40px;width:380px;box-shadow:0 20px 60px rgba(0,0,0,.4)}
+.login-card h1{font-size:22px;font-weight:700;margin-bottom:6px;text-align:center}
+.login-card p{color:var(--muted);font-size:13px;text-align:center;margin-bottom:28px}
+
+.field{margin-bottom:16px}
+.field label{display:block;font-size:13px;color:var(--muted);margin-bottom:6px;font-weight:500}
+.input-wrap{position:relative}
+.input-wrap input,.input-wrap select{width:100%;background:var(--input);border:1px solid var(--border);border-radius:8px;padding:10px 14px;color:var(--text);font-size:14px;outline:none;transition:border-color .2s}
+.input-wrap input:focus,.input-wrap select:focus{border-color:var(--primary)}
+.input-wrap .eye{position:absolute;right:10px;top:50%;transform:translateY(-50%);background:none;border:none;color:var(--muted);cursor:pointer;padding:4px;font-size:16px}
+select{cursor:pointer;appearance:none;background-image:url("data:image/svg+xml,%3Csvg xmlns='http://www.w3.org/2000/svg' width='12' height='12' fill='%238899ab'%3E%3Cpath d='M6 8L1 3h10z'/%3E%3C/svg%3E");background-repeat:no-repeat;background-position:right 12px center}
+
+.btn{display:inline-flex;align-items:center;gap:6px;padding:9px 18px;border-radius:8px;border:none;font-size:14px;font-weight:500;cursor:pointer;transition:all .15s}
+.btn-primary{background:var(--primary);color:#fff}.btn-primary:hover{background:var(--primary-hover)}
+.btn-green{background:var(--green);color:#fff}.btn-green:hover{opacity:.9}
+.btn-red{background:transparent;color:var(--red);border:1px solid var(--red)}.btn-red:hover{background:var(--red);color:#fff}
+.btn-ghost{background:transparent;color:var(--muted);border:1px solid var(--border)}.btn-ghost:hover{color:var(--text);border-color:var(--muted)}
+.btn-sm{padding:6px 12px;font-size:13px;border-radius:6px}
+.btn-block{width:100%;justify-content:center}
+.btn:disabled{opacity:.5;cursor:not-allowed}
+
+#dashboard{display:none}
+header{background:var(--surface);border-bottom:1px solid var(--border);padding:14px 0;position:sticky;top:0;z-index:50}
+header .inner{display:flex;align-items:center}
+header h1{font-size:17px;font-weight:700;flex:1}
+header .right{display:flex;align-items:center;gap:12px}
+.status{font-size:12px;padding:4px 10px;border-radius:20px;background:rgba(34,197,94,.15);color:var(--green)}
+main{padding:28px 0 60px}
+
+.card{background:var(--card);border:1px solid var(--border);border-radius:var(--radius);padding:24px;margin-bottom:24px}
+.card-header{display:flex;align-items:center;justify-content:space-between;margin-bottom:20px}
+.card-header h2{font-size:16px;font-weight:600}
+.card-header .badge{font-size:12px;color:var(--muted);background:var(--input);padding:3px 10px;border-radius:12px}
+.hint{font-size:12px;color:var(--muted);margin-top:4px}
+
+.mapping-list{display:flex;flex-direction:column;gap:10px}
+.mapping-item{background:var(--surface);border:1px solid var(--border);border-radius:8px;padding:16px;transition:border-color .2s}
+.mapping-item:hover{border-color:var(--primary)}
+.mapping-top{display:flex;align-items:center;gap:12px;flex-wrap:wrap}
+.mapping-name{font-weight:600;font-size:15px;color:var(--primary)}
+.mapping-arrow{color:var(--muted);font-size:13px}
+.mapping-upstream{font-size:14px;color:var(--text)}
+.mapping-meta{display:flex;gap:8px;flex-wrap:wrap}
+.tag{font-size:11px;padding:2px 8px;border-radius:4px;font-weight:500}
+.tag-anthropic{background:rgba(249,115,22,.15);color:#fb923c}
+.tag-openai{background:rgba(16,185,129,.15);color:#34d399}
+.tag-auto{background:rgba(139,92,246,.15);color:#a78bfa}
+.tag-override{background:rgba(59,130,246,.1);color:var(--primary)}
+.mapping-actions{margin-left:auto;display:flex;gap:6px}
+.empty{text-align:center;padding:40px;color:var(--muted)}
+
+.modal-overlay{display:none;position:fixed;inset:0;background:rgba(0,0,0,.6);z-index:100;align-items:center;justify-content:center}
+.modal-overlay.active{display:flex}
+.modal{background:var(--card);border:1px solid var(--border);border-radius:14px;padding:28px;width:520px;max-width:90vw;max-height:85vh;overflow-y:auto;box-shadow:0 20px 60px rgba(0,0,0,.5)}
+.modal h3{font-size:17px;font-weight:600;margin-bottom:20px}
+.modal-footer{display:flex;justify-content:flex-end;gap:10px;margin-top:24px;padding-top:16px;border-top:1px solid var(--border)}
+
+.toast-area{position:fixed;top:20px;right:20px;z-index:200;display:flex;flex-direction:column;gap:8px}
+.toast{padding:12px 20px;border-radius:8px;font-size:14px;font-weight:500;box-shadow:0 8px 24px rgba(0,0,0,.3);animation:slideIn .3s ease}
+.toast-ok{background:#065f46;color:#a7f3d0}
+.toast-err{background:#7f1d1d;color:#fca5a5}
+@keyframes slideIn{from{transform:translateX(100px);opacity:0}to{transform:none;opacity:1}}
diff --git a/static/admin.html b/static/admin.html
new file mode 100644
index 0000000..1a9c28b
--- /dev/null
+++ b/static/admin.html
@@ -0,0 +1,127 @@
+<!DOCTYPE html>
+<html lang="zh-CN">
+<head>
+<meta charset="UTF-8">
+<meta name="viewport" content="width=device-width, initial-scale=1.0">
+<title>API 2 Cursor - 管理面板</title>
+<link rel="stylesheet" href="/static/admin.css">
+</head>
+<body>
+
+<!-- 登录 -->
+<div id="login">
+  <div class="login-card">
+    <h1>API 2 Cursor</h1>
+    <p>模型映射管理面板</p>
+    <div class="field">
+      <label>ACCESS_API_KEY</label>
+      <div class="input-wrap">
+        <input type="password" id="loginKey" placeholder="请输入访问密钥" onkeydown="if(event.key==='Enter')doLogin()">
+        <button class="eye" onclick="togglePwd('loginKey')">&#128065;</button>
+      </div>
+    </div>
+    <button class="btn btn-primary btn-block" onclick="doLogin()">登 录</button>
+  </div>
+</div>
+
+<!-- 仪表盘 -->
+<div id="dashboard">
+  <header>
+    <div class="container inner">
+      <h1>API 2 Cursor</h1>
+      <div class="right">
+        <span class="status" id="statusBadge">已连接</span>
+        <button class="btn btn-ghost btn-sm" onclick="doLogout()">退出</button>
+      </div>
+    </div>
+  </header>
+
+  <main class="container">
+    <!-- 全局设置 -->
+    <div class="card">
+      <div class="card-header">
+        <h2>全局设置</h2>
+      </div>
+      <div class="field">
+        <label>中转站地址 (Proxy Target URL)</label>
+        <div class="input-wrap"><input type="text" id="targetUrl" placeholder="https://your-relay.com"></div>
+        <div class="hint" id="envUrl"></div>
+      </div>
+      <div class="field">
+        <label>中转站 API Key</label>
+        <div class="input-wrap">
+          <input type="password" id="proxyKey" placeholder="sk-xxx 或 Bearer token">
+          <button class="eye" onclick="togglePwd('proxyKey')">&#128065;</button>
+        </div>
+        <div class="hint" id="envKey"></div>
+      </div>
+      <button class="btn btn-primary" onclick="saveSettings()">保存设置</button>
+    </div>
+
+    <!-- 模型映射 -->
+    <div class="card">
+      <div class="card-header">
+        <h2>模型映射</h2>
+        <button class="btn btn-green btn-sm" onclick="openAddModal()">+ 添加映射</button>
+      </div>
+      <div class="hint" style="margin-top:-12px;margin-bottom:16px">
+        Cursor 发送请求时会带上模型名 → 代理根据映射表找到上游实际模型名和后端协议进行转发。
+        <br>提示：建议在 Cursor 里使用 Claude 风格的模型名（如 <code>claude-sonnet-4-5-20250929</code>），这样 Cursor 会走 <code>/v1/chat/completions</code> 格式；GPT 风格的模型名会走 <code>/v1/responses</code> 格式，两种都已支持。
+      </div>
+      <div id="mappingList"></div>
+    </div>
+  </main>
+</div>
+
+<!-- 弹窗 -->
+<div class="modal-overlay" id="modal">
+  <div class="modal">
+    <h3 id="modalTitle">添加模型映射</h3>
+    <div class="field">
+      <label>Cursor 模型名 <span style="color:var(--red)">*</span></label>
+      <div class="input-wrap"><input type="text" id="mName" placeholder="例: claude-sonnet-4-5-20250929"></div>
+      <div class="hint">在 Cursor 自定义模型中添加的名称</div>
+    </div>
+    <div class="field">
+      <label>上游实际模型名 <span style="color:var(--red)">*</span></label>
+      <div class="input-wrap"><input type="text" id="mUpstream" placeholder="例: gpt-5.4 或 claude-sonnet-4-5-20250929"></div>
+      <div class="hint">发送到中转站的真实模型名称</div>
+    </div>
+    <div class="field">
+      <label>后端类型</label>
+      <div class="input-wrap">
+        <select id="mBackend">
+          <option value="auto">自动检测</option>
+          <option value="anthropic">Anthropic (/v1/messages)</option>
+          <option value="openai">OpenAI (/v1/chat/completions)</option>
+        </select>
+      </div>
+      <div class="hint">
+        <b>anthropic</b>：转换为 Anthropic Messages 格式 — 适用于中转站通过 <code>/v1/messages</code> 提供 Claude 模型<br>
+        <b>openai</b>：保持 OpenAI Chat Completions 格式 — 适用于 GPT、DeepSeek、Codex 或通过 <code>/v1/chat/completions</code> 提供所有模型的中转站<br>
+        <b>自动检测</b>：根据上游模型名判断（含 claude → anthropic，其他 → openai）
+      </div>
+    </div>
+    <div class="field">
+      <label>自定义地址 <span style="color:var(--muted)">（可选，留空使用全局设置）</span></label>
+      <div class="input-wrap"><input type="text" id="mUrl" placeholder="留空则使用全局中转站地址"></div>
+    </div>
+    <div class="field">
+      <label>自定义 API Key <span style="color:var(--muted)">（可选，留空使用全局设置）</span></label>
+      <div class="input-wrap">
+        <input type="password" id="mKey" placeholder="留空则使用全局 API Key">
+        <button class="eye" onclick="togglePwd('mKey')">&#128065;</button>
+      </div>
+    </div>
+    <div class="modal-footer">
+      <button class="btn btn-ghost" onclick="closeModal()">取消</button>
+      <button class="btn btn-primary" id="modalSaveBtn" onclick="saveMapping()">保存</button>
+    </div>
+  </div>
+</div>
+
+<div class="toast-area" id="toasts"></div>
+
+<script src="/static/admin.js"></script>
+</body>
+</html>
diff --git a/static/admin.js b/static/admin.js
new file mode 100644
index 0000000..9b7b10c
--- /dev/null
+++ b/static/admin.js
@@ -0,0 +1,247 @@
+const API = '';
+let authKey = '';
+let editingName = null;
+
+function togglePwd(id) {
+  const el = document.getElementById(id);
+  el.type = el.type === 'password' ? 'text' : 'password';
+}
+
+function toast(msg, ok = true) {
+  const area = document.getElementById('toasts');
+  const el = document.createElement('div');
+  el.className = 'toast ' + (ok ? 'toast-ok' : 'toast-err');
+  el.textContent = msg;
+  area.appendChild(el);
+  setTimeout(() => el.remove(), 3000);
+}
+
+async function api(path, opts = {}) {
+  const headers = { 'Content-Type': 'application/json' };
+  if (authKey) headers['Authorization'] = 'Bearer ' + authKey;
+  const res = await fetch(API + path, { ...opts, headers });
+  const ct = res.headers.get('content-type') || '';
+  if (!ct.includes('application/json')) {
+    const text = await res.text();
+    if (!res.ok) throw new Error('HTTP ' + res.status + ': ' + text.substring(0, 100));
+    throw new Error('服务器返回了非 JSON 响应');
+  }
+  const data = await res.json();
+  if (!res.ok) {
+    const e = data.error;
+    const msg = (typeof e === 'object' && e !== null) ? (e.message || JSON.stringify(e)) : (e || data.message || 'HTTP ' + res.status);
+    throw new Error(msg);
+  }
+  return data;
+}
+
+// ─── 登录 ───────────────────────────────────────────
+async function doLogin() {
+  const key = document.getElementById('loginKey').value.trim();
+  if (!key) { toast('请输入密钥', false); return; }
+  try {
+    const r = await api('/api/admin/login', { method: 'POST', body: JSON.stringify({ key }) });
+    if (r.ok) {
+      authKey = key;
+      sessionStorage.setItem('_ak', key);
+      document.getElementById('login').style.display = 'none';
+      document.getElementById('dashboard').style.display = 'block';
+      loadDashboard();
+    }
+  } catch (e) {
+    toast('密钥无效', false);
+  }
+}
+
+function doLogout() {
+  authKey = '';
+  sessionStorage.removeItem('_ak');
+  document.getElementById('dashboard').style.display = 'none';
+  document.getElementById('login').style.display = 'flex';
+}
+
+// ─── 仪表盘 ─────────────────────────────────────────
+async function loadDashboard() {
+  try {
+    const s = await api('/api/admin/settings');
+    document.getElementById('targetUrl').value = s.proxy_target_url || '';
+    document.getElementById('proxyKey').value = s.proxy_api_key || '';
+    document.getElementById('envUrl').textContent = s.env_target_url ? '环境变量: ' + s.env_target_url : '';
+    document.getElementById('envKey').textContent = s.env_api_key ? '环境变量: (已配置)' : '环境变量: (未设置)';
+    await loadMappings();
+    checkHealth();
+  } catch (e) {
+    toast('加载设置失败: ' + e.message, false);
+  }
+}
+
+async function checkHealth() {
+  try {
+    const r = await fetch(API + '/health');
+    const d = await r.json();
+    const b = document.getElementById('statusBadge');
+    if (d.status === 'ok') {
+      b.textContent = '已连接';
+      b.style.background = 'rgba(34,197,94,.15)';
+      b.style.color = 'var(--green)';
+    } else {
+      b.textContent = '异常';
+    }
+  } catch {
+    const b = document.getElementById('statusBadge');
+    b.textContent = '离线';
+    b.style.background = 'rgba(239,68,68,.15)';
+    b.style.color = 'var(--red)';
+  }
+}
+
+async function saveSettings() {
+  try {
+    await api('/api/admin/settings', {
+      method: 'PUT',
+      body: JSON.stringify({
+        proxy_target_url: document.getElementById('targetUrl').value.trim(),
+        proxy_api_key: document.getElementById('proxyKey').value.trim(),
+      }),
+    });
+    toast('设置已保存');
+  } catch (e) {
+    toast('保存失败: ' + e.message, false);
+  }
+}
+
+// ─── 模型映射 ───────────────────────────────────────
+async function loadMappings() {
+  const mappings = await api('/api/admin/mappings');
+  const el = document.getElementById('mappingList');
+  const keys = Object.keys(mappings);
+
+  if (!keys.length) {
+    el.innerHTML = '<div class="empty">暂无模型映射<br><span style="font-size:13px">点击「+ 添加映射」开始配置</span></div>';
+    return;
+  }
+
+  el.innerHTML = '<div class="mapping-list">' + keys.map(name => {
+    const m = mappings[name];
+    const backend = m.backend || 'auto';
+    const tagClass = backend === 'anthropic' ? 'tag-anthropic' : backend === 'openai' ? 'tag-openai' : 'tag-auto';
+    const tagLabel = backend === 'auto' ? '自动' : backend;
+    const hasOverride = m.target_url || m.api_key;
+    return `<div class="mapping-item">
+      <div class="mapping-top">
+        <span class="mapping-name">${esc(name)}</span>
+        <span class="mapping-arrow">&rarr;</span>
+        <span class="mapping-upstream">${esc(m.upstream_model || name)}</span>
+        <div class="mapping-meta">
+          <span class="tag ${tagClass}">${tagLabel}</span>
+          ${hasOverride ? '<span class="tag tag-override">自定义地址</span>' : ''}
+        </div>
+        <div class="mapping-actions">
+          <button class="btn btn-ghost btn-sm" onclick="openEditModal('${esc(name)}')">编辑</button>
+          <button class="btn btn-red btn-sm" onclick="deleteMapping('${esc(name)}')">删除</button>
+        </div>
+      </div>
+    </div>`;
+  }).join('') + '</div>';
+}
+
+function esc(s) { return String(s).replace(/&/g,'&amp;').replace(/</g,'&lt;').replace(/>/g,'&gt;').replace(/"/g,'&quot;').replace(/'/g,'&#39;'); }
+
+// ─── 弹窗 ──────────────────────────────────────────
+function openAddModal() {
+  editingName = null;
+  document.getElementById('modalTitle').textContent = '添加模型映射';
+  document.getElementById('mName').value = '';
+  document.getElementById('mName').disabled = false;
+  document.getElementById('mUpstream').value = '';
+  document.getElementById('mBackend').value = 'auto';
+  document.getElementById('mUrl').value = '';
+  document.getElementById('mKey').value = '';
+  document.getElementById('modal').classList.add('active');
+}
+
+async function openEditModal(name) {
+  editingName = name;
+  document.getElementById('modalTitle').textContent = '编辑模型映射';
+  try {
+    const mappings = await api('/api/admin/mappings');
+    const m = mappings[name];
+    if (!m) { toast('映射未找到', false); return; }
+    document.getElementById('mName').value = name;
+    document.getElementById('mName').disabled = false;
+    document.getElementById('mUpstream').value = m.upstream_model || '';
+    document.getElementById('mBackend').value = m.backend || 'auto';
+    document.getElementById('mUrl').value = m.target_url || '';
+    document.getElementById('mKey').value = m.api_key || '';
+    document.getElementById('modal').classList.add('active');
+  } catch (e) {
+    toast('错误: ' + e.message, false);
+  }
+}
+
+function closeModal() {
+  document.getElementById('modal').classList.remove('active');
+  editingName = null;
+}
+
+async function saveMapping() {
+  const name = document.getElementById('mName').value.trim();
+  const upstream = document.getElementById('mUpstream').value.trim();
+  if (!name) { toast('请填写 Cursor 模型名', false); return; }
+  if (!upstream) { toast('请填写上游模型名', false); return; }
+
+  const payload = {
+    name,
+    upstream_model: upstream,
+    backend: document.getElementById('mBackend').value,
+    target_url: document.getElementById('mUrl').value.trim(),
+    api_key: document.getElementById('mKey').value.trim(),
+  };
+
+  try {
+    if (editingName) {
+      await api('/api/admin/mappings/' + encodeURIComponent(editingName), {
+        method: 'PUT', body: JSON.stringify(payload),
+      });
+      toast('映射已更新');
+    } else {
+      await api('/api/admin/mappings', {
+        method: 'POST', body: JSON.stringify(payload),
+      });
+      toast('映射已添加');
+    }
+    closeModal();
+    await loadMappings();
+  } catch (e) {
+    toast('操作失败: ' + e.message, false);
+  }
+}
+
+async function deleteMapping(name) {
+  if (!confirm('确定要删除映射「' + name + '」吗？')) return;
+  try {
+    await api('/api/admin/mappings/' + encodeURIComponent(name), { method: 'DELETE' });
+    toast('映射已删除');
+    await loadMappings();
+  } catch (e) {
+    toast('删除失败: ' + e.message, false);
+  }
+}
+
+// ─── 初始化 ─────────────────────────────────────────
+(function init() {
+  const saved = sessionStorage.getItem('_ak');
+  if (saved) {
+    authKey = saved;
+    document.getElementById('login').style.display = 'none';
+    document.getElementById('dashboard').style.display = 'block';
+    loadDashboard();
+  }
+})();
+
+document.getElementById('modal').addEventListener('click', function(e) {
+  if (e.target === this) closeModal();
+});
+document.addEventListener('keydown', function(e) {
+  if (e.key === 'Escape') closeModal();
+});
diff --git a/utils/__init__.py b/utils/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/utils/http.py b/utils/http.py
new file mode 100644
index 0000000..9793ed8
--- /dev/null
+++ b/utils/http.py
@@ -0,0 +1,131 @@
+"""HTTP 工具 - 请求头构建、上游转发、SSE 流解析、响应构建"""
+
+import json
+import uuid
+import logging
+
+import requests
+from flask import Response, jsonify
+
+from config import Config
+
+logger = logging.getLogger(__name__)
+
+
+def gen_id(prefix=''):
+    """生成唯一 ID"""
+    return f'{prefix}{uuid.uuid4().hex[:24]}'
+
+
+# ─── 请求头构建 ────────────────────────────────────
+
+
+def build_openai_headers(api_key):
+    """构建 OpenAI 兼容请求头"""
+    return {
+        'Authorization': f'Bearer {api_key}',
+        'Content-Type': 'application/json',
+    }
+
+
+def build_anthropic_headers(api_key):
+    """构建 Anthropic 请求头，根据密钥前缀自动选择鉴权方式"""
+    headers = {
+        'anthropic-version': '2023-06-01',
+        'Content-Type': 'application/json',
+    }
+    if api_key.startswith('sk-'):
+        headers['x-api-key'] = api_key
+    else:
+        headers['Authorization'] = f'Bearer {api_key}'
+    return headers
+
+
+# ─── 响应构建 ──────────────────────────────────────
+
+
+def sse_response(generator):
+    """将生成器包装为 SSE 流式响应"""
+    return Response(
+        generator,
+        content_type='text/event-stream',
+        headers={'Cache-Control': 'no-cache', 'X-Accel-Buffering': 'no'},
+    )
+
+
+def error_json(message, error_type='proxy_error', status=502):
+    """构建 JSON 错误响应"""
+    return jsonify({'error': {'message': str(message), 'type': error_type}}), status
+
+
+# ─── 上游请求转发 ──────────────────────────────────
+
+
+def forward_request(url, headers, payload, stream=False):
+    """转发请求到上游 API
+
+    返回值:
+        成功: (response, None)
+        失败（流式）: (None, error_body_str)
+        失败（非流式）: (None, Flask Response)
+    """
+    try:
+        resp = requests.post(
+            url, headers=headers, json=payload,
+            timeout=Config.API_TIMEOUT, stream=stream,
+        )
+        if resp.status_code != 200:
+            body = resp.content.decode('utf-8', errors='replace')
+            logger.warning(f'上游返回 {resp.status_code}: {body[:300]}')
+            if stream:
+                return None, f'上游错误 {resp.status_code}: {body}'
+            return None, Response(
+                resp.content, status=resp.status_code,
+                content_type=resp.headers.get('Content-Type', 'application/json'),
+            )
+        return resp, None
+    except requests.RequestException as e:
+        logger.error(f'请求上游失败: {e}')
+        if stream:
+            return None, str(e)
+        return None, error_json(str(e))
+
+
+# ─── SSE 流解析 ───────────────────────────────────
+
+
+def iter_openai_sse(response):
+    """解析 OpenAI SSE 流，yield chunk 字典；yield None 表示 [DONE]"""
+    for line in response.iter_lines():
+        if not line:
+            continue
+        decoded = line.decode('utf-8', errors='replace')
+        if not decoded.startswith('data:'):
+            continue
+        data_str = decoded[5:].strip()
+        if data_str == '[DONE]':
+            yield None
+            return
+        try:
+            yield json.loads(data_str)
+        except json.JSONDecodeError:
+            continue
+
+
+def iter_anthropic_sse(response):
+    """解析 Anthropic SSE 流，yield (event_type, data_dict) 元组"""
+    event_type = ''
+    for line in response.iter_lines():
+        if not line:
+            continue
+        decoded = line.decode('utf-8', errors='replace')
+        if decoded.startswith('event:'):
+            event_type = decoded[6:].strip()
+        elif decoded.startswith('data:'):
+            data_str = decoded[5:].strip()
+            if not data_str:
+                continue
+            try:
+                yield event_type, json.loads(data_str)
+            except json.JSONDecodeError:
+                continue
diff --git a/utils/think_tag.py b/utils/think_tag.py
new file mode 100644
index 0000000..05158ef
--- /dev/null
+++ b/utils/think_tag.py
@@ -0,0 +1,101 @@
+"""<think> 标签提取器
+
+部分上游 API（如 DeepSeek）使用 <think>...</think> 标签包裹思考内容，
+而 Cursor 期望 reasoning_content 字段。本模块在流式和非流式响应中
+提取 <think> 标签内容并转为 reasoning_content。
+"""
+
+import re
+
+_THINK_RE = re.compile(r'<think>(.*?)</think>', re.DOTALL)
+
+
+def extract_from_text(content):
+    """从文本中提取 <think> 标签（非流式）
+
+    返回: (cleaned_content, reasoning_content)
+    """
+    if not isinstance(content, str) or '<think>' not in content:
+        return content, None
+    m = _THINK_RE.search(content)
+    if not m:
+        return content, None
+    reasoning = m.group(1).strip()
+    cleaned = (content[:m.start()] + content[m.end():]).strip() or None
+    return cleaned, reasoning
+
+
+class ThinkTagExtractor:
+    """流式 <think> 标签提取器
+
+    处理跨 chunk 的 <think>...</think> 标签，将标签内的文本
+    转为 reasoning_content delta，标签外的文本保持为 content delta。
+    """
+
+    def __init__(self):
+        self._in_thinking = False
+
+    def process_chunk(self, chunk):
+        """处理一个流式 chunk，返回转换后的 chunk 列表"""
+        for choice in (chunk.get('choices') or []):
+            delta = choice.get('delta') or {}
+            if delta.get('reasoning_content'):
+                return [chunk]
+            content = delta.get('content')
+            if content is None or content == '':
+                return [chunk]
+            return self._split(chunk, content)
+        return [chunk]
+
+    def _split(self, chunk, text):
+        """根据 <think> 标签拆分文本为多个 chunk"""
+        results = []
+
+        if self._in_thinking:
+            end = text.find('</think>')
+            if end >= 0:
+                self._in_thinking = False
+                if text[:end]:
+                    results.append(self._make(chunk, reasoning=text[:end]))
+                rest = text[end + 8:].lstrip('\n')
+                if rest:
+                    results.append(self._make(chunk, content=rest))
+            else:
+                results.append(self._make(chunk, reasoning=text))
+        else:
+            start = text.find('<think>')
+            if start >= 0:
+                before = text[:start]
+                after = text[start + 7:]
+                if before:
+                    results.append(self._make(chunk, content=before))
+                end = after.find('</think>')
+                if end >= 0:
+                    if after[:end]:
+                        results.append(self._make(chunk, reasoning=after[:end]))
+                    rest = after[end + 8:].lstrip('\n')
+                    if rest:
+                        results.append(self._make(chunk, content=rest))
+                else:
+                    self._in_thinking = True
+                    if after:
+                        results.append(self._make(chunk, reasoning=after))
+            else:
+                results.append(chunk)
+
+        return results or [chunk]
+
+    @staticmethod
+    def _make(template, content=None, reasoning=None):
+        """根据模板 chunk 构造新的 delta chunk"""
+        delta = {}
+        if content is not None:
+            delta['content'] = content
+        if reasoning is not None:
+            delta['reasoning_content'] = reasoning
+        return {
+            'id': template.get('id', ''),
+            'object': 'chat.completion.chunk',
+            'model': template.get('model', ''),
+            'choices': [{'index': 0, 'delta': delta, 'finish_reason': None}],
+        }
diff --git a/utils/tool_fixer.py b/utils/tool_fixer.py
new file mode 100644
index 0000000..b699b73
--- /dev/null
+++ b/utils/tool_fixer.py
@@ -0,0 +1,134 @@
+"""工具参数修复
+
+修复 LLM 生成的工具调用参数中的常见问题：
+  - 智能引号 → 普通引号
+  - file_path → path 字段映射
+  - StrReplace 工具的 old_string 精确匹配修复
+  - Anthropic tool_use 块的 ID 和 stop_reason 修复
+"""
+
+import os
+import re
+import uuid
+
+# 智能引号字符集
+_SMART_DOUBLE = frozenset('«»\u201c\u201d\u275e\u201f\u201e\u275d')
+_SMART_SINGLE = frozenset('\u2018\u2019\u201a\u201b')
+
+
+def normalize_args(args):
+    """规范化工具参数：file_path → path"""
+    if isinstance(args, dict) and 'file_path' in args and 'path' not in args:
+        args['path'] = args.pop('file_path')
+    return args
+
+
+def repair_str_replace_args(tool_name, args):
+    """修复 StrReplace/search_replace 工具的精确匹配问题
+
+    当 old_string 包含智能引号导致无法精确匹配文件内容时，
+    用容错正则在文件中查找唯一匹配并替换为实际内容。
+    """
+    if not isinstance(args, dict):
+        return args
+
+    name_lower = (tool_name or '').lower()
+    if 'str_replace' not in name_lower and 'search_replace' not in name_lower:
+        return args
+
+    old_str = args.get('old_string') or args.get('old_str')
+    if not old_str:
+        return args
+
+    file_path = args.get('path') or args.get('file_path')
+    if not file_path or not os.path.isfile(file_path):
+        return args
+
+    try:
+        with open(file_path, 'r', encoding='utf-8', errors='replace') as f:
+            content = f.read()
+    except Exception:
+        return args
+
+    # 已精确匹配，无需修复
+    if old_str in content:
+        return args
+
+    # 构建容错正则尝试匹配
+    pattern = _build_fuzzy_pattern(old_str)
+    try:
+        matches = list(re.finditer(pattern, content))
+    except re.error:
+        return args
+
+    # 仅在唯一匹配时修复，避免歧义
+    if len(matches) != 1:
+        return args
+
+    matched = matches[0].group()
+    if 'old_string' in args:
+        args['old_string'] = matched
+    elif 'old_str' in args:
+        args['old_str'] = matched
+
+    # 同步修复 new_string 中的智能引号
+    new_str = args.get('new_string') or args.get('new_str')
+    if new_str:
+        fixed = _replace_smart_quotes(new_str)
+        if 'new_string' in args:
+            args['new_string'] = fixed
+        elif 'new_str' in args:
+            args['new_str'] = fixed
+
+    return args
+
+
+def fix_anthropic_tool_use(response_data):
+    """修复 Anthropic 响应中的 tool_use 块（补全 ID、修正 stop_reason）"""
+    if not isinstance(response_data, dict):
+        return response_data
+
+    content = response_data.get('content', [])
+    if not isinstance(content, list):
+        return response_data
+
+    has_tool_use = False
+    for block in content:
+        if isinstance(block, dict) and block.get('type') == 'tool_use':
+            has_tool_use = True
+            if not block.get('id'):
+                block['id'] = f'toolu_{uuid.uuid4().hex[:24]}'
+
+    if has_tool_use and response_data.get('stop_reason') != 'tool_use':
+        response_data['stop_reason'] = 'tool_use'
+
+    return response_data
+
+
+# ─── 内部辅助 ──────────────────────────────────────
+
+
+def _build_fuzzy_pattern(text):
+    """构建容错正则：智能引号可互换、空白可伸缩、反斜杠可重复"""
+    parts = []
+    for ch in text:
+        if ch in _SMART_DOUBLE or ch == '"':
+            parts.append('["\u00ab\u201c\u201d\u275e\u201f\u201e\u275d\u00bb]')
+        elif ch in _SMART_SINGLE or ch == "'":
+            parts.append("['\u2018\u2019\u201a\u201b]")
+        elif ch in (' ', '\t'):
+            parts.append(r'\s+')
+        elif ch == '\\':
+            parts.append(r'\\{1,2}')
+        else:
+            parts.append(re.escape(ch))
+    return ''.join(parts)
+
+
+def _replace_smart_quotes(text):
+    """将智能引号替换为普通 ASCII 引号"""
+    return ''.join(
+        '"' if ch in _SMART_DOUBLE else
+        "'" if ch in _SMART_SINGLE else
+        ch for ch in text
+    )