prompt-manager-client 0.2.0__tar.gz

prompt_manager_client-0.2.0/PKG-INFO

@@ -0,0 +1,22 @@
+Metadata-Version: 2.4
+Name: prompt-manager-client
+Version: 0.2.0
+Summary: 提示词管理系统 - 客户端 SDK。支持三维管理、变量解析、热更新。
+Author: prompt-manager
+License-Expression: MIT
+Project-URL: Homepage, http://localhost:8900
+Project-URL: Source, https://github.com/your-org/prompt-manager
+Keywords: prompt,prompt-manager,llm
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.8
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.8
+Provides-Extra: ws
+Requires-Dist: websocket-client; extra == "ws"
+Provides-Extra: dev
+Requires-Dist: build; extra == "dev"
+Requires-Dist: twine; extra == "dev"
+Requires-Dist: websocket-client; extra == "dev"

prompt_manager_client-0.2.0/README.md

@@ -0,0 +1,227 @@
+# Prompt Manager - 提示词 & 知识库管理系统
+
+统一的提示词和知识库管理平台，支持多科室、多平台的三维管理，提供 Web 管理界面和 Python 客户端 SDK。
+
+## 项目结构
+
+```
+prompt_manager/
+├── backend/                  # 后端服务
+│   ├── main.py               # FastAPI 主应用（API 路由）
+│   ├── database.py           # SQLite 数据库层（CRUD + 缓存）
+│   ├── models.py             # Pydantic 数据模型
+│   ├── requirements.txt      # Python 依赖
+│   ├── start.sh              # 后端启动脚本
+│   └── prompt_manager.db     # SQLite 数据库文件（自动创建）
+├── frontend/
+│   └── index.html            # 单页面管理界面（Vue 3）
+├── client_sdk/               # 客户端 SDK
+│   ├── __init__.py
+│   ├── prompt_client.py      # PromptClient 类
+│   └── pyproject.toml        # SDK 构建配置
+├── knowledge_retriever.py    # 知识库检索模块
+├── pyproject.toml            # 项目构建配置
+└── start.sh                  # 一键启动脚本
+```
+
+## 功能概览
+
+### 提示词管理
+
+| 功能 | 说明 |
+|------|------|
+| 三维管理 | 科室（口腔/植发/皮肤/眼科/儿科/医美）× 平台（百度/小红书/抖音/快手/微信）× 场景 |
+| 变量引擎 | 支持 `{公司}`、`{域中文}`、`{时间}`、`{轮次}` 等占位符自动解析 |
+| 版本管理 | 自动版本递增，支持一键回滚到历史版本 |
+| 热更新 | 修改即时生效，通过轮询或 WebSocket 推送到客户端 |
+
+### 知识库管理
+
+| 功能 | 说明 |
+|------|------|
+| 唯一索引 | 每个科室+平台组合只保留一个知识库 |
+| 条目管理 | 逐条编辑、删除、新增，每页 10 条分页 |
+| 本地持久化 | 数据存储在 SQLite，自动持久化 |
+| 智能检索 | 支持关键词检索（TF-IDF）和语义检索（可选） |
+
+## 快速开始
+
+### 1. 启动服务
+
+```bash
+# 方式一：一键启动
+cd prompt_manager
+bash start.sh
+
+# 方式二：手动启动
+cd backend
+pip install -r requirements.txt
+python main.py
+```
+
+服务启动后访问：
+- 管理界面：http://localhost:8900
+- API 文档：http://localhost:8900/docs
+
+### 2. 客户端 SDK 使用
+
+#### 安装
+
+```bash
+# 方式一：本地安装
+cd client_sdk
+pip install -e .
+
+# 方式二：直接复制 prompt_client.py 到项目中
+```
+
+#### 提示词相关
+
+```python
+from prompt_client import PromptClient
+
+client = PromptClient(base_url="http://localhost:8900")
+
+# 预加载
+client.preload()
+
+# 获取提示词内容
+content = client.get_content("hair_xhs_system")
+
+# 获取已解析变量的提示词（最常用）
+prompt = client.get_resolved(
+    name="hair_xhs_system",
+    robot_id="9125",
+    current_round=3,
+    action_desc="\n回复意图：问诊、套联"
+)
+
+# 启动后台热更新（30秒轮询一次）
+client.start_auto_update(interval=30)
+```
+
+#### 知识库相关
+
+```python
+# 获取知识库条目列表
+items = client.get_knowledge(department="hair", platform="xhs")
+# -> ["雄激素性脱发表现为...", "米诺地尔是常用药物...", ...]
+
+# 获取合并文本
+text = client.get_knowledge_text(department="hair", platform="xhs")
+# -> "雄激素性脱发表现为...\n米诺地尔是常用药物..."
+
+# 刷新知识库缓存
+client.refresh_knowledge(department="hair", platform="xhs")
+
+# 预加载全部（提示词 + 知识库）
+client.preload()
+```
+
+### 3. 知识库检索
+
+```python
+from knowledge_retriever import KnowledgeRetriever
+
+# 关键词检索（默认，无需额外依赖）
+retriever = KnowledgeRetriever(base_url="http://localhost:8900")
+retriever.preload(department="hair", platform="xhs")
+
+history = [
+    {"role": "user", "content": "最近掉头发很严重怎么办"},
+    {"role": "assistant", "content": "请问您掉发持续多长时间了？"}
+]
+
+# 检索相关知识
+results = retriever.retrieve(history, department="hair", platform="xhs", top_k=5)
+
+# 获取拼接文本（直接插入 prompt）
+knowledge_text = retriever.retrieve_as_text(
+    history, department="hair", platform="xbs",
+    top_k=5, separator="\n"
+)
+```
+
+#### 语义检索（可选）
+
+```bash
+pip install sentence-transformers
+```
+
+```python
+retriever = KnowledgeRetriever(
+    base_url="http://localhost:8900",
+    use_semantic=True,
+    semantic_model="shibing624/text2vec-base-chinese"
+)
+# 使用方式与关键词检索完全相同
+```
+
+## API 接口
+
+### 提示词 API
+
+| 方法 | 路径 | 说明 |
+|------|------|------|
+| POST | `/api/prompts` | 创建提示词 |
+| GET | `/api/prompts` | 列表查询（支持科室/平台/场景/关键词过滤） |
+| GET | `/api/prompts/{id}` | 获取详情 |
+| PUT | `/api/prompts/{id}` | 更新内容 |
+| DELETE | `/api/prompts/{id}` | 删除 |
+| GET | `/api/prompts/{id}/versions` | 获取版本历史 |
+| POST | `/api/prompts/{id}/rollback` | 回滚到指定版本 |
+| GET | `/api/v1/fetch/{name}` | 按名称获取（客户端用） |
+| GET | `/api/v1/sync` | 全量同步（客户端用） |
+| POST | `/api/v1/resolve` | 变量解析 |
+
+### 知识库 API
+
+| 方法 | 路径 | 说明 |
+|------|------|------|
+| POST | `/api/knowledge` | 创建知识库（科室+平台唯一） |
+| GET | `/api/knowledge` | 列表查询（支持科室/平台/关键词过滤） |
+| GET | `/api/knowledge/{id}` | 获取详情 |
+| PUT | `/api/knowledge/{id}` | 更新内容 |
+| DELETE | `/api/knowledge/{id}` | 删除 |
+
+### WebSocket
+
+| 路径 | 说明 |
+|------|------|
+| `/ws/updates` | 实时推送提示词/知识库变更 |
+
+## 科室与平台编码
+
+| 科室代码 | 中文名 | 平台代码 | 中文名 |
+|----------|--------|----------|--------|
+| hair | 植发科 | xhs | 小红书 |
+| dentistry | 口腔科 | bd | 百度 |
+| dermatology | 皮肤科 | dy | 抖音 |
+| ophthalmology | 眼科 | kuaishou | 快手 |
+| pediatrics | 儿科 | wechat | 微信 |
+| beauty | 医美 | general | 通用 |
+| general | 通用 | | |
+
+## 客户端 SDK 构建
+
+```bash
+cd client_sdk
+
+# 构建
+pip install build
+python -m build
+
+# 本地安装
+pip install -e .
+
+# 发布到 PyPI（可选）
+pip install twine
+twine upload dist/*
+```
+
+## 技术栈
+
+- **后端**：FastAPI + SQLite + Pydantic
+- **前端**：Vue 3（CDN）+ 单 HTML 文件
+- **客户端 SDK**：纯 Python，零依赖（语义检索可选 sentence-transformers）
+- **数据库**：SQLite，自动迁移

prompt_manager_client-0.2.0/client_sdk/__init__.py

@@ -0,0 +1,3 @@
+from .prompt_client import PromptClient
+
+__all__ = ["PromptClient"]

prompt_manager_client-0.2.0/client_sdk/prompt_client.py

@@ -0,0 +1,431 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+"""
+提示词 & 知识库管理系统 - 客户端 SDK
+
+专为模型服务设计，支持：
+1. 按科室+平台+场景获取提示词
+2. 按科室+平台获取知识库内容
+3. 变量自动解析（公司、时间、轮次等）
+4. 热更新（轮询/WebSocket）
+5. 本地缓存，读取零延迟
+
+使用示例：
+    from prompt_client import PromptClient
+
+    client = PromptClient(base_url="http://localhost:8900")
+
+    # 获取已解析的提示词
+    prompt = client.get_resolved(
+        name="hair_xhs_system",
+        robot_id="9125",
+        current_round=3,
+        action_desc="\\n回复意图：问诊、套联"
+    )
+
+    # 获取知识库
+    kb_items = client.get_knowledge(department="hair", platform="xhs")
+    # kb_items -> ["知识条目1", "知识条目2", ...]
+
+    # 启动后台热更新
+    client.start_auto_update(interval=30)
+"""
+
+import time
+import threading
+import logging
+import json
+from typing import Optional, Dict, List, Callable
+from urllib.request import urlopen, Request
+from urllib.error import URLError, HTTPError
+
+logger = logging.getLogger("prompt_client")
+
+
+class PromptClient:
+    """
+    提示词 & 知识库管理客户端
+
+    特性：
+    - 三维管理：科室 + 平台 + 场景
+    - 知识库管理：科室 + 平台 唯一索引
+    - 本地缓存，读取零延迟
+    - 变量自动解析
+    - 支持轮询和WebSocket热更新
+    - 线程安全
+    """
+
+    def __init__(self, base_url: str = "http://localhost:8900"):
+        self.base_url = base_url.rstrip("/")
+        # 提示词缓存: {name: {content, version, department, platform, scene, variables, updated_at}}
+        self._cache: Dict[str, dict] = {}
+        # 已解析缓存: {cache_key: (resolved_content, version)}
+        self._resolved_cache: Dict[str, tuple] = {}
+        # 知识库缓存: {(department, platform): {id, content, updated_at}}
+        self._kb_cache: Dict[tuple, dict] = {}
+        self._lock = threading.Lock()
+        self._running = False
+        self._poll_thread: Optional[threading.Thread] = None
+        self._on_update_callback: Optional[Callable] = None
+
+    # ==================== 提示词接口 ====================
+
+    def get(self, name: str) -> Optional[dict]:
+        """获取提示词原始数据（从缓存，零延迟）"""
+        with self._lock:
+            if name in self._cache:
+                return dict(self._cache[name])
+        self._fetch_one(name)
+        with self._lock:
+            return dict(self._cache[name]) if name in self._cache else None
+
+    def get_content(self, name: str, default: str = "") -> str:
+        """获取提示词原始内容"""
+        data = self.get(name)
+        return data["content"] if data else default
+
+    def get_resolved(
+        self,
+        name: str,
+        robot_id: str = "",
+        department: str = "",
+        current_round: int = 0,
+        action_desc: str = "",
+        knowledge_desc: str = "",
+        warmup_desc: str = "",
+        connect_desc: str = "",
+        extra_variables: dict = None,
+    ) -> str:
+        """
+        获取已解析变量的提示词（最常用接口）
+
+        内部会自动解析 {公司}, {域中文}, {时间}, {轮次} 等变量，
+        并调用远程 resolve API 处理更复杂的变量替换。
+        """
+        extra_variables = extra_variables or {}
+
+        # 先确保缓存中有数据
+        data = self.get(name)
+        if not data:
+            return ""
+
+        # 构建缓存key
+        cache_key = f"{name}::{robot_id}::{department}::{current_round}::{action_desc}::{knowledge_desc}::{warmup_desc}::{connect_desc}"
+
+        with self._lock:
+            if cache_key in self._resolved_cache:
+                resolved, version = self._resolved_cache[cache_key]
+                if version == data["version"]:
+                    return resolved
+
+        # 调用远程解析API
+        try:
+            payload = json.dumps({
+                "name": name,
+                "robot_id": robot_id,
+                "department": department,
+                "current_round": current_round,
+                "action_desc": action_desc,
+                "knowledge_desc": knowledge_desc,
+                "warmup_desc": warmup_desc,
+                "connect_desc": connect_desc,
+                "extra_variables": extra_variables
+            }).encode()
+            req = Request(
+                f"{self.base_url}/api/v1/resolve",
+                data=payload,
+                headers={"Content-Type": "application/json"}
+            )
+            with urlopen(req, timeout=10) as resp:
+                result = json.loads(resp.read().decode())
+                resolved = result["resolved_content"]
+                with self._lock:
+                    self._resolved_cache[cache_key] = (resolved, result["version"])
+                return resolved
+        except Exception as e:
+            logger.warning(f"远程解析失败，使用本地缓存: {e}")
+            return data.get("content", "")
+
+    def get_version(self, name: str) -> int:
+        """获取提示词版本号"""
+        with self._lock:
+            return self._cache.get(name, {}).get("version", 0)
+
+    def get_all_names(self) -> list:
+        """获取所有缓存中的提示词名称"""
+        with self._lock:
+            return list(self._cache.keys())
+
+    # ==================== 知识库接口 ====================
+
+    def get_knowledge(self, department: str, platform: str) -> List[str]:
+        """
+        根据科室和平台获取知识库内容列表
+
+        Args:
+            department: 科室代码，如 hair, dentistry, dermatology 等
+            platform: 平台代码，如 xhs, bd, dy 等
+
+        Returns:
+            知识条目列表，如 ["条目1", "条目2", ...]，未找到返回空列表
+        """
+        key = (department, platform)
+        with self._lock:
+            if key in self._kb_cache:
+                content = self._kb_cache[key].get("content", "[]")
+                return self._parse_kb_content(content)
+
+        # 从服务端拉取
+        self._fetch_knowledge(department, platform)
+        with self._lock:
+            if key in self._kb_cache:
+                content = self._kb_cache[key].get("content", "[]")
+                return self._parse_kb_content(content)
+        return []
+
+    def get_knowledge_text(self, department: str, platform: str, separator: str = "\n") -> str:
+        """
+        根据科室和平台获取知识库内容，合并为单个字符串
+
+        Args:
+            department: 科室代码
+            platform: 平台代码
+            separator: 条目之间的分隔符，默认换行
+
+        Returns:
+            合并后的知识库文本
+        """
+        items = self.get_knowledge(department, platform)
+        return separator.join(items) if items else ""
+
+    def get_all_knowledge(self) -> Dict[str, List[str]]:
+        """
+        获取所有缓存中的知识库
+
+        Returns:
+            字典，key 为 "科室/平台"，value 为条目列表
+        """
+        result = {}
+        with self._lock:
+            for (dept, plat), data in self._kb_cache.items():
+                result[f"{dept}/{plat}"] = self._parse_kb_content(data.get("content", "[]"))
+        return result
+
+    def refresh_knowledge(self, department: str = None, platform: str = None):
+        """
+        刷新知识库缓存
+
+        Args:
+            department: 科室代码，为空则刷新全部
+            platform: 平台代码，为空则刷新对应科室的全部或全部
+        """
+        if department and platform:
+            self._fetch_knowledge(department, platform)
+        else:
+            self._sync_all_knowledge(department=department, platform=platform)
+
+    @staticmethod
+    def _parse_kb_content(content_json: str) -> List[str]:
+        """解析知识库 content JSON 为列表"""
+        try:
+            arr = json.loads(content_json or "[]")
+            return arr if isinstance(arr, list) else []
+        except (json.JSONDecodeError, TypeError):
+            return []
+
+    def _fetch_knowledge(self, department: str, platform: str) -> bool:
+        """从服务端获取单个知识库"""
+        result = self._http_get(f"/api/knowledge?department={department}&platform={platform}")
+        if not result or not result.get("items"):
+            return False
+        item = result["items"][0]
+        key = (department, platform)
+        with self._lock:
+            old = self._kb_cache.get(key, {})
+            self._kb_cache[key] = item
+        if old.get("updated_at") != item.get("updated_at"):
+            logger.info(f"知识库 [{department}/{platform}] 已更新")
+            if self._on_update_callback:
+                try:
+                    self._on_update_callback(f"kb:{department}/{platform}", item)
+                except Exception as e:
+                    logger.error(f"知识库更新回调失败: {e}")
+        return True
+
+    def _sync_all_knowledge(self, department: str = None, platform: str = None):
+        """同步所有知识库"""
+        params = []
+        if department:
+            params.append(f"department={department}")
+        if platform:
+            params.append(f"platform={platform}")
+        query = "&".join(params)
+        path = f"/api/knowledge?{query}" if query else "/api/knowledge"
+        result = self._http_get(path)
+        if not result:
+            return
+        for item in result.get("items", []):
+            key = (item.get("department", ""), item.get("platform", ""))
+            if not key[0] or not key[1]:
+                continue
+            with self._lock:
+                old = self._kb_cache.get(key, {})
+                self._kb_cache[key] = item
+                if old.get("updated_at") != item.get("updated_at"):
+                    logger.info(f"[同步] 知识库 [{key[0]}/{key[1]}] 已更新")
+
+    # ==================== 通用接口 ====================
+
+    def preload(self):
+        """全量预加载所有提示词和知识库"""
+        self._sync_all()
+        self._sync_all_knowledge()
+
+    def on_update(self, callback: Callable[[str, dict], None]):
+        """注册更新回调: callback(key, data)，key 对提示词为名称，对知识库为 kb:科室/平台"""
+        self._on_update_callback = callback
+
+    def start_auto_update(self, interval: int = 30, names: list = None):
+        """启动轮询热更新（提示词+知识库）"""
+        if self._running:
+            return
+        self._running = True
+        self._poll_thread = threading.Thread(
+            target=self._poll_loop, args=(interval, names),
+            daemon=True, name="prompt-poll"
+        )
+        self._poll_thread.start()
+        logger.info(f"轮询热更新已启动，间隔 {interval}s")
+
+    def start_ws_update(self):
+        """启动WebSocket实时更新"""
+        try:
+            import websocket
+        except ImportError:
+            logger.error("需要安装 websocket-client: pip install websocket-client")
+            return
+        if self._ws_thread and hasattr(self, '_ws_thread') and self._ws_thread.is_alive():
+            return
+        self._running = True
+        self._ws_thread = threading.Thread(
+            target=self._ws_loop, daemon=True, name="prompt-ws"
+        )
+        self._ws_thread.start()
+
+    def stop(self):
+        """停止后台更新"""
+        self._running = False
+        if self._poll_thread:
+            self._poll_thread.join(timeout=5)
+        logger.info("热更新已停止")
+
+    # ========== 内部方法 ==========
+    def _http_get(self, path: str) -> Optional[dict]:
+        try:
+            req = Request(f"{self.base_url}{path}")
+            with urlopen(req, timeout=10) as resp:
+                return json.loads(resp.read().decode())
+        except (URLError, HTTPError, TimeoutError) as e:
+            logger.warning(f"请求失败 {path}: {e}")
+            return None
+
+    def _fetch_one(self, name: str) -> bool:
+        with self._lock:
+            current_version = self._cache.get(name, {}).get("version", 0)
+        result = self._http_get(f"/api/v1/fetch/{name}?current_version={current_version}")
+        if result is None:
+            if current_version == 0:
+                result = self._http_get(f"/api/v1/fetch/{name}")
+                if not result:
+                    return False
+            else:
+                return True
+        with self._lock:
+            old_version = self._cache.get(name, {}).get("version", 0)
+            self._cache[name] = result
+        if old_version != result.get("version", 0):
+            logger.info(f"提示词 '{name}' 已更新到 v{result['version']}")
+            if self._on_update_callback:
+                try:
+                    self._on_update_callback(name, result)
+                except Exception as e:
+                    logger.error(f"更新回调失败: {e}")
+        return True
+
+    def _sync_all(self):
+        result = self._http_get("/api/v1/sync")
+        if not result:
+            return
+        for item in result.get("prompts", []):
+            name = item.get("name")
+            if not name:
+                continue
+            with self._lock:
+                old_v = self._cache.get(name, {}).get("version", 0)
+                if item.get("version", 0) > old_v:
+                    self._cache[name] = item
+                    logger.info(f"[同步] '{name}' v{old_v} -> v{item['version']}")
+
+    def _poll_loop(self, interval, names):
+        while self._running:
+            try:
+                if names:
+                    for n in names:
+                        self._fetch_one(n)
+                else:
+                    self._sync_all()
+                self._sync_all_knowledge()
+            except Exception as e:
+                logger.error(f"轮询异常: {e}")
+            for _ in range(interval):
+                if not self._running:
+                    break
+                time.sleep(1)
+
+    def _ws_loop(self):
+        import websocket
+
+        def on_message(ws, message):
+            try:
+                msg = json.loads(message)
+                name = msg.get("name")
+                if name and msg.get("event") in ("created", "updated", "rollback"):
+                    self._fetch_one(name)
+                    # 清除已解析缓存
+                    with self._lock:
+                        keys_to_remove = [k for k in self._resolved_cache if k.startswith(f"{name}::")]
+                        for k in keys_to_remove:
+                            del self._resolved_cache[k]
+                # 知识库更新
+                if msg.get("type") == "knowledge_updated":
+                    dept = msg.get("department")
+                    plat = msg.get("platform")
+                    if dept and plat:
+                        self._fetch_knowledge(dept, plat)
+            except Exception as e:
+                logger.error(f"WS消息处理失败: {e}")
+
+        def on_error(ws, error):
+            logger.warning(f"WS错误: {error}")
+
+        def on_close(ws, *args):
+            logger.info("WS连接关闭")
+
+        ws_url = self.base_url.replace("http://", "ws://").replace("https://", "wss://")
+        while self._running:
+            try:
+                ws = websocket.WebSocketApp(
+                    f"{ws_url}/ws/updates",
+                    on_message=on_message, on_error=on_error, on_close=on_close
+                )
+                ws.run_forever(ping_interval=30, ping_timeout=10)
+            except Exception as e:
+                logger.error(f"WS异常: {e}")
+            if self._running:
+                time.sleep(5)
+
+    def __repr__(self):
+        with self._lock:
+            return (f"PromptClient(cached_prompts={len(self._cache)}, "
+                    f"cached_kb={len(self._kb_cache)}, "
+                    f"names={list(self._cache.keys())})")

prompt_manager_client-0.2.0/prompt_manager_client.egg-info/PKG-INFO

@@ -0,0 +1,22 @@
+Metadata-Version: 2.4
+Name: prompt-manager-client
+Version: 0.2.0
+Summary: 提示词管理系统 - 客户端 SDK。支持三维管理、变量解析、热更新。
+Author: prompt-manager
+License-Expression: MIT
+Project-URL: Homepage, http://localhost:8900
+Project-URL: Source, https://github.com/your-org/prompt-manager
+Keywords: prompt,prompt-manager,llm
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.8
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Operating System :: OS Independent
+Requires-Python: >=3.8
+Provides-Extra: ws
+Requires-Dist: websocket-client; extra == "ws"
+Provides-Extra: dev
+Requires-Dist: build; extra == "dev"
+Requires-Dist: twine; extra == "dev"
+Requires-Dist: websocket-client; extra == "dev"

prompt_manager_client-0.2.0/prompt_manager_client.egg-info/SOURCES.txt

@@ -0,0 +1,9 @@
+README.md
+pyproject.toml
+client_sdk/__init__.py
+client_sdk/prompt_client.py
+prompt_manager_client.egg-info/PKG-INFO
+prompt_manager_client.egg-info/SOURCES.txt
+prompt_manager_client.egg-info/dependency_links.txt
+prompt_manager_client.egg-info/requires.txt
+prompt_manager_client.egg-info/top_level.txt

prompt_manager_client-0.2.0/prompt_manager_client.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

prompt_manager_client-0.2.0/prompt_manager_client.egg-info/requires.txt

@@ -0,0 +1,8 @@
+
+[dev]
+build
+twine
+websocket-client
+
+[ws]
+websocket-client

prompt_manager_client-0.2.0/prompt_manager_client.egg-info/top_level.txt

@@ -0,0 +1 @@
+client_sdk

prompt_manager_client-0.2.0/pyproject.toml

@@ -0,0 +1,33 @@
+[build-system]
+requires = ["setuptools>=61.0"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "prompt-manager-client"
+version = "0.2.0"
+description = "提示词管理系统 - 客户端 SDK。支持三维管理、变量解析、热更新。"
+requires-python = ">=3.8"
+license = "MIT"
+keywords = ["prompt", "prompt-manager", "llm"]
+authors = [
+    {name = "prompt-manager"},
+]
+classifiers = [
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.8",
+    "Programming Language :: Python :: 3.9",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Operating System :: OS Independent",
+]
+
+[project.urls]
+Homepage = "http://localhost:8900"
+Source = "https://github.com/your-org/prompt-manager"
+
+[project.optional-dependencies]
+ws = ["websocket-client"]
+dev = ["build", "twine", "websocket-client"]
+
+[tool.setuptools.packages.find]
+include = ["client_sdk*"]

prompt_manager_client-0.2.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+