← 返回 Skills 市场
214
总下载
0
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install chatbot-engine
功能描述
智能对话引擎 - 多轮对话与意图识别 | Chatbot Engine - Multi-turn dialogue and intent recognition
使用说明 (SKILL.md)
Chatbot Engine - 智能对话引擎
企业级对话系统解决方案，支持多轮对话、意图识别、上下文管理和知识库检索。
核心功能
| 功能模块 | 说明 |
|---|---|
| 意图识别 | 基于规则/机器学习的意图分类 |
| 实体抽取 | 命名实体识别（人名、地点、时间等） |
| 多轮对话 | 上下文感知的多轮交互 |
| 知识库检索 | 基于向量检索的知识问答 |
| 对话管理 | 对话状态跟踪和流程控制 |
快速开始
from scripts.dialogue_manager import DialogueManager
# 创建对话管理器
bot = DialogueManager()
# 处理用户输入
response = bot.process("我想预订明天北京的酒店")
print(response)
安装
pip install -r requirements.txt
项目结构
chatbot-engine/
├── SKILL.md # Skill说明文档
├── README.md # 完整文档
├── requirements.txt # 依赖列表
├── scripts/ # 核心模块
│ ├── dialogue_manager.py # 对话管理器
│ ├── intent_classifier.py # 意图分类器
│ ├── entity_extractor.py # 实体抽取器
│ └── knowledge_base.py # 知识库
├── examples/ # 使用示例
│ └── basic_usage.py
└── tests/ # 单元测试
└── test_chatbot.py
安全使用建议
This package largely implements a local chatbot, but review and proceed cautiously:
- Missing declared env vars: The LLM adapter will read env vars like OPENAI_API_KEY or ANTHROPIC_API_KEY if you initialize with those providers; these were not listed as required. Do not install or run the skill with sensitive API keys in your environment unless you trust it.
- Missing dependencies: The code imports sentence_transformers and fuzzywuzzy, but requirements.txt does not list them. Either add those to requirements.txt or run in 'mock' or 'local' mode to avoid installing them. Unexpected manual dependency installation increases risk if you copy commands from third-party sources.
- Network calls and data flow: If you configure provider='openai' or 'anthropic' the adapter will use the respective client libraries to call external APIs and send prompts/context. That means user messages and context can be transmitted to those services — expected for a chatbot but important to know for sensitive data.
- Run in a sandbox first: Execute unit tests and examples in an isolated environment (virtualenv/container) and try provider='mock' to observe behavior without network calls.
- If you need to use production LLMs, explicitly set and scope API keys and audit network usage; consider adding explicit declarations to requirements and SKILL.md documenting env var names and what will be transmitted.
- If you want to be stricter: update requirements.txt to include sentence_transformers and fuzzywuzzy (or modify code to optionally import them), and document the env vars the adapter will read. If you cannot verify these, treat the package as untrusted code and do not provide real API keys or sensitive data.
功能分析
Type: OpenClaw Skill
Name: chatbot-engine
Version: 1.0.0
The chatbot-engine skill bundle is a standard implementation of a conversational AI framework featuring intent recognition, entity extraction, and knowledge base management. The code in scripts/chatbot.py and scripts/llm_adapter.py follows best practices for session management and API key handling via environment variables. No evidence of malicious behavior, data exfiltration, or prompt injection was found across the documentation or implementation files.
能力评估
Purpose & Capability
Name/description describe a multi-turn chatbot and the included Python modules implement exactly that (dialogue manager, intent classifier, KB, LLM adapter). The included LLMAdapter (openai/anthropic/local/mock) and knowledge-base/vector logic align with the stated purpose.
Instruction Scope
SKILL.md gives only standard usage/installation instructions (pip install -r requirements.txt and usage examples). It doesn't instruct the agent to read unrelated files or exfiltrate data. Runtime code does perform file I/O (save/load sessions, save/load KB/intents) which is expected for this purpose.
Install Mechanism
There is no install spec beyond 'pip install -r requirements.txt'. The requirements.txt does not include packages that the code imports (e.g., sentence_transformers and fuzzywuzzy are used but not listed). That mismatch may cause runtime failures or unexpected manual installs. No remote downloads are present, but missing deps increase friction and hidden manual steps.
Credentials
The skill declares no required environment variables, but LLMAdapter reads provider-specific env vars like OPENAI_API_KEY / ANTHROPIC_API_KEY (os.getenv(f"{provider.upper()}_API_KEY")). Those keys are not declared in metadata. The adapter will call external LLM providers if provided API keys are available — this is expected for a chatbot but the omission in declared env requirements is a transparency/regression issue and could lead to accidental exposure of keys if the agent supplies them.
Persistence & Privilege
Skill does not request always:true and does not modify other skills or system-wide agent settings. It stores/loads sessions and KB under paths provided by callers; that is normal for this kind of skill.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install chatbot-engine - 安装完成后，直接呼叫该 Skill 的名称或使用
/chatbot-engine触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
Initial release: Multi-turn dialogue and intent recognition engine
元数据
常见问题
Chatbot Engine 是什么？
智能对话引擎 - 多轮对话与意图识别 | Chatbot Engine - Multi-turn dialogue and intent recognition. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 214 次。
如何安装 Chatbot Engine？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install chatbot-engine」即可一键安装，无需额外配置。
Chatbot Engine 是免费的吗？
是的，Chatbot Engine 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Chatbot Engine 支持哪些平台？
Chatbot Engine 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Chatbot Engine？
由 Lv Lancer（@kaiyuelv）开发并维护，当前版本 v1.0.0。
推荐 Skills