Merge pull request #106 from ant-xuexiao/feat_chat_create_bot

Author: RaoHai

lui/src/services/ChatController.ts

@@ -8,7 +8,7 @@ export async function streamChat(
   messages: IPrompt[],
   host = 'http://127.0.0.1:8000',
 ): Promise<Response> {
-  return fetch(`${host}/api/chat/qa`, {
+  return fetch(`${host}/api/chat/builder`, {
     method: 'POST',
     headers: {
       'Content-Type': 'application/json',

server/agent/bot_builder.py

@@ -1,18 +1,19 @@
 from typing import AsyncIterator
 from data_class import ChatData
 from agent.base import AgentBuilder
+from tools import bot_builder
 
 PROMPT = """
 # 角色
-你是一 GitHub 答疑机器创建助手。你擅长根据用户提供的 Github 仓库信息创建一个答疑机器人。
+你是一名 GitHub 答疑机器创建助手。你擅长根据用户提供的 Github 仓库信息创建一个答疑机器人。
 
 ## 技能
-### 技能1：获取并确认仓库信息
-- 引导用户提供他们的GitHub仓库信息。
-- 根据提供的信息确认这个仓库存在并且可以访问。
+### 技能1：获取 GitHub 仓库名
+- 引导用户提供他们的 GitHub 仓库名或地址。
+- 根据提供的 GitHub 地址提取 GitHub 仓库名。
 
 ### 技能2：创建答疑机器人
-- 使用bot_builder工具根据用户提供的Github仓库信息创建机器人。
+- 使用 bot_builder 工具根据用户提供的 Github 仓库名创建机器人。
 
 ### 技能3：修改机器人的配置
 - 根据用户的描述进行机器人的配置信息修改。
@@ -24,8 +25,10 @@
 """
 
 
-TOOL_MAPPING = {}
+TOOL_MAPPING = {
+    "bot_builder": bot_builder.create_bot,
+}
 
 def agent_chat(input_data: ChatData) -> AsyncIterator[str]:
-    agent = AgentBuilder(prompt=PROMPT, tools={}, enable_tavily=False)
+    agent = AgentBuilder(prompt=PROMPT, tools=TOOL_MAPPING, enable_tavily=False)
     return agent.run_chat(input_data)

server/agent/qa_chat.py

@@ -1,35 +1,10 @@
 from typing import AsyncIterator
 from data_class import ChatData
 from agent.base import AgentBuilder
+from prompts.bot_template import generate_prompt_by_repo_name
 from tools import issue, sourcecode, knowledge
 
 
-PROMPT = """
-# Character
-You are a skilled assistant dedicated to Ant Design, capable of delivering comprehensive insights and solutions pertaining to Ant Design. You excel in fixing code issues correlated with Ant Design.
-
-## Skills
-### Skill 1: Engaging Interaction
-Your primary role involves engaging with users, offering them in-depth responses to their Ant Design inquiries in a conversational fashion.
-
-### Skill 2: Insightful Information Search
-For queries that touch upon unfamiliar zones, you are equipped with two powerful knowledge lookup tools, used to gather necessary details:
-   - search_knowledge: This is your initial resource for queries concerning ambiguous topics about Ant Design. While using this, ensure to retain the user's original query language for the highest accuracy possible. Therefore, a specific question like 'Ant Design的新特性是什么?' should be searched as 'Ant Design的新特性是什么?'.
-   - tavily_search_results_json: Should search_knowledge fail to accommodate the required facts, this tool would be the next step.
-
-### Skill 3: Expert Issue Solver
-In case of specific issues reported by users, you are to aid them using a selection of bespoke tools, curated as per the issue nature and prescribed steps. The common instances cater to:
-   - Routine engagement with the user.
-   - Employment of certain tools such as create_issue, get_issues, search_issues, search_code etc. when the user is facing a specific hurdle.
-
-## Constraints:
-- Maintain a strict focus on Ant Design in your responses; if confronted with unrelated queries, politely notify the user of your confines and steer them towards asking questions relevant to Ant Design.
-- Your tool utilization choices should be driven by the nature of the inquiry and recommended actions.
-- While operating tools for searching information, keep the user's original language to attain utmost precision.
-- With your multilingual capability, always respond in the user's language. If the inquiry popped is in English, your response should mirror that; same goes for Chinese or any other language.
-"""
-
-
 TOOL_MAPPING = {
     "search_knowledge": knowledge.search_knowledge,
     "create_issue": issue.create_issue,
@@ -39,5 +14,6 @@
 }
 
 def agent_chat(input_data: ChatData) -> AsyncIterator[str]:
-    agent = AgentBuilder(prompt=PROMPT, tools=TOOL_MAPPING)
-    return agent.run_chat(input_data)
+   prompt = generate_prompt_by_repo_name("ant-design")
+   agent = AgentBuilder(prompt=prompt, tools=TOOL_MAPPING)
+   return agent.run_chat(input_data)

server/prompts/bot_builder.py

@@ -0,0 +1,30 @@
+
+
+
+PROMPT = """
+# Character
+You are a skilled assistant dedicated to {repo_name}, capable of delivering comprehensive insights and solutions pertaining to {repo_name}. You excel in fixing code issues correlated with {repo_name}.
+
+## Skills
+### Skill 1: Engaging Interaction
+Your primary role involves engaging with users, offering them in-depth responses to their {repo_name} inquiries in a conversational fashion.
+
+### Skill 2: Insightful Information Search
+For queries that touch upon unfamiliar zones, you are equipped with two powerful knowledge lookup tools, used to gather necessary details:
+   - search_knowledge: This is your initial resource for queries concerning ambiguous topics about {repo_name}. While using this, ensure to retain the user's original query language for the highest accuracy possible. Therefore, a specific question like '{repo_name} 的新特性是什么?' should be searched as '{repo_name} 的新特性是什么?'.
+   - tavily_search_results_json: Should search_knowledge fail to accommodate the required facts, this tool would be the next step.
+
+### Skill 3: Expert Issue Solver
+In case of specific issues reported by users, you are to aid them using a selection of bespoke tools, curated as per the issue nature and prescribed steps. The common instances cater to:
+   - Routine engagement with the user.
+   - Employment of certain tools such as create_issue, get_issues, search_issues, search_code etc. when the user is facing a specific hurdle.
+
+## Constraints:
+- Maintain a strict focus on {repo_name} in your responses; if confronted with unrelated queries, politely notify the user of your confines and steer them towards asking questions relevant to {repo_name}.
+- Your tool utilization choices should be driven by the nature of the inquiry and recommended actions.
+- While operating tools for searching information, keep the user's original language to attain utmost precision.
+- With your multilingual capability, always respond in the user's language. If the inquiry popped is in English, your response should mirror that; same goes for Chinese or any other language.
+"""
+
+def generate_prompt_by_repo_name(repo_name: str):
+    return PROMPT.format(repo_name=repo_name)

server/prompts/bot_generator.py

@@ -0,0 +1,46 @@
+from langchain.tools import tool
+from github import Github
+from db.supabase.client import get_client
+from prompts.bot_template import generate_prompt_by_repo_name
+
+g = Github()
+
+@tool
+def create_bot(
+    repo_name: str,
+    starters: list[str] = ["介绍一下项目", "快速上手", "贡献指南"],
+):
+    """
+    create a bot based on the given github repository.
+
+    :param repo_name: The name of the repository, e.g., "octocat/Hello-World"
+    :param starters: The Opening Dialog, e.g.["介绍一下项目", "快速上手", "贡献指南"]
+    """
+    try:
+        # Step1:Get the repository object
+        repo = g.get_repo(repo_name)
+
+        # Step2: Generate the prompt
+        prompt = generate_prompt_by_repo_name(repo_name)
+        
+        # Step3: Create bot based on the prompt
+        bot_data = {
+          "name":  repo.name,
+          "description": repo.description,
+          "avatar": repo.organization.avatar_url if repo.organization else None,
+          "prompt": prompt,
+          "uid": "u123456", # TODO get from auth
+          "enable_img_generation": False,
+          "label": "Assistant",
+          "starters": starters,
+          "enable_img_generation": False,
+          "public": False,
+        }
+        print('bot_data', bot_data)
+        supabase = get_client()
+        response = supabase.table("bots").insert(bot_data).execute()
+        return response
+    except Exception as e:
+        print(f"An error occurred: {e}")
+        return e
+

server/prompts/bot_template.py

@@ -54,8 +54,8 @@ def get_issues(
 
             issues_list = [
                 {
-                    'issue_name': f'Issue #{issue.number} - {issue.title}',
-                    'issue_url': issue.html_url
+                    "issue_name": f"Issue #{issue.number} - {issue.title}",
+                    "issue_url": issue.html_url
                 }
                 for issue in issues
             ]

server/tools/bot_builder.py

@@ -4,7 +4,7 @@
 
 @tool
 def search_knowledge(
-      query: str,
+    query: str,
 ):
     """
     Search for information based on the query.  When use this tool, do not translate the search query. Use the original query language to search. eg: When user's question is 'Ant Design 有哪些新特性?', the query should be 'Ant Design 有哪些新特性?'.

server/tools/issue.py

@@ -12,7 +12,7 @@ class BotCreateRequest(BaseModel):
     name: Optional[str] = None
     starters: Optional[List[str]] = None  # 假设为起始文本列表
     voice: Optional[str] = None
-    public: Optional[bool] = None
+    public: Optional[bool] = False
 class BotUpdateRequest(BaseModel):
     avatar: Optional[str] = None
     description: Optional[str] = None
@@ -23,4 +23,4 @@ class BotUpdateRequest(BaseModel):
     name: Optional[str] = None
     starters: Optional[List[str]] = None
     voice: Optional[str] = None
-    public: Optional[bool] = None
+    public: Optional[bool] = None