• Agent
    智能体是具备自主感知、记忆、决策、交互与执行能力的智能系统，是人工智能产品及服务的重要形态。 -- 豆包
    AI 智能体是能够感知并响应其所处环境、采取行动以实现预设目标的自动化实体。

    五大核心能力界定：
    - 自主感知：可主动 / 被动从环境中获取并理解信息，无需人类逐次手动输入
    - 记忆：可存储、检索和复用历史信息，保证行为与上下文的连贯性
    - 决策：可自主拆解任务、规划执行路径、选择行动方案
    - 交互：可与人类、其他智能体、系统设备进行双向协同沟通
    - 执行：可调用外部工具、软件系统或硬件设备，实际改变环境状
    感知输入 → 记忆检索 → 决策生成 → 行动执行 → 反馈学习 -- CSDN AI助手

  • agent是强化版的prompt？agent 与 prompt 的区别
    - Agent = 大模型 + 记忆模块 + 工具集（Tool/Skills） + 规划引擎 + 安全机制 + 各类内置 Prompt 指令 -- 豆包
    - Prompt：一问一答；A跟他：一问+思考+确认+决策+执行+答案/结果
    - Agent = 高度自治、自驱的Prompt + 任务执行/实现
    - 个人理解：好的prompt（清晰 明确）是agent的基础，然而单纯的prompt不可能构成agent

  • 最简Agent实现

    API_URL = {
    "answer_new"                :  f"http://localhost:7091/api/answer",
    "models_getlist"            :  f"http://localhost:7091/api/models",
    "conversations_getlist"     :  f"http://localhost:7091/api/get_conversations",   
    "conversations_getsingle"   :  f"http://localhost:7091/api/get_single_conversation", 
    "conversations_rename"      :  f"http://localhost:7091/api/update_conversation_name",
    }

    1）感知输入 , select model + question ai 
         --> select Model --> creat new conversation --> rename conversation

    # check model list, to file "model_id"
def get_model():
    url = API_URL["models_getlist"]
    model_list = []
    response = requests.get(url)

    result = response.json()
    for model in result['models']:
        mn = f"Model_Name: {model['display_name']}, Model_Id: {model['id']}"
        model_list.append([mn])
    return model_list   

# creat new blan conversation without any question
def creat_new_chat():
    url = API_URL["answer_new"]
    payload = {
      "question": f"",
##        "history": chat_history,
##        "conversation_id": "6a2524f29404e7c57b19e98b",          
##        "prompt_id": "",
##        "chunks": 2,
##        "retriever": "",
##        "api_key": "string",
##        "agent_id": "string",
##        "active_docs": "",
##        "isNoneDoc": True,
##        "save_conversation": True,
##        "visibility": "hidden",
      "model_id": "docsgpt-local",
##        "passthrough": passthrough,
##        "temperature": 0.0,           
##        "top_k": 5,
    }
    
    response = requests.post(url, json=payload)

# get conversations list
def get_conversations():
    url = API_URL["conversations_getlist"]
    conversation_list = []
    response = requests.get(url)

    result = response.json()
##    print(f"--> Conversation: {response}\n{result}")  
    for conversation in result:
        cn = {"Conversaton_Name": f"{conversation['name']}", "Conversaton_Id": f"{conversation['id']}"}
        conversation_list.append(cn)
    print(f"--> {conversation_list}")
    return conversation_list

# get single conversation, for creat_new_chat() used
def get_single_conversation(covid):
    url = API_URL["conversations_getsingle"]
    response = requests.get(f"{url}?id={covid}")

    result = response.json()
    
    return result

# rename conversion name, make sure show clear
def rename_chat(chat_name):
    url = API_URL["conversations_rename"]

    conversation_list = get_conversation()

    covid = conversation_list[0]['Conversaton_Id']

    payload = {
      "id": covid,
      "name": chat_name,    
        }
    
    requests.post(url, json=payload)
    
    return covid

    2）记忆检索 , history + compress
          --> store user question + assitant(system reply) --> compress store减少上下文容量（减少token使用，减少理解偏差）

    # ask question, get history
def chat(prompt):
    url = API_URL["answer_new"]
    
    question = prompt.strip()
    chat_history = []
    passthrough = {
        "Time": time.ctime(),
        "sysprompt": "以中文回复以下问题：",   # 分析以下问题，如问题不明确则列出不明确之处，如问题明确则输出结果
    }        
        
    payload = {
      "question": f"{passthrough['sysprompt']}{question }",
      "history": chat_history,
      "conversation_id": "6a39310ac2149c3dbc243a04",          
##        "prompt_id": "",
      "chunks": 2,
      "retriever": "",
##        "api_key": "string",
##        "agent_id": "string",
      "active_docs": "",
      "isNoneDoc": True,
      "save_conversation": True,
##          "visibility": "hidden",
      "model_id": "docsgpt-local",
      "passthrough": passthrough,
      "temperature": 0.0,           
      "top_k": 5,
    }

    chat_history.append(
             {"role":    "user", 
              "content": question}
             )
    
    # call AI Answer
    response = requests.post(url, json=payload)

    if response.status_code == 200:
        result = response.json()
        # add question reply into history
        chat_history.append({"role": "assistant", "content": result['answer']})
        return result['answer']
    else:
        return f"{response.status_code}, Error"

# compress conversation history, keep context
def compress(chat_history : list):
    message_json = json.dumps(chat_history, ensure_ascii=False)
##    if len(message_json) <= 10000:
##        return messages
    prompt = f"压缩以下对话历史，保留核心信息和关键事实，去除冗余，不能超过100个字符：\n{message_json}"

    result = chat(prompt)
    
    return result

# loop conversation, build basic agent construction
def chat_loop():
    url = API_URL["answer_new"]
    
    while True:  # while loop, different for chat()
        question = input("\nPlease input your question：").strip()
        chat_history = []
        passthrough = {
            "Time": time.ctime(),
            "sysprompt": "以中文回复以下问题：",   # 分析以下问题，如问题不明确则列出不明确之处，如问题明确则输出结果
        }        
            
        payload = {
          "question": f"{passthrough['sysprompt']}{question }",
          "history": chat_history,
          "conversation_id": "6a39310ac2149c3dbc243a04",          
##        "prompt_id": "",
          "chunks": 2,
          "retriever": "",
##        "api_key": "string",
##        "agent_id": "string",
          "active_docs": "",
          "isNoneDoc": True,
          "save_conversation": True,
##          "visibility": "hidden",
          "model_id": "docsgpt-local",
          "passthrough": passthrough,
          "temperature": 0.0,           
          "top_k": 5,
        }

        chat_history.append(
                 {"role":    "user", 
                  "content": question}
                 )
        
        # quit conversation
##        if user_input == "/quit":
##            print("End Agent >>")
##            break

        
        # call AI Answer
        response = requests.post(url, json=payload)

        if response.status_code == 200:
            result = response.json()
            # add question reply into history
            chat_history.append({"role": "assistant", "content": result['answer']})
            return result['answer']
        else:
            return f"{response.status_code}, Error"

    3）决策生成 , think + select tool
         - 生成tool --> 按格式生成

    ---
name: read-json-file
description: |
读取json文件时使用
tags： read, json
适用：json文件读取
license: rz01
metadata:
  author: roy zhu
  version: "0.1"
---

## 技能概述
读取json格式文件时使用。例如：
- 读取json文件
- 提取json文件内容

## 执行步骤
1. 打开json文件
2. 读取json文件
3. 分析json文件内容
4. 关闭json文件

## 注意事项
- 读取异常给出提示

## Examples

         - 选择/执行tool --> 路由，参考 (4 封私信 / 30 条消息) 第六章：手搓 Skill 系统 —— 能力模块化与编排 - 知乎 或 LLM大模型源码

    # 文件目录
│  skills-map.json    # skill map list
│
└─read-json-file      # skill: read-json-file
    │  changelog.md
    │  metadata.json
    │  readme.md
    │  SKILL.md       # core for skill description
    │
    ├─assets
    ├─examples
    │      skills.md
    │
    ├─references
    ├─scripts
    │      read_json_file.py
    │
    └─tests

1) SKILL.md --> read-json-file
2) skills-map.json
{
	"read-json-file": "read json file，读取json文件"
}

3) python: read json file
    def read_json_file(file_name:str) -> str:
        """读json文件"""
        try:
            with open(file_name, "r", encoding="utf-8") as f:
                data = json.load(f)
                return data
    ##            return f.read()
        except Exception as e:
            return f"fail：{str(e)}"

4-1) tool(skill) route: get and select skill --> 此处仅为模拟方式，通过skill关键字与prompt及answer的包含关系确定skill的选择
    def get_skills():
        skills = read_json_file("./skills/skills-map.json")

        def get_skill_name():
            return list(skills.keys)

        def get_skill_values()->list:
            return list(skills.values)

        return skills
4-1) 选中skill，读入skill.md内容，执行选中skill操作
    def sel_skill(query):
        for skill in get_skills():
            if skill in query or get_skills()[skill] in query or query in skill or query in get_skills()[skill]:
                return read_md_file(f"./skills/{skill}/skill.md")
            else:
                return "cannot get skill."

    4）反馈学习, chat loop

    1) 循环对话
   - 使用while循环，当出现特定指令，如 /quit 时通出
   - 使用LLM工具所提供的函数，调用指定的 chat(conversation) ID持续对话
2) 压缩上下文，存入 system prompt / user prompt / assisttan promt
    chat_history = []

    chat_history.append(
        {"role":    "system",
         "content": sysprmpt}
                 )  

    chat_history.append(
        {"role":    "user",
         "content": question}
                 )

    chat_history.append(
        {"role":    "assistand",
         "content": answer}
                 )

def compress(chat_history : list):
    message_json = json.dumps(chat_history, ensure_ascii=False)
##    if len(message_json) <= 10000:
##        return messages
    prompt = f"压缩以下对话历史，保留核心信息和关键事实，去除冗余，不能超过100个字符：\n{message_json}"

    result = chat(prompt)  # get chat reply: chat_history
    
    return result
  • DocsGPT Agent
    - 最简单省事的方式，使用大模型工具提供的agent 和 skill

参考：

# AIGC # 人工智能
Logo
亚马逊云科技技术品牌专区

更多推荐

cover

2025-2026 CSDN年度技术趋势预测：AI、云原生与开发者工具演进

avatar 亚马逊云科技技术品牌专区
cover

【硬件+APP+云平台】53.1.智能投喂器-V2-硬件定时-基于STM32嵌入式物联网单片机软硬件毕业生系统设计

avatar 亚马逊云科技技术品牌专区
cover

人工智能训练师-考试大纲深度解读+分值分布

avatar 亚马逊云科技技术品牌专区