up

README.md

@@ -48,8 +48,32 @@ Welcome to MetaChain! MetaChain is a **Fully-Automated** and highly **Self-Devel
     </ul>
 </div>
 
+<span id='table-of-contents'/>
+
+## 📑 Table of Contents
+
+* <a href='#features'>✨ Features</a>
+* <a href='#news'>🔥 News</a>
+* <a href='#quick-start'>⚡ Quick Start</a>
+  * <a href='#installation'>Installation</a>
+  * <a href='#api-keys-setup'>API Keys Setup</a>
+  * <a href='#start-with-cli-mode'>Start with CLI Mode</a>
+* <a href='#how-to-use'>🔍 How to Use MetaChain</a>
+  * <a href='#user-mode'>1. User Mode (SOTA Open Deep Research)</a>
+  * <a href='#agent-editor'>2. Agent Editor</a>
+  * <a href='#workflow-editor'>3. Workflow Editor</a>
+* <a href='#todo'>☑️ Todo List</a>
+* <a href='#documentation'>📖 Documentation</a>
+* <a href='#community'>🤝 How to Join the Community</a>
+* <a href='#acknowledgements'>🙏 Acknowledgements</a>
+* <a href='#cite'>🌟 Cite</a>
+
+<span id='quick-start'/>
+
 ## ⚡ Quick Start
 
+<span id='installation'/>
+
 ### Installation
 
 #### MetaChain Installation
@@ -68,6 +92,8 @@ We use Docker to containerize the agent-interactive environment. So please insta
 docker pull tjbtech1/metachain:latest
 ```
 
+<span id='api-keys-setup'/>
+
 ### API Keys Setup
 
 Create a environment variable file, just like `.env.template`, and set the API keys for the LLMs you want to use. Not every LLM API Key is required, use what you need.
@@ -86,6 +112,8 @@ GROQ_API_KEY=
 XAI_API_KEY=
 ```
 
+<span id='start-with-cli-mode'/>
+
 ### Start with CLI Mode
 Just run the following command to start the CLI mode. (use shell script `cd path/to/MetaChain && sh playground/cli/metachain_cli.sh`)
 
@@ -119,8 +147,12 @@ After the CLI mode is started, you can see the start page of MetaChain:
 
 More features coming soon! 🚀 **Web GUI interface** under development.
 
+<span id='how-to-use'/>
+
 ## 🔍 How to Use MetaChain
 
+<span id='user-mode'/>
+
 ### 1. `user mode` (SOTA 🏆 Open Deep Research)
 
 MetaChain have a out-of-the-box multi-agent system, which you could choose `user mode` in the start page to use it. This multi-agent system is a general AI assistant, having the same functionality with **OpenAI's Deep Research** and the comparable performance with it in [GAIA](https://gaia-benchmark-leaderboard.hf.space/) benchmark. 
@@ -129,6 +161,7 @@ MetaChain have a out-of-the-box multi-agent system, which you could choose `user
 - 🔄 **Model Flexibility**: Compatible with any LLM (including Deepseek-R1, Grok, Gemini, etc.)
 - 💰 **Cost-Effective**: Open-source alternative to Deep Research's $200/month subscription
 - 🎯 **User-Friendly**: Easy-to-deploy CLI interface for seamless interaction
+- 📁 **File Support**: Handles file uploads for enhanced data interaction
 
 <table>
 <tr align="center">
@@ -149,6 +182,27 @@ MetaChain have a out-of-the-box multi-agent system, which you could choose `user
     </td>
 </tr>
 </table>
+<table>
+<tr align="center">
+    <td width="33%">
+        <img src="./assets/user_mode/4-upload.png" alt="Input" width="100%"/>
+        <br>
+        <em>@Upload_files will help you upload the files.</em>
+    </td>
+    <td width="33%">
+        <img src="./assets/user_mode/5-select.png" alt="Output" width="100%"/>
+        <br>
+        <em>Select the files you want to use.</em>
+    </td>
+    <td width="33%">
+        <img src="./assets/user_mode/6-success.png" alt="Select Agent" width="100%"/>
+        <br>
+        <em>Successfully uploaded the files.</em>
+    </td>
+</tr>
+</table>
+
+<span id='agent-editor'/>
 
 ### 2. `agent editor` (Agent Creation without Workflow)
 
@@ -195,6 +249,8 @@ You can use `agent editor` as shown in the following figure.
 </tr>
 </table>
 
+<span id='workflow-editor'/>
+
 ### 3. `workflow editor` (Agent Creation with Workflow)
 
 You can also create the agent workflows using natural language description with the `workflow editor` mode, as shown in the following figure. (Tips: this mode does not support tool creation temporarily.)
@@ -233,13 +289,28 @@ You can also create the agent workflows using natural language description with
 </tr>
 </table>
 
+<span id='todo'/>
+
 ## ☑️ Todo List
 
+MetaChain is continuously evolving! Here's what's coming:
 
+- 📊 **More Benchmarks**: Expanding evaluations to **SWE-bench**, **WebArena**, and more
+- 🖥️ **GUI Agent**: Supporting *Computer-Use* agents with GUI interaction
+- 🔧 **Tool Platforms**: Integration with more platforms like **Composio**
+- 🏗️ **Code Sandboxes**: Supporting additional environments like **E2B**
+- 🎨 **Web Interface**: Developing comprehensive GUI for better user experience
+
+Have ideas or suggestions? Feel free to open an issue! Stay tuned for more exciting updates! 🚀
+
+<span id='documentation'/>
 
 ## 📖 Documentation
 
 
+
+<span id='community'/>
+
 ## 🤝 How to Join the Community
 
 OpenHands is a community-driven project, and we welcome contributions from everyone. We do most of our communication
@@ -251,6 +322,7 @@ through Slack, so this is the best place to start, but we also are happy to have
 
 See more about the community in [COMMUNITY.md](./COMMUNITY.md) or find details on contributing in [CONTRIBUTING.md](./CONTRIBUTING.md).
 
+<span id='acknowledgements'/>
 
 ## 🙏 Acknowledgements
 
@@ -258,8 +330,15 @@ OpenHands is built by a large number of contributors, and every contribution is
 
 For a list of open source projects and licenses used in OpenHands, please see our [CREDITS.md](./CREDITS.md) file.
 
+<span id='cite'/>
+
 ## 🌟 Cite
 
 ```
 
 ```
+
+
+
+
+

playground/cli/file_select.py

@@ -0,0 +1,53 @@
+import tkinter as tk
+from tkinter import filedialog
+import shutil
+import os
+from rich.console import Console
+
+def select_and_copy_files(dest_dir, console: Console):
+    # 创建 tkinter 根窗口但隐藏它
+    root = tk.Tk()
+    root.withdraw()
+
+    # 打开文件选择对话框
+    files = filedialog.askopenfilenames(
+        title='Select files to copy',
+        filetypes=[
+            # ('Text files', '*.txt'),
+            ('All files', '*.*'),
+            ('PDF files', '*.pdf'),
+            ('Docx files', '*.docx'),
+            ('Txt files', '*.txt'),
+            ('Zip files', '*.zip'),
+            ('Text files', '*.txt'),
+        ]
+    )
+    
+    if not files:
+        print("No files selected")
+        return
+
+    # 选择目标文件夹
+    # dest_dir = filedialog.askdirectory(
+    #     title='Select destination folder'
+    # )
+    
+    if not dest_dir:
+        print("No destination folder selected")
+        return
+
+    # 复制文件
+    for file_path in files:
+        file_name = os.path.basename(file_path)
+        dest_path = os.path.join(dest_dir, file_name)
+        try:
+            shutil.copy2(file_path, dest_path)
+            console.print(f"[bold green]Uploaded: {file_name}[/bold green]")
+        except Exception as e:
+            console.print(f"[bold red]Error uploading {file_name}: {e}[/bold red]")
+
+    console.print(f"[bold green]Successfully uploaded {len(files)} files[/bold green]")
+
+if __name__ == "__main__":
+    dest_dir = "/Users/tangjiabin/Documents/reasoning/metachain/workspace_meta_showcase/showcase_nl2agent_showcase/workplace"
+    select_and_copy_files(dest_dir)

playground/cli/metachain_cli.py

@@ -34,6 +34,7 @@ from rich.panel import Panel
 import re
 from playground.cli.metachain_meta_agent import meta_agent
 from playground.cli.metachain_meta_workflow import meta_workflow
+from playground.cli.file_select import select_and_copy_files
 def get_args(): 
     parser = argparse.ArgumentParser(description="working@tjb-tech")
     parser.add_argument('--container_name', type=str, default='gpu_test')
@@ -107,6 +108,7 @@ def user_mode(model: str, context_variables: dict, debug: bool = True):
     agents = {system_triage_agent.name.replace(' ', '_'): system_triage_agent}
     for agent_name in system_triage_agent.agent_teams.keys():
         agents[agent_name.replace(' ', '_')] = system_triage_agent.agent_teams[agent_name]("placeholder").agent
+    agents["Upload_files"] = "select"
     style = Style.from_dict({
         'bottom-toolbar': 'bg:#333333 #ffffff',
     })
@@ -140,24 +142,35 @@ def user_mode(model: str, context_variables: dict, debug: bool = True):
                 # print(word, end=' ')
                 pass
         print()
-        agent_name = agent.name
-        console.print(f"[bold green][bold magenta]@{agent_name}[/bold magenta] will help you, be patient...[/bold green]")
-        messages.append({"role": "user", "content": query})
-        response = client.run(agent, messages, context_variables, debug=debug)
-        messages.extend(response.messages)
-        model_answer_raw = response.messages[-1]['content']
+        
+        if hasattr(agent, "name"): 
+            agent_name = agent.name
+            console.print(f"[bold green][bold magenta]@{agent_name}[/bold magenta] will help you, be patient...[/bold green]")
+            messages.append({"role": "user", "content": query})
+            response = client.run(agent, messages, context_variables, debug=debug)
+            messages.extend(response.messages)
+            model_answer_raw = response.messages[-1]['content']
 
-        # attempt to parse model_answer
-        if model_answer_raw.startswith('Case resolved'):
-            model_answer = re.findall(r'<solution>(.*?)</solution>', model_answer_raw, re.DOTALL)
-            if len(model_answer) == 0:
+            # attempt to parse model_answer
+            if model_answer_raw.startswith('Case resolved'):
+                model_answer = re.findall(r'<solution>(.*?)</solution>', model_answer_raw, re.DOTALL)
+                if len(model_answer) == 0:
+                    model_answer = model_answer_raw
+                else:
+                    model_answer = model_answer[0]
+            else: 
                 model_answer = model_answer_raw
-            else:
-                model_answer = model_answer[0]
+            console.print(f"[bold green][bold magenta]@{agent_name}[/bold magenta] has finished with the response:\n[/bold green] [bold blue]{model_answer}[/bold blue]")
+            agent = response.agent
+        elif agent == "select": 
+            code_env: DockerEnv = context_variables["code_env"]
+            local_workplace = code_env.local_workplace
+            files_dir = os.path.join(local_workplace, "files")
+            os.makedirs(files_dir, exist_ok=True)
+            select_and_copy_files(files_dir, console)
         else: 
-            model_answer = model_answer_raw
-        console.print(f"[bold green][bold magenta]@{agent_name}[/bold magenta] has finished with the response:\n[/bold green] [bold blue]{model_answer}[/bold blue]")
-        agent = response.agent
+            console.print(f"[bold red]Unknown agent: {agent}[/bold red]")
+
 
 def tool_to_table(tool_dict: dict):
     table = Table(title="Tool List", show_lines=True)