Update doc research readme (#717)

Author: wangxingjun778

projects/doc_research/README.md

@@ -27,6 +27,7 @@
   - 🚀 **High Efficiency** - Leverage powerful LLMs for fast and accurate research, leveraging key information extraction techniques to further optimize token usage
   - ⚙️ **Flexible Deployment** - Support local run and [ModelScope Studio](https://modelscope.cn/studios) on both CPU and GPU environments.
   - 💰 **Free Model Inference** - Free LLM API inference calls for ModelScope users, refer to [ModelScope API-Inference](https://modelscope.cn/docs/model-service/API-Inference/intro)
+  - 📚 **Export & Upload** - Support exporting research reports to PDF, PPTX, DOCX, HTML formats; support uploading reports to ModelScope, HuggingFace, GitHub
 
 
 <br>
@@ -102,6 +103,84 @@ ms-agent app --doc_research \
 > When running locally, the default address is http://0.0.0.0:7860/. If the page can't be accessed, try disabling proxy.
 
 
+### 4. Report Export
+DocResearch supports exporting markdown reports to various formats for easy archiving and sharing:
+- PDF
+- PPTX
+- HTML
+- DOCX
+
+**Examples of exporting using Python script:**
+```python
+from ms_agent.utils.markdown_converter import MarkdownConverter
+
+report_path: str = '/path/to/report_dir'
+output_path: str = '/path/to/output_dir'
+
+# Call export functions
+MarkdownConverter.markdown_to_html(report_path, output_path + "/html")
+MarkdownConverter.markdown_to_docx(report_path, output_path + "/docx")
+MarkdownConverter.markdown_to_pptx(report_path, output_path + "/pptx")
+MarkdownConverter.markdown_to_pdf(report_path, output_path + "/pdf")
+```
+
+
+### 5. Report Uploading & Sharing
+DocResearch supports uploading generated markdown research reports to multiple platforms for easy sharing and collaboration:
+- **ModelScope**
+- **HuggingFace**
+- **GitHub**
+
+**Upload to ModelScope:**
+```python
+from ms_agent.utils.push_to_hub import PushToModelScope
+
+# Get the ModelScope token： https://modelscope.cn/my/myaccesstoken
+push_to_ms = PushToModelScope(token='ms-xxx')
+
+push_to_ms.push(
+    repo_id='your-username/your-repo-name',
+    folder_path='/path/to/report_dir',
+    path_in_repo='report',
+    commit_message='Upload research report to ModelScope',
+)
+```
+
+**Upload to HuggingFace:**
+```python
+from ms_agent.utils.push_to_hub import PushToHuggingFace
+
+# Get the HuggingFace token： https://huggingface.co/settings/tokens
+push_to_hf = PushToHuggingFace(token='hf_xxx')
+
+push_to_hf.push(
+    repo_id='your-username/your-repo-name',
+    folder_path='/path/to/report_dir',
+    path_in_repo='report',
+    commit_message='Upload research report to HuggingFace',
+)
+```
+
+**Upload to GitHub:**
+```python
+from ms_agent.utils.push_to_hub import PushToGitHub
+
+# Get the GitHub PAT(Personal Access Tokens): https://github.com/settings/tokens
+push_to_git = PushToGitHub(
+    user_name = 'your-username',
+    repo_name = 'your-repo-name',
+    token = 'xxx',
+    visibility = 'public',  # `public` or `private`
+    description = 'A repository for research reports generated by MS-Agent DocResearch',)
+
+push_to_git.push(
+    folder_path='/path/to/report_dir',
+    path_in_repo='report',
+    commit_message='Upload research report to GitHub',
+)
+```
+
+
 <br>
 
 ## Usage Instructions

projects/doc_research/README_zh.md

@@ -27,6 +27,7 @@
 - 🚀 **精准高效** - 利用强大的LLM进行快速准确的研究，采用关键信息抽取技术进一步优化了token使用
 - ⚙️ **灵活部署** - 支持本地运行和魔搭创空间运行模式（CPU-Only），同时也兼容GPU环境
 - 💰 **免费模型推理** - 魔搭ModelScope用户可免费调用LLM API推理，参考 [ModelScope API-Inference](https://modelscope.cn/docs/model-service/API-Inference/intro)
+- 📚 **导出与上传** - 支持将研究报告导出为PDF、PPTX、DOCX、HTML格式；支持将报告上传至ModelScope、HuggingFace、GitHub
 
 
 <br>
@@ -104,9 +105,86 @@ ms-agent app --doc_research \
   > 本地运行时，默认访问地址为 `http://0.0.0.0:7860/` ，如无法访问，可尝试关闭VPN  <br>
 
 
+### 4. 报告导出
+DocResearch支持将markdown格式报告导出为多种格式，方便用户进行存档和分享：
+- PDF
+- PPTX
+- HTML
+- DOCX
+
+**使用Python脚本导出示例：**
+```python
+from ms_agent.utils.markdown_converter import MarkdownConverter
+
+report_path: str = '/path/to/report_dir'
+output_path: str = '/path/to/output_dir'
+
+# Call export functions
+MarkdownConverter.markdown_to_html(report_path, output_path + "/html")
+MarkdownConverter.markdown_to_docx(report_path, output_path + "/docx")
+MarkdownConverter.markdown_to_pptx(report_path, output_path + "/pptx")
+MarkdownConverter.markdown_to_pdf(report_path, output_path + "/pdf")
+```
+
+### 5. 报告上传分享
+DocResearch支持将生成的markdown格式研究报告上传至多个平台，方便用户进行分享和协作：
+- **ModelScope**
+- **HuggingFace**
+- **GitHub**
+
+**上传到ModelScope：**
+```python
+from ms_agent.utils.push_to_hub import PushToModelScope
+
+# ModelScope令牌获取： https://modelscope.cn/my/myaccesstoken
+push_to_ms = PushToModelScope(token='ms-xxx')
+
+push_to_ms.push(
+    repo_id='your-username/your-repo-name',
+    folder_path='/path/to/report_dir',
+    path_in_repo='report',
+    commit_message='Upload research report to ModelScope',
+)
+```
+
+**上传到HuggingFace：**
+```python
+from ms_agent.utils.push_to_hub import PushToHuggingFace
+
+# HuggingFace令牌获取： https://huggingface.co/settings/tokens
+push_to_hf = PushToHuggingFace(token='hf_xxx')
+
+push_to_hf.push(
+    repo_id='your-username/your-repo-name',
+    folder_path='/path/to/report_dir',
+    path_in_repo='report',
+    commit_message='Upload research report to HuggingFace',
+)
+```
+
+**上传到GitHub：**
+```python
+from ms_agent.utils.push_to_hub import PushToGitHub
+
+# 获取GitHub PAT(Personal Access Tokens): https://github.com/settings/tokens
+push_to_git = PushToGitHub(
+    user_name = 'your-username',
+    repo_name = 'your-repo-name',
+    token = 'xxx',
+    visibility = 'public',  # `public` or `private`
+    description = 'A repository for research reports generated by MS-Agent DocResearch',)
+
+push_to_git.push(
+    folder_path='/path/to/report_dir',
+    path_in_repo='report',
+    commit_message='Upload research report to GitHub',
+)
+```
+
+
 <br>
 
-## 使用说明
+## Gradio UI使用说明
 
 1. **用户提示**：在文本框中输入您的研究目标或问题
 2. **文件上传**：选择需要分析的文件（支持多选）

requirements/research.txt

@@ -5,6 +5,7 @@ exa-py
 google-search-results
 gradio>=5.0.0
 json5
+markdown
 mcp
 modelscope
 openai