What is Shuyi Qingzhou Agent
Shuyi Qingzhou Agent is a localized AI agent focused on Excel data processing. Users do not need to memorize complex function formulas; they can simply describe their needs in natural language, and the agent will automatically complete operations such as data cleaning, format standardization, dynamic grouping and summarization, cross-sheet association, and batch file processing. The product is fully compatible with local large model deployment, with all data running in a local closed loop throughout the process and zero upload to the cloud, meeting the compliance requirements of heavily regulated industries such as finance, healthcare, and government affairs.
Main features of Shuyi Qingzhou Agent
-
Natural language-driven data processing: Users do not need to memorize complex function formulas and can directly describe their needs in colloquial terms, such as "merge the names in column A and the phone numbers in column B into one column, separated by a Chinese enumeration comma." The Agent automatically parses the intent and generates an execution plan.
-
Intelligent data cleaning and format standardization: Automatically identifies common issues such as inconsistent date formats, mixed text and numbers, and empty value filling, supports batch normalization processing, and provides a before-and-after comparison preview of cleaning.
-
Dynamic grouping, summarization, and pivot analysis: Automatically generates grouped statistics based on semantic understanding, such as "calculate sales by region and their proportions," without manually dragging pivot table fields.
-
Cross-sheet association and data merging: Supports semantic expression of association logic such as VLOOKUP and XLOOKUP, automatically matches key fields, and handles multi-sheet data integration scenarios.
-
Batch file processing: Supports executing the same operation process on multiple Excel files in a folder, such as unifying formats, extracting specific fields, and merging and summarizing, suitable for periodic report processing.
How to use Shuyi Qingzhou Agent
-
Download and install: Visit the Shuyi Qingzhou Agent official website https://www.syqzai.com/,下载 Windows desktop installer, and complete the local deployment by following the guide. Ordinary administrative personnel can independently complete the installation under the guidance of technical documentation.
-
Configure a local large model: Connect mainstream local large models such as Qwen, DeepSeek, and Kimi. All data runs in a local closed-loop environment throughout the process, with no need to upload to the cloud and support for offline use.
-
Open the Excel file: Directly import the Excel spreadsheet to be processed. The Agent automatically identifies the table structure, field types, and data characteristics.
-
Input requirements in natural language: Describe operation instructions in colloquial terms in the dialog box, such as "merge the names in column A and the phone numbers in column B into one column separated by a Chinese enumeration comma" or "calculate sales by region and their proportions," without needing to memorize function formulas.
-
Preview and confirm execution: The Agent automatically generates an execution plan and displays a before-and-after comparison preview of cleaning. After confirmation, complete data cleaning, format standardization, pivot analysis, or cross-sheet association with one click.
-
Batch processing (optional): Select multiple Excel files in a folder and execute unified format conversion, field extraction, or merging and summarization, suitable for batch processing of periodic reports.
-
Save instruction templates: Save commonly used operation processes as templates. The next time you encounter similar tasks, such as HR attendance statistics or monthly financial reconciliation, you can directly reuse them with one click to automate periodic work.
-
Export results: After processing is complete, directly export a standard-format Excel file, or generate a Word analysis report with charts and formatting in one click, for work reporting or system import.
Core Advantages of 数以轻舟Agent
-
Zero-threshold operation with natural language: No formulas to write and no syntax to memorize throughout the process. Use conversational descriptions, and the Agent automatically parses intent and executes, enabling even Excel beginners to complete professional-grade data processing.
-
Local deployment and data sovereignty: Fully compatible with local large models such as Qwen, DeepSeek, and Kimi. Excel files and business data run entirely in a local closed loop throughout the process, with zero outbound network requests, meeting the compliance requirements of highly regulated industries such as finance, healthcare, and government affairs.
-
Intelligent cleaning and format standardization: Automatically identifies common issues such as chaotic date formats, mixed text and numbers, and null value filling. Supports batch normalization processing and provides before-and-after cleaning comparison previews, saving manual troubleshooting time.
-
Semantic pivot and grouped summarization: Automatically generates grouped statistics based on natural language semantic understanding, with no need to manually drag pivot table fields, completing multidimensional analysis in one click.
-
Semantic expression of cross-table relationships: Supports natural language descriptions of complex relationship logic such as VLOOKUP and XLOOKUP, automatically matches key fields, and easily handles multi-table data integration and merge/split scenarios.
-
Efficient batch file processing: Supports executing the same operation workflow on multiple Excel files within a folder, such as unifying formats, extracting specific fields, and merging and summarizing. Suitable for batch processing of periodic reports such as monthly and quarterly reports.
-
One-click reuse of instruction templates: Commonly used operation workflows can be saved as templates. Periodic tasks such as HR attendance, financial reconciliation, and sales statistics can be called directly next time, enabling automated pipelines.
Comparison of 数以轻舟Agent with Similar Competing Products
| Comparison Dimension | 数以轻舟Agent | ChatExcel | WPS AI |
|---|---|---|---|
| Development Team | Beijing Qianze Shuzhi Technology Co., Ltd. | Peking University Team | Kingsoft Office (Kingsoft) |
| Core Positioning | Localized Excel data processing Agent | Online AI spreadsheet processing and data analysis platform | AI assistant built into the WPS office suite |
| Deployment Method | Local installation on the Windows desktop, supports offline operation | Online use on the web, requires uploading files to the cloud | Integrated into the WPS Office client, some features require an internet connection |
| Data Security | Data runs in a local closed loop throughout the process, with zero outbound network requests, meeting finance/healthcare/government compliance requirements | Files need to be uploaded to the cloud for processing, relying on the platform's data security policy | Data flows between WPS cloud services and local devices, and the enterprise edition has security hardening |
| Natural Language Operation | Supported; conversational descriptions can generate complex formulas and pivot analysis | Supported; deep Chinese semantic adaptation, completing cleaning/summarization/charts in one sentence | Supported; enter instructions within the spreadsheet through the sidebar or dialog box |
| Local Large Models | Fully compatible with local models such as Qwen, DeepSeek, and Kimi | Mainly relies on cloud large models, with limited localization capabilities | Relies on Kingsoft's self-developed or partnered cloud models, does not support third-party local model integration |
| Batch file processing | Supports batch execution of the same workflow across multiple Excel files in a folder | Supports multi-sheet collaborative processing, but the degree of batch automation is limited | Supports batch format conversion and basic processing; complex batch scenarios require manual operations |
| Cross-sheet association | Semantic expression of VLOOKUP/XLOOKUP, automatically matching key fields | Supports cross-sheet collaborative processing, data merging and comparison | Supports multi-sheet data referencing and basic association; complex associations require manual formulas |
| Instruction template reuse | Supports saving commonly used instruction templates for one-click reuse of periodic tasks | Supports conversation history and partial operation reuse | Supports quick invocation of common functions; in-depth custom templates are limited |
Application scenarios of Shuyi Qingzhou Agent
- Customer list organization: Customer information collected by the sales department from various channels has inconsistent formats. Through natural language instructions, deduplication, completion, and format unification are quickly completed, generating a standard customer table that can be imported into CRM.
- Quick quotation generation: Input basic product data and pricing rules, automatically generate dynamic quotations with formula calculations, and support batch replacement of customer information to output multiple independent files.
- Financial report preprocessing: Perform account classification, monthly summary, and abnormal value marking on exported raw financial data, generating an analysis base table readable by management.
- Batch processing of HR data: Periodic Excel operations such as onboarding information forms, attendance statistics, and salary calculation, with one-click reuse achieved by saving commonly used instruction templates.
- Internal analysis of sensitive data: Analysis scenarios involving customer privacy, trade secrets, and undisclosed financial data, running locally throughout the entire process, meeting the enterprise's requirements for data sovereignty.