tencent cloud

Tencent Cloud Agent Development Platform

Release Notes and Announcements
Release Notes
Product Announcement
Product Introduction
Product Overview
Advantages
Use Cases
Model Introduction
Purchase Guide
Package Subscription
Previous Version
Getting Started
Agent Application and Its Three Modes
Creating a "Content Summary Assistant" in Standard Mode
Creating a “Webpage Scraping Assistant” in Single Workflow Mode
Creating a “Stand-up Comedy Content Creation Assistant” in Multi-Agent Mode
Operation Guide
Application Development
Workflow
Multi-Agent
Knowledge Base
Widget
Plugin Marketplace
Model List 
Prompt Templates
Application Templates
Platform Management
Business, Workspace and Permissions
API Documentation
History
API Category
Making API Requests
Atomic Capability APIs
Operation Optimization APIs
Document Library APIs
Q&A Database APIs
Knowledge Tag APIs
Application Management APIs
Enterprise Management APIs
Billing APIs
Release Management APIs
Dialogue Endpoint APIs
Data Statistics APIs
Data Types
Error Codes
Application API Documentation
Dialogue API Overview
Dialog API Documentation (WebSocket)
Dialog API Documentation (HTTP SSE)
Image Chat or File Chat (Real-time Document Parsing + Chat)
Offline Document Upload
Tencent Cloud Agent Development Platform Operation COS Guide
ADP Document Parsing Protocol
FAQs
Product FAQs
Technical FAQs
Related Agreements
Tencent Cloud Agent Development Platform Service Level Agreement
Tencent Cloud Agent Development Platform Service Specific Terms
Tencent Cloud Agent Development Platform Privacy Policy
Tencent Cloud Agent Development Platform Data Processing and Security Agreement
Open-Source License Statement
Lighthouse OpenClaw Connector Plugin Service Agreement
Contact Us
Glossary
ドキュメントTencent Cloud Agent Development PlatformGetting StartedCreating a “Webpage Scraping Assistant” in Single Workflow Mode

Creating a “Webpage Scraping Assistant” in Single Workflow Mode

PDF
フォーカスモード
フォントサイズ
最終更新日: 2025-09-09 17:35:18
Agent Development Platform supports building single-workflow mode applications. This mode provides an intuitive visual canvas, allowing quick addition of multiple nodes such as large models, code, and plug-ins via drag and drop to construct complex business processes and achieve stable, controllable business impact.
This document uses the creation of a "webpage scraping assistant" as an example to introduce in detail how to create a single-workflow mode application on Agent Development Platform.

Prerequisites

A registered Tencent Cloud account with authentication completed.
To register a Tencent Cloud account, please click Sign Up - Tencent Cloud.
To complete authentication, please click Sign in - Tencent Cloud.

Operation Steps

Step 1: Creating a New App

2. Create a new app
2.1 Click App Dev in the menu on the left.



2.2. Click New Application, enter application name and application description, and select Single Workflow Mode for application mode.



2.3. Click Create, and it will automatically redirect to a new workflow page.




Step 2: Creating and Setting Up a Workflow

1. Create a workflow.
Enter the workflow name and workflow description, then click Confirm to create a workflow.



You can also use AI One-click optimization to optimize the workflow description you entered, allowing the LLM to refine the content entered by user into structured content so that the model can understand when to trigger workflow better.
Note:
The One-click optimization feature will consume token resources.



2. Add a start node.
The start node is used to set the information required to launch the workflow. In this scene, users need to provide the webpage link to be parsed. The link is imported by default through the SYS.UserQuery system variable of the start node, no need to add other variables.



3. Add a webpage scraping plugin node.
The plugin node enables quick setup of listed plugins in the Agent Development Platfrom store or third-party plugins, making application development simple and enhance the power of the agents.
3.1 On the left side of the canvas, select the plugin node in information collection, search for the Fetch plugin, and add it.



3.2. Set the variable url to refer to the system variable SYS.UserQuery, and use defaults values for other input variables.



4. Add a response node.
The response node is used to reply to the dialogue. The content you set will be presented to users.
4.1. Select the response node from the basic nodes on the left side of the canvas and add it.



4.2. Create a text variable, select the data source as the output variable Output.result.content.text from the fetch1 node.



4.3. In the response content, type / to insert node variables, then select the input variable text from the current node.



5. Add an end node.
The end node is the final node of the workflow, used for returning the workflow result.
Note:
The output result of the end node is only used as an output variable. To present the output content to users, you need to add a response node.
Create an output variable called result , select Output.Answer from Reply 1 node as the reference of the output variable.



6. Debug the workflow.
6.1. Click Debug on the right side of the canvas.



6.2. Manually set the VisitorBizId value to indicate current session user integration, then click Debug.



6.3. Enter the website link and press Enter to check the output of the workflow. During debugging, you can observe the workflow execution on the canvas to verify whether the workflow is configured correctly.



7. Enable workflow.
Return to the workflow settings page and enable the created workflow.




Step 3: Setting the App

1. Set the model.
Enter the Application Settings Page, click Model Settings to select a model for understanding user input.



2. Add workflow.
Click Click to Add Workflow to add the Web Scraping Assistant workflow.
Note:
The workflow must be debugged successfully before it can be added.



Moreover, you can choose to add welcome words, voice, and avatar to enhance the experience.

Step 4: Debugging the Application

After setting up the application, you can test whether it meets the expectations in the debug area on the right.




Step 5: Releasing the Application

Once the debug result meets the expected outcome, click Release to release the application.



After the application is released, go to Release > Service Status to get the Experience Link and API informations.
To release your application on platforms such as WeChat or WeCom, go to Release - Release Publishing Channels. For more details, check Application Release Overview.

You have now created a "Webpage Scraping Assistant" using "Single Workflow Mode". For instructions on creating applications in Standard mode and Multi-Agent mode, refer to:

ヘルプとサポート

この記事はお役に立ちましたか?

フィードバック