Release Notes
Product Introduction
Purchase Guide
Getting Started
Buddy
- Buddy - Data Analytics Scenarios
- Data Engineering Agent
- Buddy Overview
- Data Governance Agent
- Buddy Data Security
- Overview of Buddy Agent Basic Capabilities
Platform Management
- Workspace Management
- Compute Resource Management
- Data Source Management Overview
Data Integration
- Supported Data Ingestion Solutions
- Supported Data Sources and Read Capabilities
- Data Source Management
- List of Data Sources
- Offline Data Synchronization
- Real-Time Data Synchronization
- File Upload
Data Engineering
- Studio
- Workflow
- CI/CD
Data Science
- Data Science Overview
- Model Experiments
- Feature Management
- Model Management
- Model Service
Data Analysis
- Data Exploration Overview
Data Governance
- Data Catalog Overview
- Data Discovery Overview
- Data Security Overview
- Data Quality Overview
- Semantic Model Overview
- Governance Tag Overview
Data Applications
- Agent Development
- MCP
- Model List
API Documentation
- History
- Introduction
- API Category
- Making API Requests
- Console APIs
- Workspace APIs
- Computational Resource APIs
- Development Studio APIs
- Workflow APIs
- Catalog APIs
- Data Types
- Error Codes
FAQs
Agreements
Contact Us
Glossary

Agent Overview

Download

Focus Mode

Font Size

Last updated: 2026-10-09 16:16:53

AI-Translated

This document describes the product positioning, types, and typical usage flows of Agents in DataBuddy, helping you build an overall understanding of the Agent module.
Product Positioning
The Agent module is an agent platform provided by the data application layer of DataBuddy, serving two types of users:
Developers: Flexibly customize complex enterprise-level Agents using high-code development templates (OpenAI/LangChain/LangGraph) and integrate them with DataBuddy data and AI capabilities.
Business personnel/Analysts: With capabilities such as intelligent tuning, business personnel can also improve Agent response quality through feedback.
The platform provides Agents with full-lifecycle capabilities from creation, development, debugging, deployment, and running to evaluation, optimization, and Ops, and invokes data and AI resources within DataBuddy in a secure and governed manner.
Two Working Modes of Agent
The industry typically classifies Agents into two types, both of which are supported by DataBuddy:
Mode
Description
Scenario
Workflow mode
Predefined workflows built into the Agent deliver predictable and stable execution results through process orchestration.
Clearly defined tasks and high requirements for result consistency (such as customer service Q&A with fixed processes)
Autonomous decision mode
Dynamically determine which tools to call and which paths to take based on large model inference.
Requires flexibility and handles open-ended tasks (such as weekly report insights and complex analysis assistants)
In real-world projects, the two modes are often used in combination. It is recommended to start simple and then scale up: first solve problems with simple API + Prompt, and adopt a full Agent only when decision-making is required.
Supported Deployment Methods for Agent Platform
The platform homepage (Agent list) provides the following creation entries:
Type
Description
Phase
High-code development Agent - OpenAI framework
Code template based on the OpenAI Agents framework
Launched
High-code development Agent - LangGraph framework
Code template based on the LangChain / LangGraph framework
Launched
End-to-End Process of Agent
1. Create an Agent (select a creation method/template).
        │
        ▼
2. Configure tools and knowledge bases (MCP).
        │
        ▼
3. Develop in Studio.
        │
        ▼
4. Debug: Preview and test in ChatUI, and view Traces.
        │
        ▼
5. Deploy: Deploy from Studio to the runtime environment.
        │
        ▼
6. Run: Invoke through ChatUI / API / by other Agents.
        │
        ▼
7. Evaluation and optimization: Evaluator scoring, human feedback, and intelligent tuning (ALHF).
My Agent List
After you go to the Agent module, the list area displays the agents for which the current user has usage or management permissions (across workspaces under the same root account). The list fields are as follows:
Field
Description
Agent name
Filled in at creation
Agent type
Custom code Agent
Agent description
Description information filled in at creation
Owner
Agent owner
Invocation Endpoint
Endpoint name through which the Agent can be called (for an App-type Agent, it is the App name)
Last updated
Last updated time, sorted from newest to oldest
Supporting Infrastructure
The Agent module depends on the following infrastructure, which is provided by default in the platform:
Data Center Operating System (DCOS)
Function
MCP service list
Standard protocol layer for Agent tool invocation. For details, see MCP Overview.
Model list
The platform hosts base models. Agents can call them directly by model ID without configuring an API Key.
MLflow experiment
Carries the Agent's traces, evaluators, evaluation runs, and versions, serving as the vehicle for evaluation and optimization.
Studio
Development and debugging environment for Agent code. For details, see Studio Overview.
Catalog
Provide assets such as Catalog functions, Volumes, and vector indexes that can be called by Agent.
Studio AI Assistance
Provide AI-assisted capabilities such as code generation and tool discovery in Studio to improve Agent development efficiency.
References
Agent Overview
Creating and Managing Agents
MCP Overview
Model List Overview

Help and Support

Was this page helpful?

You can also Contact sales or Submit a Ticket for help.

Help us improve! Rate your documentation experience in 5 mins.

Feedback

tencent cloud

Tencent Cloud DataBuddy

Agent Overview

Product Positioning

Two Working Modes of Agent

Supported Deployment Methods for Agent Platform

End-to-End Process of Agent

My Agent List

Supporting Infrastructure

References

Help and Support

Mode	Description	Scenario
Workflow mode	Predefined workflows built into the Agent deliver predictable and stable execution results through process orchestration.	Clearly defined tasks and high requirements for result consistency (such as customer service Q&A with fixed processes)
Autonomous decision mode	Dynamically determine which tools to call and which paths to take based on large model inference.	Requires flexibility and handles open-ended tasks (such as weekly report insights and complex analysis assistants)

Type	Description	Phase
High-code development Agent - OpenAI framework	Code template based on the OpenAI Agents framework	Launched
High-code development Agent - LangGraph framework	Code template based on the LangChain / LangGraph framework	Launched

Field	Description
Agent name	Filled in at creation
Agent type	Custom code Agent
Agent description	Description information filled in at creation
Owner	Agent owner
Invocation Endpoint	Endpoint name through which the Agent can be called (for an App-type Agent, it is the App name)
Last updated	Last updated time, sorted from newest to oldest

Data Center Operating System (DCOS)	Function
MCP service list	Standard protocol layer for Agent tool invocation. For details, see MCP Overview.
Model list	The platform hosts base models. Agents can call them directly by model ID without configuring an API Key.
MLflow experiment	Carries the Agent's traces, evaluators, evaluation runs, and versions, serving as the vehicle for evaluation and optimization.
Studio	Development and debugging environment for Agent code. For details, see Studio Overview.
Catalog	Provide assets such as Catalog functions, Volumes, and vector indexes that can be called by Agent.
Studio AI Assistance	Provide AI-assisted capabilities such as code generation and tool discovery in Studio to improve Agent development efficiency.