Skip to main content

Basic Agent Creation

Basic Agent Creation​

  1. On the AI Studio page, click "Create" at the top right corner to create a basic agent

  1. Creation steps

    • Enter the agent name, select the agent avatar, select the model group, select the agent category, add an agent description:

      ① Agent Name: Enter the name of the agent as its identifier.
      ② Agent Avatar: Select the default avatar for the agent; uploading avatars is currently not supported.
      ③ Model Group: Configure an appropriate model group for the agent.
      ④ Agent Category: Select the group(s) where the new agent belongs; multiple selections are allowed.
      ⑤ Agent Description: Enter a brief description explaining the agent's functions and purposes.

    • Click "Create". After the agent is created, it will enter the basic orchestration agent configuration page. Configure and publish it before putting it into use.

  1. Agent Configuration

There are two ways to enter agent configuration:

  • Directly enter the agent configuration page after creating the agent;

  • Hover the mouse over the agent card to see the "✏️" icon, click it to enter the configuration page.

    ① Prompt: Enter the agent prompt. Intelligent generation of prompts from existing prompts is also supported. The prompt is limited to 2000 characters.
    ② Opening Statement: Enter the agent's opening statement. Intelligent generation based on the prompt or existing opening statement is also supported. The opening statement is limited to 2000 characters.
    ③ Model Group: Click "+" to add a model group; multiple selectable models are supported.

Note: Model groups need to be added by administrators in System Management first. Multiple different models are added to the same model group, which is then configured to the agent.

Adding Model Groups

  • Path: Management → Model Management → Model Group → New Model Group (only administrators can add models)

  • Adding Steps:

    • Click "New Model Group"
    • Complete the following configurations:
      • Enter the model group name
      • Select models to add to the model group; multiple selections allowed
      • Choose whether to enable adaptive model deployment
      • Choose whether to enable deep thinking model
    • Click "Save"

  • Adaptive Model Deployment: Automatically adjusts computing resources based on traffic to ensure stable and smooth service;

  • Deep Thinking Model: Intelligently calls more powerful AI for complex problems, significantly improving answer quality.

④ Skills: Click "+" to add one or more skills, or add recommended skills

There are 5 default skills: Google Search, Tencent Search, Text-to-Image, News Query Tool, Webpage Reader.

  • Google Search: Obtains real-time and accurate web information through the Google search engine, supporting global webpage content retrieval.
  • Tencent Search: Based on Tencent search technology, provides search services optimized for the Chinese internet environment, especially for Chinese content retrieval.
  • News Query Tool: A dedicated tool for searching and obtaining various news information.
  • Webpage Reader: Extracts webpage text, data, and other content, parsing webpage information.
  • Text-to-Image: Automatically generates corresponding image content based on text descriptions, transforming textual creativity into visual presentation.

Note: Additional skills can be added by administrators through operations and configurations.

⑤ MCP Service

  • MCP service manages the connection permissions between AI agents and external tools or data sources within the system

    • Capability Extension: Enables AI agents with practical functions such as search, calculation, and visualization.
    • Rich Ecosystem: Continuously integrates various tool services to meet diversified needs.
    • Standardized Access: Integrates internal system resources through personal MCP.

⑥ Conversation Experience:

  • Conversation Settings: You can enable settings such as "User Question Suggestions, Question Guidance, Chat History, Conversation Feedback, Keyword Review"
    • User Question Suggestions: After the agent answers, it provides some question suggestions to the user based on the previous context.
    • Question Guidance: During user-agent conversations, related question guidance is provided, using model capabilities to infer possible user questions and complete user queries.
    • Chat History: Whether to retain the agent's chat history. If turned off, the agent's chat history cannot be retrieved.
    • Conversation Feedback: Users can like or dislike the agent's answers to optimize responses.
    • Enable Keyword Review: At least one of input content review or output content review must be enabled. Once enabled, sensitive word detection will be performed on prompts or AI feedback results. Sensitive words can be maintained in advance.

⑦ Knowledge Base:

  • Knowledge Base: Click "+" to add a knowledge base
    • Allow File Upload:
      • If enabled, content from knowledge bases cannot be added as knowledge sources
      • If disabled, knowledge bases from personal or enterprise spaces can be selectively added as knowledge sources
  • Knowledge Base Configuration: You can modify detailed settings such as "Retrieval Strategy, Private Domain Q&A, Retrieval Method"

    1. Retrieval Strategy: Hybrid Search, Embedding Search, Text Search

      • Hybrid Retrieval: Combines vector retrieval and full-text retrieval results, returning reranked results
      • Embedding Retrieval: Finds fragments based on similarity, with some cross-language generalization ability
      • Text Retrieval: Finds fragments based on keywords, suitable for retrieval containing specific keywords or noun fragments

    2. Maximum Recall Count: Range 1–10; setting too high or too low is not recommended; suggested value is 3–5

    3. Metadata Filtering: None, Filter, Weight

    4. Force Private Domain File Q&A: When enabled, network search and other skills will not be used; the agent's answers only target knowledge base content

    5. Document Matching Similarity: Range 0–1; the higher the similarity, the more similar the recalled document content is; suggested value is about 0.8 (i.e., 80%)

    6. QnA Matching Similarity: Range 0–1; similar to document content similarity matching; suggested value is about 0.9 (i.e., 90%)

    7. Show References: When enabled, the agent will list referenced documents in answers to improve answer credibility

💡 Tip: Neither maximum recall count, document matching similarity, nor QnA matching similarity is better the higher or lower. It is recommended to set according to actual needs. If no special requirements, keep the default values.

⑧ Data Sources

  • Data Sources: Click "+" to add data sources as the agent's Q&A data source

  • Enable Template: Whether to enable preset mapping templates between natural language and SQL.

    • When a user inputs a natural language question (e.g., "What was the sales last month?"), the system will first try to match a preset template.
    • If a matching template is found (e.g., a general question like "Query sales for a certain period"), the existing SQL structure in the template will be used as a reference, combined with specific fields/table names to generate the final SQL statement.

  • Enable Parameterized Template: When enabled, parameterized queries are enabled on the template basis, enhancing query flexibility and security.

  • Question Rewriting: When enabled, the user's input question will be automatically optimized to ensure accurate data queries

    • Original user question: Check the sales (incomplete information)
    • After rewriting: Query the total sales of all products in July 2024 (added time and scope)

  • Step-by-Step Thinking: When this feature is enabled, before generating the final query result, the system will output detailed thinking steps explaining how it analyzes the question and constructs the SQL query.

    • Step 1: Identify keywords "July 2024" and "sales"
    • Step 2: Determine data table Orders, fields order_date and sales_amount
    • Step 3: Construct date range condition from 2024-07-01 to 2024-07-31
    • Step 4: Generate SQL