Powerplatform MCP

🚀 PowerPlatform MCP Prompts

PowerPlatform is a powerful low-code development platform for building business applications. To assist developers in using the platform more efficiently, MCP (Microsoft Cloud Platform) offers various prompt functions.

✨ Features

• Entity Overview: Provides detailed information about specific entities.
• Query Templates: Helps generate and optimize data query statements.
• Relationship Maps: Displays associations between different entities.
• Field Details: Explains the data types and usage scenarios of specific fields.

💻 Usage Examples

Basic Usage

1. Entity Understanding

Users can inquire about information regarding a specific entity through natural language:

User: Please tell me about the Account entity in our PowerPlatform environment.
Response: [Internally uses the ENTITY_OVERVIEW prompt]
Here is an overview of the Account entity:
- It is a core business entity representing customers and potential customers.
- The primary identifier is 'accountid', and the primary display field is 'name'.
- Key fields include:
  * name (Account Name)
  * emailaddress1 (Email)
  * telephone1 (Primary Phone Number)
  * revenue (Annual Revenue)
- It has 42 one-to-many relationships and 8 many-to-many relationships.

2. Query Building

When assistance is needed in writing queries, users can seek support from the AI:

User: Please help me write a query to find active accounts in California with an annual revenue of over $1 million.
Response: [Internally uses the QUERY_TEMPLATE prompt]
Here is the required OData query:
accounts?$select=name,revenue,statecode,address1_stateorprovince
&$filter=statecode eq 0 
    and revenue gt 1000000 
    and address1_stateorprovince eq 'California'
&$orderby=revenue desc

This query functions as follows:
1. Select relevant fields: name, revenue, statecode, address1_stateorprovince
2. Filter conditions:
   - The status code is 0 (indicating active)
   - The annual revenue exceeds $1 million
   - The state is California
3. Sort in descending order by annual revenue

3. Relationship Understanding

When exploring relationships between entities, users can ask questions like the following:

User: How are the Account and Contact entities associated in the system?
Response: [Internally uses the RELATIONSHIP_MAP prompt to analyze the two entities separately]
Here is an overview of the key relationships:
1. Primary Contact (One-to-Many):
   - account_primary_contact: Account (1) → Contact (N)
     Each account can have multiple contacts, but one is designated as PRIMARY.
2. Customer Relationship:
   - contact_customer_accounts: Contact (N) ← Account (1)
     Contacts can be associated with customer accounts.
3. Other Associations:
   - accountleads_association: Used to track potential customers
   - account_connections: Used to manage contacts

4. Field/Attribute Investigation

When information about a specific field is needed, users can make inquiries like this:

User: What type of field is the'revenue' field on the Account entity?
Response: [Internally uses the ATTRIBUTE_DETAILS prompt]
Here is the information about this field:
- Data Type: Money
- Format: Displayed in currency format
- Description: Represents the annual revenue of the account
- Required: Yes
- Default Value: null

5. Debugging Assistance

When encountering problems, users can seek debugging support from the AI:

User: I'm having an issue with the following query. Please help me analyze it:
accounts?$select=name,revenue&$filter=year(PropertyName=’revenue’) gt 2020

Response:

The current query has the following issues:
1. Syntax error in PropertyName: The correct field reference should be `(PropertyName='revenue')`, but the Property Name needs to be placed inside single quotes.
2. Incorrect function call: The way the `year(PropertyName)` function is called is incorrect. The correct format should be `year(PropertyName)`.
3. Field selection issue: You may need to add more relevant fields to meet the filtering criteria.

The recommended corrected query is as follows:
accounts?$select=name,revenue&$filter=year(PropertyName='revenue') gt 2020

📚 Documentation

By using the MCP prompt functions of PowerPlatform, developers can conduct entity analysis, query optimization, and problem diagnosis more efficiently. These prompts not only save development time but also improve code quality and application performance.