Copilot Agents - Declarative vs Custom
Graham Hosking
Snr. Technical Specialist @ Microsoft | Copilot & AI Solutions | Professional Services Industry
When does a Declarative Agent become a Custom Engine Agent?
A question I've also been wondering and trying to get to the bottom of is as I'm building complex agents for some customers, I've thought to myself "Is this now a custom engine agent?"...
A Declarative Agent will become a Custom Engine Agent when:
In summary, your agent remains declarative as long as it primarily uses Microsoft's default models, orchestration, and low-code tools. However, once you incorporate substantial customisation at the foundational, orchestration, or logic level, it transitions into a Custom Engine Agent.
Comparison Matrix (in picture form, because IN articles can't do tables)...
Extending Custom Engine Agents
To Extend the capabilities you can use other methods in order to us OpenAI assistants API and custom OpenAPIs for example:
AI Agent Build an AI agent chatbot for Teams that can make decisions and perform actions based on LLM reasoning. AI Agent template has options to start from scratch or use OpenAI Assistants API.
Chat with your data Expand AI bot's knowledge with your content to get more accurate answers to your questions. Chat with your data supports Azure AI Search, Microsoft Graph Search with Microsoft 365 and SharePoint content, and custom OpenAPI and data ingestion options.
Other Methods - Connect Data to Azure OpenAI for generative answers
You can also connect a Copilot Studio agent to your data by selecting Deploy to and then A new Microsoft Copilot Studio bot. The agent is created and automatically connects to your Azure resource in your default environment.