Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
In the agent details and instructions, define the agent and its behavior. This definition shapes the agent's tone, scope, and response style.
Agents powered by the GitHub Copilot harness use instructions as the primary mechanism for controlling agent behavior. The enhanced orchestration runtime interprets these instructions when responding to users. Instructions tell the agent what to do, how to respond, and what to avoid.
Configure the following elements from the Build tab:
- Agent name: An editable name field where you set the agent's display name.
- Agent icon: An icon for your agent. You can replace the default icon and its color with something more representative.
- Instructions: Natural language guidance that shapes the agent's behavior, tone, and scope.
Configure your agent's details and instructions
Open your agent in Copilot Studio.
Select the Build tab.
To update the agent name, select the name field and enter a new name.
To replace or customize the agent icon, select the existing icon. In the Change agent icon dialog, select a new icon from the catalog or upload the icon you want. The icon must be in PNG format, and its size must be 100 KB or less. Select Colors to choose a different color or add an accent color. Select Apply to save your changes.
In the Instructions section, select the space under Instructions and begin typing. Enter natural language instructions that guide the agent's behavior. Be specific about the agent's purpose, tone, and boundaries. Select Save.
Write effective instructions
Instructions are the most important tool for shaping agent behavior. Well-written instructions help the agent stay on topic, respond in the right tone, and handle edge cases appropriately.
Consider including:
- The agent's primary role and purpose.
- The tone and style the agent should use (formal, friendly, concise).
- Subjects or tasks the agent should decline or redirect.
- How the agent should handle unclear or ambiguous user input.
- Any escalation or handoff triggers.
Tip
Start with a clear, simple description of the agent's purpose and expand from there. You can update instructions iteratively as you test the agent's behavior.
For general guidance on writing agent instructions, see Write agent instructions. (Some guidance in that article is specific to the standard harness's topic-based model.)