Skip to main content
This is preview documentation. The official release is not yet available.
DESIGN → Agents manages identity, behavior and execution. Reuse an Agent in Chat, Issues, Teams and Endpoints. Saving its definition does not start work.

​
Choose an integration path

  • Managed Agent: create a cloud Agent in the console, then configure its model and resources.
  • AgentScope framework: register your own deployed application as an External Agent; task execution requires the corresponding adapter support.
  • Hosted: connect Runtime Host and select an installed, authenticated Coding Agent provider.
The three submenus cover creation, registration and verification. Detailed parameters, resource configuration and execution principles are collected in Reference. After connecting an Agent, submit work through an Endpoint or create an Issue in the console.

​
Interface tour

Agent catalog with Managed, Hosted and External filters

Current console UI with fixed demonstration data.

Use the filters to distinguish runtime types and the cards to identify each Agent’s purpose. Open a card to configure it, or choose New agent to add one. The example shows all three types; verify runtime availability for each Agent before using it.

​
Choose an execution type

The work entry points are shared, but models, tools, recovery and configuration projection differ. Catalog visibility does not guarantee every conversation or dispatch capability. See the dedicated Managed, Hosted and External guides.

​
Create an Agent

  1. Start creation and enter its name, purpose and Instructions.
  2. Link a Workspace when it should reuse operating guidance, skills, tools or subagent definitions.
  3. Explicitly choose Runtime under Execution. An online Hosted provider may be selected initially; choose AgentScope Managed for a Managed Agent.
  4. Leave Model empty for the runtime default, or enter a model identifier supported by that provider.
  5. For Managed Agents, choose an Environment in Advanced settings. The automatic local default requires administrator permission for Local execution.
  6. Select Create & open agent, then review the resulting configuration.
Agent key is a stable identity within the scope. The display name communicates purpose to colleagues.

​
Configure the detail pages

​
Add capabilities incrementally

Verify conversation first, then file access, an external tool and a specialist skill. Store credentials in Vault and shared knowledge in Memory. Configuration does not automatically install executable tools or grant external permissions. A shared Workspace can affect multiple Agents. Check its consumers before editing and verify with a new Chat or Issue. Do not assume that running work switches definitions immediately. Next: Teams · Workspaces.