Agent Management
Manage AI agents with multi-model collaboration, skill binding, and MCP tool integration for complex autonomous task execution.
An Agent is Cherry Studio Enterprise's advanced AI capability designed for development scenarios. Unlike assistants that focus on prompt-driven conversations and knowledge bases, agents feature multi-model collaboration (primary/planning/fast models), permission mode control, skill binding, and MCP tool integration, enabling autonomous completion of complex multi-step tasks.
With this module, admins can create and configure specialized AI agents for enterprise teams and distribute them to user groups.
Core value:
- Multi-model collaboration: Configure optimal models for different task phases — primary model for core reasoning, planning model for complex analysis, fast model for quick operations.
- Permission governance: Admins can precisely control the default and maximum permission modes to enforce enterprise security policies.
- Skill extensions: Bind pre-built skills from the resource library to give agents code analysis, file processing, and other specialized capabilities.
- MCP tool integration: Connect MCP servers to let agents call external tools and data sources.
The agent list displays all created agents, with search by name and group filtering. Each agent shows its name, group, bound model, and description.
Validation and Publishing
Run functional tests, RAG validation, and tool verification before publishing, then authorize distribution to user groups.
Create and Configure Agents
Create agents through 7 configuration tabs covering quick setup, model selection, permission control, skill binding, and MCP tool integration.