[FEATURE]: Wildcard supports for agent configuration

[huybui38]

Feature hasn't been suggested before.

• I have verified this feature I'm about to request hasn't been suggested before.

Describe the enhancement you want to request

Summary

I’d like to have wildcard support in agent configuration so I can target groups of subagents by naming pattern instead of configuring each one individually.

Problem

In my setup, I have multiple subagents organized by category, for example:

researcher/agent-a
researcher/agent-b
reviewer/agent-c
reviewer/agent-d

Right now, if I want to assign the same model or behavior to all researcher/* agents, I need to configure each one explicitly. This becomes repetitive and hard to maintain as the number of subagents grows.

Proposed Solution

I propose supporting wildcard-based matching in agent configuration, so I can apply settings to a category of agents using a pattern.

Example:

{
  "agent": {
    "researcher/*": {
      "model": "anthropic/claude-opus-4.6"
    },
   "reviewer/*": {
      "model": "opencode-go/glm-5"
    }
  }
}

With this, all agents whose names match researcher/* would inherit the configured model.