Polyagent FAQs

Question 1

Does Polyagent require extensive configuration?

Accepted Answer

No, Polyagent is designed for zero-configuration integration. Once the MCP server is set up in your client (Codex or Gemini CLI), it automatically discovers and uses your existing `.claude/agents/` definitions.

Question 2

How does Polyagent improve performance and context optimization?

Accepted Answer

Polyagent spawns agents in isolated CLI sessions. This approach optimizes context by only passing essential information, not intermediary tokens. It also improves performance and prompt adherence by allowing each agent to operate with its own context and rules, adhering to the single responsibility principle.

Question 3

What are the core requirements to use Polyagent?

Accepted Answer

To use Polyagent, you need Python 3.10+, at least one of the `codex`, `claude`, or `gemini` CLI tools installed, and an existing `.claude/agents/` directory containing your Claude Code agent definitions.

Question 4

What is Polyagent and what problem does it solve?

Accepted Answer

Polyagent is an MCP server that seamlessly integrates Claude Code agents into AI coding assistants like Codex and Gemini CLI. It solves the problem of reusing existing `.claude/agents/` definitions across different AI coding environments without any additional porting or configuration.

Question 5

Which AI coding assistants and clients does Polyagent support?

Accepted Answer

Polyagent supports Codex, Gemini CLI, and any other MCP-compatible client. It's designed to be a drop-in solution for these environments to leverage existing Claude Code agent definitions.

Polyagent

Polyagent

主要功能

使用案例

主要功能

使用案例