CLI

The AgentPool CLI provides a comprehensive set of commands to manage and interact with AI agents. It's designed around the concept of an "active agent file" - a YAML configuration that defines your agents and their settings. This avoids the need to pass the config file path each time you want to run a command.

Active Agent File¶

The CLI maintains an "active agent file" setting which determines which agents are available for commands like run, task, or watch. You can:

• Add agent files with agentpool add <name> <path>
• Set the active file with agentpool set <name>
• List agents from the active config with agentpool list

Most commands will use the currently active agent file by default, but can be overridden with the --config option.

Available Commands¶

Configuration¶

Agent Management¶

Execution¶

Server Commands¶

History Management¶

Quick Start¶

Option 1: Automatic Config Discovery¶

AgentPool automatically discovers config files. Just create agentpool.yml in your project:

# Create a starter config
agentpool config init

# Edit the config, then run
agentpool run myagent "Hello"

Option 2: Named Configurations¶

For managing multiple config files:

1. Add and activate an agent configuration:

agentpool add myconfig agents.yml
agentpool set myconfig

1. List available agents:

agentpool list

1. Run a prompt with an agent:

agentpool run analyzer "Analyze this text"

Command Examples¶

Running Agents¶

# Run a single agent with a prompt
agentpool run myagent "Analyze this"

# Run a team
agentpool run myteam "Process this"

# Show detailed output with costs
agentpool run myagent "Hello" --detail full --costs

Executing Tasks¶

# Execute a defined task
agentpool task docs write_api_docs

# Execute with additional prompt
agentpool task docs write_api_docs --prompt "Include code examples"

Server Commands¶

# ACP server for IDE integration (Zed, Toad)
agentpool serve-acp config.yml

# OpenCode server for OpenCode TUI/Desktop
agentpool serve-opencode config.yml --port 4096

# MCP server (stdio transport)
agentpool serve-mcp config.yml

# MCP server with SSE transport
agentpool serve-mcp config.yml --transport sse --port 3001

# AG-UI server
agentpool serve-agui config.yml --port 8002

# OpenAI-compatible API server
agentpool serve-api config.yml --port 8000

History Commands¶

# Show last 5 conversations
agentpool history show -n 5

# Show conversations from last 24 hours
agentpool history show --period 24h

# Show stats grouped by model
agentpool history stats --period 1w --group-by model

# Clear history for specific agent
agentpool history reset --agent myagent

Global Options¶

Configuration Files¶

Agent configurations are YAML files that define:

• Available agents and their capabilities
• System prompts and knowledge sources
• Tool configurations
• Response types
• And more

Example:

agents:
  analyzer:
    display_name: "Text Analyzer"
    model: openai:gpt-4o
    description: "Analyzes text and provides structured output"
    tools:
      - type: file_access

See the Configuration Guide for detailed information about agent configuration.