Navigation: Documentation Home → Reference → CLI Quick Reference
Quick reference for the most commonly used AgenticGoKit CLI commands.
# Research assistant with web search
agentcli create research-bot --template research-assistant
# RAG knowledge base
agentcli create knowledge-base --template rag-system
# Data processing pipeline
agentcli create data-flow --template data-pipeline
# Chat system with memory
agentcli create chat-bot --template chat-system
# Basic multi-agent system
agentcli create my-project --template basic# Custom RAG system
agentcli create my-kb --memory pgvector --embedding openai --rag 1500
# Production MCP setup
agentcli create my-bot --mcp standard --visualize
# Interactive setup
agentcli create --interactive# Use template but customize
agentcli create my-research --template research-assistant --agents 5 --mcp advanced# List all templates
agentcli template list
# Create custom template
agentcli template create my-template
# Validate template
agentcli template validate my-template.yaml
# Show search paths
agentcli template paths| Flag | Description | Examples |
|---|---|---|
--template, -t |
Project template | basic, research-assistant, rag-system |
--memory |
Memory provider | memory, pgvector, weaviate |
--embedding |
Embedding provider | openai, ollama:nomic-embed-text |
--enable-mcp |
Enable MCP minimal | (boolean flag) |
--mcp |
MCP integration | minimal, standard, advanced |
--rag |
RAG chunk size | default, 1000, 2000 |
--orchestration |
Agent coordination | sequential, collaborative, loop |
--agents, -a |
Number of agents | 2, 3, 5 |
--provider, -p |
LLM provider | openai, azure, ollama |
--visualize |
Generate diagrams | (boolean flag) |
--interactive, -i |
Interactive mode | (boolean flag) |
| Template | Description | Use Case |
|---|---|---|
basic |
2 agents, sequential | Learning, simple projects |
research-assistant |
3 agents, web search | Research, analysis |
rag-system |
3 agents, vector DB | Knowledge bases, Q&A |
data-pipeline |
4 agents, sequential | ETL, data processing |
chat-system |
2 agents, memory | Chatbots, conversations |
# Version info
agentcli version
# List sessions
agentcli list
# View traces
agentcli trace session-id
# Memory debug
agentcli memory --stats
# MCP management
agentcli mcp servers
# Cache management
agentcli cache stats
# Shell completion
agentcli completion bash > /etc/bash_completion.d/agentcli
agentcli completion zsh > "${fpath[1]}/_agentcli"
agentcli completion powershell > agentcli.ps1# Quick development project
agentcli create dev-project --template basic --provider openai# Production-ready RAG system
agentcli create prod-kb --template rag-system --mcp advanced --visualize# Research assistant with custom agents
agentcli create research-team --template research-assistant --agents 4# ETL pipeline with visualization
agentcli create etl-system --template data-pipeline --visualize# General help
agentcli --help
# Command-specific help
agentcli create --help
agentcli template --help
# Template details
agentcli create help-templates
agentcli template listEnable intelligent tab completion for faster CLI usage:
# Load for current session
source <(agentcli completion bash)
# Install permanently (Linux)
agentcli completion bash > /etc/bash_completion.d/agentcli
# Install permanently (macOS with Homebrew)
agentcli completion bash > $(brew --prefix)/etc/bash_completion.d/agentcli# Enable completion support
echo "autoload -U compinit; compinit" >> ~/.zshrc
# Install completion
agentcli completion zsh > "${fpath[1]}/_agentcli"# Load for current session
agentcli completion powershell | Out-String | Invoke-Expression
# Install permanently
agentcli completion powershell > agentcli.ps1
# Add to your PowerShell profile# Load for current session
agentcli completion fish | source
# Install permanently
agentcli completion fish > ~/.config/fish/completions/agentcli.fishFor complete documentation, see the Full CLI Reference.