AgenticGoKit

AgenticGoKit

Production-ready Go framework for building intelligent multi-agent AI systems

AgenticGoKit enables developers to build sophisticated agent workflows with dynamic tool integration, multi-provider LLM support, and enterprise-grade orchestration patterns. Designed for Go developers who need the performance and reliability of compiled binaries with the flexibility of modern AI agent systems.

⚠️ Alpha Release: AgenticGoKit is currently in alpha development. APIs may change, and some features are still being developed. Suitable for experimentation and early adoption, but not recommended for production use yet.

5-Minute Demo

Create a collaborative multi-agent system with one command:

# Install the CLI
go install github.com/kunalkushwaha/agenticgokit/cmd/agentcli@latest

# Create a multi-agent research team
agentcli create research-team --template research-assistant --visualize

cd research-team

# Set your API key
export AZURE_OPENAI_API_KEY=your-key-here
export AZURE_OPENAI_ENDPOINT=https://your-resource.openai.azure.com/
export AZURE_OPENAI_DEPLOYMENT=your-deployment-name

# Run the collaborative system
go run . -m "Research the latest developments in AI agent frameworks"

What you get:

• Complete Go project with main.go, agentflow.toml, and go.mod
• Three specialized agents working in parallel
• Automatic result synthesis and error handling
• Mermaid workflow diagrams generated
• Production-ready project structure

That's it! Multi-agent collaboration with one CLI command.

Installation

One-Line Installation (Recommended)

Windows (PowerShell)

iwr -useb https://raw.githubusercontent.com/kunalkushwaha/agenticgokit/master/install.ps1 | iex

Linux/macOS (Bash)

curl -fsSL https://raw.githubusercontent.com/kunalkushwaha/agenticgokit/master/install.sh | bash

Alternative Methods

Go Install

go install github.com/kunalkushwaha/agenticgokit/cmd/agentcli@latest

Specific Version

# Linux/macOS
curl -fsSL https://raw.githubusercontent.com/kunalkushwaha/agenticgokit/master/install.sh | bash -s -- --version v0.3.0

# Windows
iwr -useb 'https://raw.githubusercontent.com/kunalkushwaha/agenticgokit/master/install.ps1' | iex -Version v0.3.0

Manual Download

1. Go to Releases
2. Download binary for your platform
3. Add to PATH

Build from Source

git clone https://github.com/kunalkushwaha/agenticgokit.git
cd agenticgokit
make build

Post-Installation

# Verify installation
agentcli version

# Enable shell completion (optional)
# Bash: source <(agentcli completion bash)
# Zsh: agentcli completion zsh > "${fpath[1]}/_agentcli"
# PowerShell: agentcli completion powershell | Out-String | Invoke-Expression

# Create your first project
agentcli create my-project --template basic

Complete Installation Guide - Detailed instructions, troubleshooting, and advanced options

Why AgenticGoKit?

Framework Comparison: See our detailed comparison with LangChain, AutoGen, CrewAI, and Semantic Kernel.

What You Can Build

agentcli create research-team \
  --template research-assistant \
  --visualize

agentcli create data-pipeline \
  --template data-pipeline \
  --visualize

agentcli create chat-system \
  --template chat-system \
  --visualize

agentcli create knowledge-base \
  --template rag-system \
  --visualize

Quick Start Paths

go get github.com/kunalkushwaha/agenticgokit

git clone https://github.com/kunalkushwaha/agenticgokit
cd examples/04-rag-knowledge-base
docker-compose up -d
go run main.go

Core Concepts

Multi-Agent Orchestration

// Collaborative agents (parallel execution)
agents := map[string]core.AgentHandler{
    "researcher": NewResearchAgent(),
    "analyzer":   NewAnalysisAgent(),
    "validator":  NewValidationAgent(),
}

runner := core.CreateCollaborativeRunner(agents, 30*time.Second)
result, err := runner.ProcessEvent(ctx, event)

Configuration-Based Setup

# agentflow.toml
[orchestration]
mode = "collaborative"
timeout_seconds = 30

[agent_memory]
provider = "pgvector"
enable_rag = true
chunk_size = 1000

[mcp]
enabled = true

Memory & RAG Integration

// Configure persistent memory with vector search
memory, err := core.NewMemory(core.AgentMemoryConfig{
    Provider: "pgvector",
    EnableRAG: true,
    EnableKnowledgeBase: true,
    ChunkSize: 1000,
})

Tool Integration (MCP)

// MCP tools are automatically discovered and integrated
// Agents can use web search, file operations, and custom tools
agent := agents.NewToolEnabledAgent("assistant", llmProvider, toolManager)

Current Features

• Multi-Agent Orchestration: Collaborative, sequential, loop, and mixed patterns
• Memory & RAG: PostgreSQL pgvector, Weaviate, and in-memory providers
• Tool Integration: MCP protocol support for dynamic tool discovery
• Configuration Management: TOML-based configuration with environment overrides
• Workflow Visualization: Automatic Mermaid diagram generation
• CLI Scaffolding: Generate complete projects with one command
• Production Patterns: Error handling, retry logic, and monitoring hooks

Documentation

Complete Documentation

Environment Setup

LLM Provider Configuration

# For Azure OpenAI (recommended)
export AZURE_OPENAI_API_KEY=your-key-here
export AZURE_OPENAI_ENDPOINT=https://your-resource.openai.azure.com/
export AZURE_OPENAI_DEPLOYMENT=your-deployment-name

# For OpenAI
export OPENAI_API_KEY=your-key-here

# For Ollama (local)
export OLLAMA_HOST=http://localhost:11434

Plugin imports (LLM, logging, orchestrator)

AgenticGoKit uses public plugin packages that self-register via blank imports. Ensure your main package (or your CLI/plugins bundle) includes the imports for the features you use:

import (
  _ "github.com/kunalkushwaha/agenticgokit/plugins/llm/ollama"
  _ "github.com/kunalkushwaha/agenticgokit/plugins/logging/zerolog"
  _ "github.com/kunalkushwaha/agenticgokit/plugins/orchestrator/default"
  _ "github.com/kunalkushwaha/agenticgokit/plugins/runner/default"
)

Troubleshooting: If you see errors like "provider not registered" or "orchestrator factory not registered", add the corresponding blank import above. The scaffolded projects and agentcli already include a plugins bundle by default.

Community & Support

License

Apache 2.0 - see LICENSE for details.

AgenticGoKit: Where Go performance meets AI agent intelligence.