agent-framework-go

module
v0.1.0 Latest Latest
Warning

This package is not in the latest version of its module.

Go to latest
Published: Sep 1, 2026 License: MIT

README

Microsoft Agent Framework

Welcome to Microsoft Agent Framework for Go!

Microsoft Foundry Discord MS Learn Documentation Go Reference GitHub stars

Microsoft Agent Framework (MAF) is an open, multi-language framework for building production-grade AI agents and multi-agent workflows. This repository contains the Go implementation of Microsoft Agent Framework, along with Go samples and documentation to help you get started.

Microsoft Agent Framework is built for teams taking agents from prototype to production. It provides a consistent foundation for building, orchestrating, and operating agent systems while keeping architecture choices open as requirements evolve, and supports a broad ecosystem including Microsoft Foundry, Azure OpenAI, OpenAI, Model Context Protocol (MCP), Agent2Agent (A2A), AG-UI, and the GitHub Copilot SDK.

The .NET and Python implementations are available in the upstream Microsoft Agent Framework repository.

Note: Microsoft Agent Framework for Go is in public preview and is currently evolving outside the core upstream codebase. As adoption and feedback grow, we expect closer alignment with the broader MAF ecosystem.

Watch the full Agent Framework introduction (30 min)

Watch the full Agent Framework introduction (30 min)

Is this the right framework for you?

MAF is a strong fit if you:

  • are building agents and workflows you expect to run in production,
  • need orchestration beyond a single prompt or stateless chat loop,
  • want graph-based patterns such as sequential, concurrent, group collaboration, and custom workflow routing,
  • care about checkpointing, restartability, observability, governance, or human-in-the-loop control,
  • need provider flexibility so your architecture can evolve without major rewrites.

Key Features

Explore new MAF capabilities and real implementation patterns on the official blog.

For a detailed .NET-to-Go feature comparison, see the .NET and Go SDK feature comparison. In short, the Go SDK covers core agents, tools, middleware, workflows, observability, and interoperability integrations, while .NET currently has broader product integrations and several features that are not implemented yet in Go.

  • Go Support: Go packages, examples, and APIs for building agents and workflows in Go.
  • Multiple Agent Provider Support: Support for various LLM and agent providers, with more being added continuously.
  • Middleware: Flexible middleware for request/response processing, logging, OpenTelemetry, context providers, tool approval, and automatic tool calling.
  • Orchestration Patterns & Workflows: Build multi-agent systems with graph-based workflows supporting sequential, concurrent, group collaboration, conditional routing, subworkflows, checkpointing, streaming, human-in-the-loop, and time-travel patterns. Handoff orchestration is not implemented yet in the Go SDK.
  • Microsoft Foundry Agents: Build project-backed Foundry agents, invoke existing server-side Foundry agents, pass Foundry client headers, capture served-model metadata, and use Foundry memory. Foundry-hosted deployment is not implemented yet in the Go SDK.
  • Observability: OpenTelemetry integration for distributed tracing, monitoring, and debugging.
  • Declarative Agents: Not implemented yet in the Go SDK.
  • Agent Skills: Build domain-specific knowledge bases from files, inline definitions, and scripts for agents to discover and use.
  • AF Labs: Not implemented yet in the Go SDK.
  • DevUI: Not implemented yet in the Go SDK.

Table of Contents

Getting Started

Installation

Go

go get github.com/microsoft/agent-framework-go
Learning Resources
  • Overview - High level overview of the framework
  • Quick Start - Get started with a simple agent
  • Tutorials - Step by step tutorials
  • User Guide - In-depth user guide for building agents and workflows
Quickstart
Basic Agent - Go

Create a simple Microsoft Foundry agent that writes a haiku about the Microsoft Agent Framework.

package main

import (
	"cmp"
	"context"
	"fmt"
	"os"

	"github.com/Azure/azure-sdk-for-go/sdk/azidentity"
	"github.com/microsoft/agent-framework-go/provider/foundryprovider"
)

func main() {
	endpoint := os.Getenv("FOUNDRY_PROJECT_ENDPOINT")
	model := cmp.Or(os.Getenv("FOUNDRY_MODEL"), "gpt-4o-mini")

	// Authenticate to Microsoft Foundry.
	token, err := azidentity.NewDefaultAzureCredential(nil)
	if err != nil {
		panic(err)
	}

	// Create a Microsoft Foundry agent.
	a := foundryprovider.NewAgent(endpoint, token, foundryprovider.ModelDeployment(model),
		foundryprovider.AgentConfig{
			Instructions: "You are a helpful assistant.",
		},
	)

	// Run the agent.
	ctx := context.Background()
	fmt.Println(a.RunText(ctx, "Write a haiku about the Microsoft Agent Framework").Collect())
}

More Examples & Samples

Go
  • Getting Started: progressive tutorial from hello world to workflows
  • Agent Concepts: deep-dive samples by topic, including tools, middleware, providers, observability, A2A, AG-UI, MCP, and skills
  • Workflows: multi-agent patterns, routing, checkpointing, observability, and workflow orchestration
  • End-to-End: full applications and demos
  • Feature Comparison: .NET and Go SDK feature parity notes

Community & Feedback

  • Found a bug? File a GitHub issue to help us improve.
  • Enjoying MAF for Go? GitHub stars to show your support and help others discover the project.
  • Have questions? Join our Discord.

Troubleshooting

Authentication
Problem Cause Fix
Authentication errors when using Azure credentials Not signed in to Azure CLI or another configured credential source Run az login before starting your app, or configure the specific credential your app uses
API key errors Wrong or missing API key Verify the key and ensure it is for the correct resource or provider
Provider endpoint errors Missing or incorrect endpoint, deployment, model, or API version Check the environment variables and constructor options used by the sample or provider

Tip: DefaultAzureCredential is convenient for development but in production, consider using a specific credential, such as managed identity, to avoid latency issues, unintended credential probing, and potential security risks from fallback mechanisms.

Environment Variables

For environment variable configuration specific to each sample, refer to the README or source in the sample directory under examples.

Contributor Resources

Important Notes

[!IMPORTANT] If you use Microsoft Agent Framework to build applications that operate with any third-party servers, agents, code, or non-Azure Direct models ("Third-Party Systems"), you do so at your own risk. Third-Party Systems are Non-Microsoft Products under the Microsoft Product Terms and are governed by their own third-party license terms. You are responsible for any usage and associated costs.

We recommend reviewing all data being shared with and received from Third-Party Systems and being cognizant of third-party practices for handling, sharing, retention, and location of data. It is your responsibility to manage whether your data will flow outside of your organization's Azure compliance and geographic boundaries and any related implications, and that appropriate permissions, boundaries, and approvals are provisioned.

You are responsible for carefully reviewing and testing applications you build using Microsoft Agent Framework in the context of your specific use cases, and making all appropriate decisions and customizations. This includes implementing your own responsible AI mitigations such as metaprompt, content filters, or other safety systems, and ensuring your applications meet appropriate quality, reliability, security, and trustworthiness standards. See also: Transparency FAQ.

Telemetry and data collection

This repository does not configure Microsoft telemetry collection by default. Some packages and samples include optional OpenTelemetry instrumentation or connect to Microsoft-hosted services. To turn off framework instrumentation, do not configure OpenTelemetry exporters or telemetry middleware in your application. Service telemetry, if any, is governed by the services you choose to call.

Preview status

The Agent Framework for Go is in public preview. Declarative agents, RAG, CodeAct, and functional workflows are not yet available. File issues on GitHub at https://github.com/microsoft/agent-framework-go/issues.

Trademarks

This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft trademarks or logos is subject to and must follow Microsoft's Trademark & Brand Guidelines. Use of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship. Any use of third-party trademarks or logos is subject to those third-party policies.

Directories

Path Synopsis
Package agent provides the provider-agnostic agent abstraction at the core of the framework.
Package agent provides the provider-agnostic agent abstraction at the core of the framework.
compaction
Package compaction reduces conversation history to fit a model's context window.
Package compaction reduces conversation history to fit a model's context window.
format/jsonformat
Package jsonformat provides JSON schema-based response formatting and validation for structured agent output, deriving schemas from Go types and validating and normalizing values against them.
Package jsonformat provides JSON schema-based response formatting and validation for structured agent output, deriving schemas from Go types and validating and normalizing values against them.
harness/agentmode
Package agentmode provides a context provider that tracks the agent's operating mode (e.g.
Package agentmode provides a context provider that tracks the agent's operating mode (e.g.
harness/loop
Package loop provides middleware that re-invokes an agent until one of its evaluators decides no more work is needed.
Package loop provides middleware that re-invokes an agent until one of its evaluators decides no more work is needed.
harness/todo
Package todo provides a context provider that gives agents todo list management tools for tracking work items during long-running complex tasks.
Package todo provides a context provider that gives agents todo list management tools for tracking work items during long-running complex tasks.
harness/toolapproval
Package toolapproval provides middleware that manages human-in-the-loop tool approval with support for "don't ask again" standing rules.
Package toolapproval provides middleware that manages human-in-the-loop tool approval with support for "don't ask again" standing rules.
harness/toolautocall
Package toolautocall provides the middleware that automatically invokes the function tools a model calls and feeds their results back for the next turn.
Package toolautocall provides the middleware that automatically invokes the function tools a model calls and feeds their results back for the next turn.
skills
Package skills provides domain knowledge bases ("skills") an agent can load on demand.
Package skills provides domain knowledge bases ("skills") an agent can load on demand.
skills/fsskills
Package fsskills discovers and loads skills from a filesystem (an fs.FS), materializing skill directories, their resources, and file-backed scripts as skills.Skill values.
Package fsskills discovers and loads skills from a filesystem (an fs.FS), materializing skill directories, their resources, and file-backed scripts as skills.Skill values.
cmd
prfromissue command
Command prfromissue opens pull requests for gh-aw fallback tracking issues and links the two together.
Command prfromissue opens pull requests for gh-aw fallback tracking issues and links the two together.
prriskclassifier command
Command prriskclassifier applies deterministic Agent Framework risk rules to a pull request and reports whether semantic agent classification is needed.
Command prriskclassifier applies deterministic Agent Framework risk rules to a pull request and reports whether semantic agent classification is needed.
verifyexamples command
examples
02-agents/agents/step09_dependency_injection command
This sample shows how to use dependency injection to register an Agent and use it from a hosted service with a user input chat loop.
This sample shows how to use dependency injection to register an Agent and use it from a hosted service with a user input chat loop.
02-agents/agents/step13_middleware command
This sample shows how to write custom agent middleware.
This sample shows how to write custom agent middleware.
02-agents/providers/azure/openai_responses_background command
This sample demonstrates long-running agent operations using the OpenAI Responses API "background" option.
This sample demonstrates long-running agent operations using the OpenAI Responses API "background" option.
demos/chat_cli command
internal
hashmap
Package hashmap provides a hash table with custom hashing and key equivalence.
Package hashmap provides a hash table with custom hashing and key equivalence.
maphash
Package maphash provides temporary hash interfaces used by hash-based containers.
Package maphash provides temporary hash interfaces used by hash-based containers.
Package message defines the message and content primitives that are the universal currency of the framework.
Package message defines the message and content primitives that are the universal currency of the framework.
messagefilter
Package messagefilter provides composable filters over slices of messages, used to select which messages a context provider stores or forwards.
Package messagefilter provides composable filters over slices of messages, used to select which messages a context provider stores or forwards.
messageworkflow
Package messageworkflow provides helpers for workflow executors that buffer incoming messages and act when a turn token arrives, used by the agent-hosting and group-chat executors in workflow/agentworkflow.
Package messageworkflow provides helpers for workflow executors that buffer incoming messages and act when a turn token arrives, used by the agent-hosting and group-chat executors in workflow/agentworkflow.
provider
a2aprovider
Package a2aprovider exposes agents over the Agent2Agent (A2A) protocol, adapting A2A messages to and from the framework message model and including server and hosting support.
Package a2aprovider exposes agents over the Agent2Agent (A2A) protocol, adapting A2A messages to and from the framework message model and including server and hosting support.
aguiprovider
Package aguiprovider exposes agents over the AG-UI protocol, streaming agent responses as AG-UI events with HTTP handler and hosting support.
Package aguiprovider exposes agents over the AG-UI protocol, streaming agent responses as AG-UI events with HTTP handler and hosting support.
anthropicprovider
Package anthropicprovider provides Anthropic (Claude)-backed agents via NewAgent, which takes an Anthropic client and an AgentConfig.
Package anthropicprovider provides Anthropic (Claude)-backed agents via NewAgent, which takes an Anthropic client and an AgentConfig.
copilotprovider
Package copilotprovider provides GitHub Copilot-backed agents via NewAgent, mapping Copilot session events to framework response updates.
Package copilotprovider provides GitHub Copilot-backed agents via NewAgent, mapping Copilot session events to framework response updates.
foundryprovider
Package foundryprovider provides Microsoft Foundry provider integrations for agents.
Package foundryprovider provides Microsoft Foundry provider integrations for agents.
geminiprovider
Package geminiprovider provides Google Gemini-backed agents via NewAgent, which takes a genai client and an AgentConfig.
Package geminiprovider provides Google Gemini-backed agents via NewAgent, which takes a genai client and an AgentConfig.
openaiprovider
Package openaiprovider provides OpenAI-backed agents.
Package openaiprovider provides OpenAI-backed agents.
otelprovider
Package otelprovider wraps an agent with OpenTelemetry tracing via NewMiddleware, recording spans and token usage for agent runs following the GenAI semantic conventions.
Package otelprovider wraps an agent with OpenTelemetry tracing via NewMiddleware, recording spans and token usage for agent runs following the GenAI semantic conventions.
Package tool defines the Tool interface and its variants (SchemaTool, FuncTool) that agents expose to models.
Package tool defines the Tool interface and its variants (SchemaTool, FuncTool) that agents expose to models.
agenttool
Package agenttool adapts an agent.Agent into a tool.FuncTool via New, so one agent can be invoked as a tool by another.
Package agenttool adapts an agent.Agent into a tool.FuncTool via New, so one agent can be invoked as a tool by another.
functool
Package functool turns a typed Go function into a tool.FuncTool via New and MustNew, deriving the tool's input and output JSON schemas from the function's parameter and return types.
Package functool turns a typed Go function into a tool.FuncTool via New and MustNew, deriving the tool's input and output JSON schemas from the function's parameter and return types.
hostedtool
Package hostedtool provides definitions for hosted tools.
Package hostedtool provides definitions for hosted tools.
mcptool
Package mcptool provides integration with the Model Context Protocol (MCP).
Package mcptool provides integration with the Model Context Protocol (MCP).
shelltool
Package shelltool provides a shell command execution tool that can be registered with an agent.
Package shelltool provides a shell command execution tool that can be registered with an agent.
Package workflow is a graph-execution engine for multi-agent orchestration.
Package workflow is a graph-execution engine for multi-agent orchestration.
agentworkflow
This file hosts an agent.Agent as a workflow workflow.Executor, so the agent can participate in graphs alongside regular executors and other hosted agents.
This file hosts an agent.Agent as a workflow workflow.Executor, so the agent can participate in graphs alongside regular executors and other hosted agents.
checkpoint
Package checkpoint provides checkpoint storage and management for workflow runs, enabling restartability and time-travel resume.
Package checkpoint provides checkpoint storage and management for workflow runs, enabling restartability and time-travel resume.
inproc
Package inproc provides the in-process workflow runtime: it instantiates a workflow's executors and drives execution, streaming or to completion, with optional checkpointing.
Package inproc provides the in-process workflow runtime: it instantiates a workflow's executors and drives execution, streaming or to completion, with optional checkpointing.
observability
Package observability defines the tracing abstraction (Tracer, spans, and attributes) used to instrument workflow execution; the opentelemetry subpackage provides an OpenTelemetry implementation.
Package observability defines the tracing abstraction (Tracer, spans, and attributes) used to instrument workflow execution; the opentelemetry subpackage provides an OpenTelemetry implementation.
observability/opentelemetry
Package opentelemetry implements the workflow observability Tracer using OpenTelemetry, emitting spans for workflow and executor activity.
Package opentelemetry implements the workflow observability Tracer using OpenTelemetry, emitting spans for workflow and executor activity.

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL