How do I get a Pulumi Access Token?

Log in to the [**Pulumi Console**](https://app.pulumi.com), go to **Account Settings > Access Tokens**, click **Create Token**, give it a name and copy the token immediately — it starts with `pul_` and won't be shown again.

What is a Pulumi stack?

A stack is an isolated, independently configurable instance of your Pulumi program. Stacks typically represent different environments like dev, staging and prod. Each stack has its own configuration, state, outputs and deployment history. Use list_stacks to discover all stacks in an organization.

Can I view the outputs of a stack?

Yes! Use `get_stack_outputs` with the org name, project name and stack name. It returns all exported output values from the latest successful deployment, such as URLs, IP addresses, resource IDs and connection strings. This is useful for discovering endpoint addresses after infrastructure deployment.

Pulumi MCP Server for AutoGen 11 tools — connect in under 2 minutes

Name: Pulumi
Availability: InStock
Author: Vinkius

microsoft.github.io/autogen/docs/tools/mcp-tools

Built by Vinkius GDPR 11 Tools Framework

Microsoft AutoGen enables multi-agent conversations where agents negotiate, delegate, and execute tasks collaboratively. Add Pulumi as an MCP tool provider through Vinkius and every agent in the group can access live data and take action.

Get MCP Server for AI Agents

ASK AI ABOUT THIS MCP SERVER

Open in ChatGPT Open in Claude Open in Perplexity

Vinkius supports streamable HTTP and SSE.

python

import asyncio
from autogen_agentchat.agents import AssistantAgent
from autogen_ext.tools.mcp import McpWorkbench

async def main():
    # Your Vinkius token. get it at cloud.vinkius.com
    async with McpWorkbench(
        server_params={"url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"},
        transport="streamable_http",
    ) as workbench:
        tools = await workbench.list_tools()
        agent = AssistantAgent(
            name="pulumi_agent",
            tools=tools,
            system_message=(
                "You help users with Pulumi. "
                "11 tools available."
            ),
        )
        print(f"Agent ready with {len(tools)} tools")

asyncio.run(main())

Fully ManagedVinkius Servers

60%Token savings

High SecurityEnterprise-grade

IAMAccess control

EU AI ActCompliant

DLPData protection

V8 IsolateSandboxed

Ed25519Audit chain

<40msKill switch

Stream every event to Splunk, Datadog, or your own webhook in real-time

* Every MCP server runs on Vinkius-managed infrastructure inside AWS - a purpose-built runtime with per-request V8 isolates, Ed25519 signed audit chains, and sub-40ms cold starts optimized for native MCP execution. See our infrastructure

About Pulumi MCP Server

Connect your Pulumi account to any AI agent and take full control of your infrastructure-as-code through natural conversation.

AutoGen enables multi-agent conversations where agents negotiate, delegate, and collaboratively use Pulumi tools. Connect 11 tools through Vinkius and assign role-based access. a data analyst queries while a reviewer validates, with optional human-in-the-loop approval for sensitive operations.

What you can do

Organization Discovery — List organizations and retrieve their details, team settings and member info
Stack Management — List, create and delete stacks (infrastructure environments) across all your projects
Deployment Tracking — Monitor stack update history with status (succeeded, failed, in-progress), resource changes and error logs
Output Inspection — View exported output values from the latest deployment (URLs, IPs, resource IDs)
Tag Management — List and set custom tags on stacks for organization and filtering (environment, team, cost-center)

The Pulumi MCP Server exposes 11 tools through the Vinkius. Connect it to AutoGen in under two minutes — no API keys to rotate, no infrastructure to provision, no vendor lock-in. Your configuration, your data, your control.

How to Connect Pulumi to AutoGen via MCP

Follow these steps to integrate the Pulumi MCP Server with AutoGen.

Install AutoGen

Run pip install "autogen-ext[mcp]"

Replace the token

Replace [YOUR_TOKEN_HERE] with your Vinkius token

Integrate into workflow

Use the agent in your AutoGen multi-agent orchestration

Explore tools

The workbench discovers 11 tools from Pulumi automatically

Why Use AutoGen with the Pulumi MCP Server

AutoGen provides unique advantages when paired with Pulumi through the Model Context Protocol.

Multi-agent conversations: multiple AutoGen agents discuss, delegate, and collaboratively use Pulumi tools to solve complex tasks

Role-based architecture lets you assign Pulumi tool access to specific agents. a data analyst queries while a reviewer validates

Human-in-the-loop support: agents can pause for human approval before executing sensitive Pulumi tool calls

Code execution sandbox: AutoGen agents can write and run code that processes Pulumi tool responses in an isolated environment

Pulumi + AutoGen Use Cases

Practical scenarios where AutoGen combined with the Pulumi MCP Server delivers measurable value.

Collaborative analysis: one agent queries Pulumi while another validates results and a third generates the final report

Automated review pipelines: a researcher agent fetches data from Pulumi, a critic agent evaluates quality, and a writer produces the output

Interactive planning: agents negotiate task allocation using Pulumi data to make informed decisions about resource distribution

Code generation with live data: an AutoGen coder agent writes scripts that process Pulumi responses in a sandboxed execution environment

Pulumi MCP Tools for AutoGen (11)

These 11 tools become available when you connect Pulumi to AutoGen via MCP:

create_stack

A stack is an isolated, independently configurable instance of your Pulumi program. Requires the org name, project name and stack name (e.g. "staging", "prod"). Returns the created stack with its URL. Create a new Pulumi stack

delete_stack

The stack must be empty (no resources) or force deletion must be enabled. Provide the org name, project name and stack name. WARNING: this action is irreversible. Delete a Pulumi stack

get_current_user

Returns the user's GitHub login, avatar URL, email and name. Use this to verify your access token is working correctly and to see which identity the API calls will appear as. Get the currently authenticated Pulumi user

get_deployment

Provide the org name, project name, stack name and deployment version number. Get details for a specific Pulumi deployment

get_organization

Provide the organization name (slug). Get details for a specific Pulumi organization

get_stack

Provide the org name, project name and stack name. Get details for a specific Pulumi stack

get_stack_outputs

Outputs are values your Pulumi program exports, such as URLs, IP addresses, resource IDs and connection strings. Useful for discovering endpoint addresses and configuration values after infrastructure deployment. Get the exported output values from a Pulumi stack

list_deployments

Each deployment shows its version number, status (succeeded, failed, in-progress), start/end time, resource changes (created, updated, deleted) and the user who triggered it. Use this to audit infrastructure changes and track deployment success/failure patterns. List deployment history for a Pulumi stack

list_stack_tags

Tags are key-value metadata labels used for organizing, filtering and managing stacks (e.g. environment=prod, team=platform, cost-center=engineering). List tags on a Pulumi stack

list_stacks

Each stack represents an isolated, independently configurable instance of your infrastructure (e.g. dev, staging, prod). Returns stack name, project name, last update info, resource count and whether updates are in progress. List all stacks in a Pulumi organization

set_stack_tag

Tags are used for organizing, filtering and managing stacks (e.g. key="environment", value="prod", key="team", value="platform"). Provide the org name, project name, stack name, tag name and tag value. Set a tag on a Pulumi stack

Example Prompts for Pulumi in AutoGen

Ready-to-use prompts you can give your AutoGen agent to start working with Pulumi immediately.

"Show me all stacks in my organization."

"What was the result of the latest deployment to my-infra/prod?"

"Show me the exported outputs from the prod stack."

Troubleshooting Pulumi MCP Server with AutoGen

Common issues when connecting Pulumi to AutoGen through the Vinkius, and how to resolve them.

McpWorkbench not found

Install: pip install "autogen-ext[mcp]"

Pulumi + AutoGen FAQ

Common questions about integrating Pulumi MCP Server with AutoGen.

How does AutoGen connect to MCP servers?

Create an MCP tool adapter and assign it to one or more agents in the group chat. AutoGen agents can then call Pulumi tools during their conversation turns.

Can different agents have different MCP tool access?

Yes. AutoGen's role-based architecture lets you assign specific MCP tools to specific agents, so a querying agent has different capabilities than a reviewing agent.