What is the Pulumi MCP Server?

The Pulumi MCP Server gives AI agents like Claude, ChatGPT, and Cursor direct access to Pulumi. Manage cloud infrastructure via Pulumi — list stacks, track deployments, audit outputs and tag resources from any AI agent. Powered by the Vinkius AI Gateway — no API keys, no infrastructure, connect in under 2 minutes.

Pulumi MCP Server: see your AI Agent in action

AI Agent→Vinkius→Pulumi
You

Vinkius AI Gateway
GDPR·High Security·Kill Switch·Ultra-Low Latency·Plug and Play

Built-in capabilities (11)

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

What this connector unlocks

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

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)

How it works

1. Subscribe to this server
2. Enter your Pulumi Access Token
3. Start managing your infrastructure from Claude, Cursor, or any MCP-compatible client

No more switching to the Pulumi Console to check deployment status or review stack outputs. Your AI acts as a dedicated infrastructure operations engineer.

Who is this for?

  • DevOps Engineers — quickly check stack status, review deployment history and inspect outputs without opening the Pulumi Console
  • Platform Teams — audit infrastructure changes, track deployment success rates and manage stack tags across organizations
  • Developers — discover available stacks, review exported endpoints and verify resource provisioning via conversation

Frequently asked questions

Give your AI agents the power of Pulumi

Access Pulumi and 2,500+ MCP servers — ready for your agents to use, right now. No glue code. No custom integrations. Just plug Vinkius AI Gateway and let your agents work.