pulumi-automation-api

Orchestrates Pulumi infrastructure operations programmatically using the Automation API.

Updated Apr 30, 2026
One-click install
npx skills add https://github.com/AdityaBorkar/igbot-fork --skill pulumi-automation-api-adityaborkar
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: pulumi-automation-api
Source: https://github.com/AdityaBorkar/igbot-fork/tree/main/.agents/skills/pulumi-automation-api
Command: npx skills add https://github.com/AdityaBorkar/igbot-fork --skill pulumi-automation-api-adityaborkar

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @pulumi/pulumi.

What problem does it solve? Managing infrastructure across multiple Pulumi stacks with Bash scripts or Makefiles leads to fragile orchestration, poor error handling, and manual sequencing. This Skill provides patterns for programmatically controlling Pulumi operations from application code. ## Core Features & Use Cases - Multi-Stack Orchestration: Deploy dependent stacks (infrastructure, platform, application) in correct order, or independent stacks in parallel with Promise.all. - Architecture Guidance: Choose between local source (existing Pulumi projects via workDir) and inline source (embedded program functions) based on team structure and distribution needs. - Programmatic Operations: Set stack configuration, read outputs, stream deployment logs via onOutput callbacks, and handle errors with structured try/catch patterns. - Use Case: Build a self-service internal platform where developers provision environments through a web portal, with the backend calling Automation API to create and destroy stacks on demand. ## Quick Start Ask the AI to write a TypeScript script using Pulumi Automation API that deploys multiple stacks in dependency order with output streaming.

Frequently Asked Questions about pulumi-automation-api

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run Pulumi deployments programmatically without the CLI?▼

Use the Pulumi Automation API by calling LocalWorkspace.createOrSelectStack with a stack name and either a workDir pointing to an existing project or an inline program function. Then call stack.up() to deploy and stack.destroy() to tear down.

How to deploy multiple Pulumi stacks in dependency order?▼

Iterate through stacks sequentially with a for loop, calling stack.up() on each in dependency order, and destroy them in reverse order. For independent stacks, use Promise.all to deploy them in parallel.

What is the difference between local source and inline source in Pulumi Automation API?▼

Local source points workDir at an existing Pulumi project, suiting separate teams and independent versioning. Inline source embeds the Pulumi program as a function in the orchestrator, suiting single teams and compiled binary distribution.

Can the Automation API orchestrator use a different language than the Pulumi programs?▼

Yes, the orchestrator and the Pulumi programs it manages can be written in different languages. For example, a Go orchestrator can manage TypeScript Pulumi programs, letting platform and application teams each use their preferred language.

When should I not use Pulumi Automation API?▼

Avoid it for single projects with standard deployment needs where the regular CLI suffices. It adds value only when you need programmatic control, multi-stack orchestration, or infrastructure embedded in applications.