pulumi-neo_skill

This skill helps you manage Pulumi Neo tasks and infrastructure via natural language, simplifying automation, analysis, and task workflows for cloud engineers.

Python

0

GitHub Stars

1

Bundled Files

3 weeks ago

Catalog Refreshed

2 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstart where the catalogue uses aiagentskills.

npx veilstart add skill dirien/claude-skills --skill pulumi-neo

SKILL.md9.2 KB

Overview

This skill provides a conversational AI agent for managing Pulumi infrastructure using natural language. It lets platform engineers create, inspect, and approve infrastructure tasks via the Pulumi Neo API or included helper scripts. Use it to analyze stacks, generate Pulumi code, and drive multi-step maintenance workflows without manual API plumbing.

How this skill works

The skill creates Neo tasks by posting user messages and optional entity context (stack, repository, pull request, policy issue) to the Pulumi Neo API. It can poll task events to surface agent responses, detect approval requests, and submit approvals or cancellations. A Python helper script is provided for task creation, listing, polling, and event retrieval; direct curl calls to the Neo endpoints are also supported.

When to use it

Ask for an infrastructure analysis or security review of a Pulumi stack
Generate or refactor Pulumi code from a natural language description
Request multi-step platform operations like upgrades, migrations, or CI/CD setup
Review infrastructure changes in a pull request or fix policy violations
When you want conversational, audit-friendly proposals that can require approval

Best practices

Always attach relevant entity context (stack, repo, PR, policy_issue) to tasks
Provide specific, constrained instructions (regions, budget, compliance) for predictable results
Use --no-poll for CI or programmatic calls to avoid blocking loops
Check task events for tool_calls and approval_request IDs rather than relying on status alone
Review Neo’s proposed changes and generated pull requests before approving

Example use cases

Analyze production stack for security and compliance issues and get remediation steps
Generate a new Pulumi TypeScript project for an AWS ECS web app
Review infrastructure changes in a GitHub PR and get suggested fixes
Create a task to upgrade a Kubernetes cluster and receive a staged plan
Detect and remediate policy violations across stacks with guided code changes

FAQ

A Pulumi Cloud account with Neo access and PULUMI_ACCESS_TOKEN set as an environment variable. The token must have access to the target organization.

How do I supply stack or repo context?

Include entity details when creating a task: stack (name+project), repository (name+org+forge), or pull_request (number+repo). The helper script accepts flags for these.

Why is my script hanging?

Interactive polling runs continuously. Use --no-poll for non-interactive use or CI to avoid blocking.