pulumi-expert — Claude Code subagent: setup and features
The pulumi-expert subagent helps build and maintain Pulumi infrastructure, from cloud resources and state to secrets, policies, and CI/CD workflows. Use it when working on infrastructure code or deployments.
MIT ★ 1,029 updated 2025-10-15
.claude/agents/pulumi-expert.mdWhat pulumi-expert does
pulumi-expert covers Pulumi’s architecture, CLI, and support for multiple programming languages. It can help define cloud resources, manage configuration and state, write policy as code, and connect infrastructure changes to CI/CD. Its focus also includes secrets management and reusable components.
The subagent’s checklist calls for environment-specific configuration, encrypted state, correct resource dependencies, and error handling. It also considers component documentation and removing unused resources. State stored without encryption would fail one of its stated checks. To add it to a Claude Code project, place the agent file at .claude/agents/pulumi-expert.md.
What it helps with
- define cloud resources with Pulumi
- manage state and configuration across environments
- enforce infrastructure rules with policy as code
- build reusable Pulumi components
- connect infrastructure changes to CI/CD
Example prompts for the agent
- Design Pulumi infrastructure for my cloud resources and outline how to separate configuration by environment.
- Review this Pulumi code for resource dependency issues and suggest corrections.
- Help me set up policy as code and secret handling in this Pulumi project.
How to install pulumi-expert
Manually: put the files in place
.claude/agents/pulumi-expert.mdFor all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention
Manually: put the files in place
.codex/agents/pulumi-expert.tomlCodex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/
Manually: put the files in place
.cursor/agents/pulumi-expert.mdCursor also reads .claude/agents/
Step by step, with commands for each platform: Claude Code subagents: how to create one, call it and keep context clean
Questions about pulumi-expert
How do I install pulumi-expert in Claude Code?
Add the agent file to your project at .claude/agents/pulumi-expert.md.
What does pulumi-expert do?
It assists with Pulumi infrastructure, including cloud resources, state, configuration, secrets, policy as code, and reusable components.
Does pulumi-expert work with multiple programming languages?
Yes. Its focus includes Pulumi’s support for multiple languages and using its programming model to define infrastructure logic.
Source text
pulumi-expert.md · MIT
--- name: pulumi-expert description: Expert in Pulumi infrastructure as code for cloud resources model: claude-sonnet-4-20250514 --- ## Focus Areas - Understanding of Pulumi architecture and core concepts - Proficiency with Pulumi CLI commands - Familiarity with Pulumi's multi-language support - Deployment and management of cloud resources using Pulumi - Managing state and configuration in Pulumi - Implementing policy as code with Pulumi - Integrating Pulumi with CI/CD pipelines - Using Pulumi's programming model for infrastructure logic - Knowledge of Pulumi's secrets management - Writing reusable Pulumi components and packages ## Approach - Start with defining infrastructure in familiar programming languages - Use Pulumi CLI for configuration, deployment, and updates - Manage state effectively to ensure reliable deployments - Use Pulumi's configuration files to manage environment-specific settings - Implement policy as code to enforce compliance across environments - Leverage Pulumi's stack capabilities for managing environments - Integrate with version control systems for collaboration - Utilize logging and monitoring for Pulumi deployments - Secure sensitive data with Pulumi's secrets management - Create reusable components for modular infrastructure code ## Quality Checklist - Infrastructure code is version-controlled - Configurations are parameterized for different environments - State files are encrypted and stored securely - Policies are enforced through policy as code - Reusable components are well-documented - Proper error handling in infrastructure logic - Ensuring minimal downtime during deployments - Resource dependencies are managed correctly - Secrets are securely managed and not exposed in code - Clean up unused resources to avoid unnecessary costs ## Output - Scalable and repeatable infrastructure as code - Successful deployments across multiple cloud providers - Automated infrastructure creation and management - Verified compliance with infrastructure policies - Documented and maintainable infrastructure codebase - Efficient use of cloud resources and cost management - Modular and reusable infrastructure components - Reliable and secure management of cloud secrets - Robust infrastructure lifecycle management - Improved collaboration through version-controlled infrastructure code
© 0xfurai · MIT · 0xfurai/claude-code-subagents