mitodl/agent-kit/pulumi-modify-existing

Claude Code and agent skill by mitodl/agent-kit

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Modify existing Pulumi infrastructure stacks safely. Use this skill when making any Pulumi IaC changes — always edit the existing stack entrypoint, never create new files, preserve assumeRole and cross-account configuration, and validate with pulumi preview before finishing.

Install

npx vskill@latest install mitodl/agent-kit/pulumi-modify-existing

SKILL.md

Pulumi: Modify Existing Files

Read the repo's conventions first

Before editing, read the repository's own AGENTS.md and any guides it points to (in mitodl/ol-infrastructure: docs/context/03-pulumi-workflows.md, 05-secrets-management.md, 06-code-style-conventions.md). They own stack naming, component rules, helpers, and secrets layout, and they change; don't restate them from memory.

Regardless of repo, these hold:

  • Environment values come from stack config, not literals in __main__.py.

A value that differs between CI, QA, and Production belongs in Pulumi.<project>.<Env>.yaml, read with Config(...).require() when it's required and not secret.

  • Secrets never land in plaintext: use the repo's SOPS files, Vault, or

pulumi config set --secret (secure: values). When a secret config value goes straight into a resource input or an export, read it with require_secret() / get_secret(): require() / get() return a plain str with no secret marking (and no warning), so anything built from it isn't tracked as a secret. Component config models that take a pydantic SecretStr (ol-infrastructure's OLDBConfig.password) need the plain value, so those call sites use require() and the component unwraps it.

  • Pinned versions live where the repo centralizes them (ol-infrastructure:

src/bridge/lib/versions.py, which Renovate updates), not inline.

  • Reuse existing components and helpers (parse_stack(), OL* component

resources, stack outputs such as shared security groups) instead of building raw resources that duplicate them.

Always modify, never create

Infrastructure changes belong in the existing entrypoint for the stack (typically __main__.py). Do not create additional Pulumi files unless the stack already follows a multi-file pattern that makes a new file the right fit.

Preserve existing functionality

When editing a Pulumi stack entrypoint, the following must never be silently removed:

  • `assumeRole` / cross-account role configuration — removing this breaks

production deployments that depend on cross-account access.

  • Existing resource exports (pulumi.export(...)).
  • Stack references to other stacks.

Before finalizing changes, scan the diff to confirm nothing was deleted unintentionally.

Validate changes before declaring done

After modifying any Pulumi code, run a preview to confirm the plan is correct:

bash
pulumi preview --stack <stack-name>

Review the output carefully:

  • Unexpected resource replacements or deletions are bugs, not acceptable

side effects. The common cause is a changed logical resource name, Helm release name, or component name on something that already exists. Keep the original name, or add opts=ResourceOptions(aliases=[...]) so Pulumi treats it as the same resource.

  • "0 changes" is only correct if you genuinely expected no changes.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/mitodl/agent-kit/pulumi-modify-existing/badge)](https://verified-skill.com/skills/mitodl/agent-kit/pulumi-modify-existing)

Something wrong with this skill or its page? Tell us.