loadout terraform.instructions.md

A set of rules for managing AWS cloud infrastructure with Terraform. Terraform is a tool that describes servers, networks, permissions, and other infrastructure in files so they can be reviewed and applied consistently.

In plain words
What is it for?
Use them when creating or changing Terraform configurations, AWS networking and identity, deployment environments, logging, alerts, or infrastructure documentation.
Why use it?
They reduce guesswork around infrastructure changes and cover areas such as security, state, environments, monitoring, and operational readiness.

Instructions file for GitHub Copilot

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add instructions/stefaniuk/loadout/terraform
Clone the repo
git clone --depth 1 https://github.com/stefaniuk/loadout

Made for: GitHub Copilot.

Per session 10,716 This file is loaded in full into every session.
When invoked 10,716 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.10716 $0.10716
Opus 5 $0.05358 $0.05358
Sonnet 5 $0.02143 $0.02143
Haiku 4.5 $0.01072 $0.01072

Measured 2d ago against content hash b06af6a439ad, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

loadout terraform.instructions.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.github/instructions/terraform.instructions.md · 787 lines

How it starts

The opening of the file, as written. The whole thing — 787 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Terraform Engineering Instructions (AWS) ☁️

These instructions define the default engineering approach for building and evolving AWS infrastructure using Terraform (Infrastructure as Code).

They must remain applicable to:

  • Terraform configuration (.tf), variables, and modules
  • Backend/state, environments, and release workflows
  • Security, networking, identity, logging, and compliance controls
  • Operational readiness (monitoring, alerting, runbooks)
  • Documentation and diagrams for infrastructure and operations

They are non-negotiable unless an exception is explicitly documented (with rationale and expiry) in an ADR/decision record.

Identifier scheme. Every normative rule carries a unique tag in the form [TF-<prefix>-NNN], where the prefix maps to the containing section (for example OP for Operating Principles, LCL for Local-first developer experience, QG for Quality Gates, continuing through AI for AI-assisted expectations). Use these identifiers when referencing, planning, or validating requirements.


0. Quick reference (apply first) 🧠

This section exists so humans and AI assistants can reliably apply the most important rules even when context is tight.

  • [TF-QR-001] Specification first: if a requirement is not specified (or agreed), do not invent it ([TF-OP-005]).
  • [TF-QR-002] Small, safe changes: prefer incremental evolution over big rewrites ([TF-OP-003]).
  • [TF-QR-003] Deterministic and reproducible: the same inputs must produce the same plan ([TF-OP-002], [TF-OP-008]–[TF-OP-009]).
  • [TF-QR-004] Run the quality gates after any Terraform change and iterate to clean ([TF-QG-001]–[TF-QG-008]).
  • [TF-QR-005] Plan-first discipline: all changes must be reviewed via terraform plan output ([TF-BEH-001]–[TF-BEH-003]).
  • [TF-QR-006] London-only residency: all infrastructure must be in eu-west-2; no other region without an ADR ([TF-REG-001]–[TF-REG-002]).
  • [TF-QR-007] Secure-by-default: least privilege, encryption, auditable changes ([TF-OP-004], [TF-SEC-001]–[TF-SEC-026]).
  • [TF-QR-008] No snowflakes: no manual changes in the AWS Console that are not represented in Terraform ([TF-OP-006]).
  • [TF-QR-009] Remote state, isolated per environment: encrypted, versioned, access-controlled ([TF-STATE-001]–[TF-STATE-008]).
  • [TF-QR-010] Observability baseline: CloudWatch logs, metrics, alarms, dashboards, and runbook references ([TF-OBS-001]–[TF-OBS-067]).
  • [TF-QR-011] IaC or nothing: every AWS change must flow through Terraform under version control ([TF-OP-007]).
  • [TF-QR-012] Avoid common anti-patterns: hardcoded ARNs, floating versions, mega-modules, secrets in .tfvars (§20).

Read the full file on GitHub · 787 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 787 lines · 10,716 tokens per session scan A b06af6a439ad

Subscribe to this mod's changes

loadout terraform.instructions.md is an instructions file published in the GitHub repository stefaniuk/loadout (1 stars, last pushed 4d ago), licensed MIT. It adds 10,716 tokens to every session, about $0.0536 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.