Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add AlexK020908/infra-designer/plugin install infra-designerWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/alexk020908/infra-designer/plan-infra)<a href="https://agentmods.dev/commands/alexk020908/infra-designer/plan-infra"><img src="https://agentmods.dev/badge/commands/alexk020908/infra-designer/plan-infra/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/alexk020908/infra-designer/plan-infra"><img src="https://agentmods.dev/badge/commands/alexk020908/infra-designer/plan-infra.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00023 | $0.02899 |
| Opus 5 | $0.00012 | $0.01450 |
| Sonnet 5 | $0.00005 | $0.00580 |
| Haiku 4.5 | $0.00002 | $0.00290 |
Grade A, and why
plan-infra 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 9d 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.
How it starts
The opening of the file, as written. The whole thing — 214 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Plan infrastructure
You are designing cloud infrastructure with the user and capturing it as a canonical, reproducible spec. The spec — not the chat — is the source of truth.
User request: $ARGUMENTS
1. Establish requirements and CONFIRM them — mandatory gate
A one-line request ("build me a URL shortener") is never enough to design from. You MUST confirm requirements with the user before designing. This gate is unconditional — it applies even in auto / non-interactive mode. Do this every time:
-
Draft the requirements you would design to. Infer sensible defaults from the request; mark anything you inferred as an assumption (don't pretend the user said it). Cover:
- Functional — decompose the system into distinct services/paths. Don't lump everything into one "API." Name each read path, write path, and background job separately, since they often scale differently and may become different nodes. (URL shortener: redirect read path — tiny, cache-heavy, huge volume; create-link write path — low-volume, durable; optional click analytics — async. Three different concerns.)
- Non-functional — the numbers that pick and size components. Reads/s and writes/s separately, read:write ratio, object count/growth, latency targets, consistency, durability, availability/HA, security/compliance.
- Constraints. Budget/month, region, cloud provider, ops appetite.
-
Present the draft back as a concise, concrete list (group it as above, and flag each inferred item with "(assumed)").
-
Ask explicitly: "Is this what you expect? Tell me what to change."
-
STOP and wait for the user's reply. Do NOT read components, write the spec, or render anything until they confirm or correct. End your turn here. This hold applies in auto mode too — emit the requirements and the question, then yield. Only proceed once the user has responded; fold their corrections in (re-confirm if they changed anything material).
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.
- 9d ago First seen · 214 lines · 23 tokens per session scan A 9aa9558a3f71
plan-infra is a command published in the GitHub repository AlexK020908/infra-designer (2 stars, last pushed 2mo ago), licensed MIT. It adds 23 tokens to every session and 2,899 once invoked, about $0.0001 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.
Other commands, from other repositories
k8s-manifest
Generate production-ready Kubernetes manifests for the current application.
setup-lambda
Configure and deploy an AWS Lambda function with proper settings.
uninstall
Uninstall filetree from this repo — un-wires CLAUDE.md/AGENTS.md, removes generated artifacts (confirmed per item), tears down self-only mode if enabled.
rclone_sync
Make source and dest identical, modifying destination only.
rclone_serve_s3
Serve remote:path over s3.
checklist
Generate a custom checklist for the current feature based on user requirements.