mcp-architecture

mcp-architecture is a cursor rule for Cursor from kaminari-ad/mcp. It costs 0 tokens per session (1,348 once invoked), scanned A, original, MIT.

An architecture guide for a TypeScript MCP project using domain-driven design and a ports-and-adapters structure. It separates business decisions, application actions, outside integrations, and connection methods.

In plain words
What is it for?
Use it when placing new code, defining interfaces, adding MCP tools, or connecting the project to an API or transport.
Why use it?
It makes dependencies easier to control and lets core logic stay separate from databases, APIs, logging, and transport details.

Cursor rule for Cursor

Written for Cursor: installed under .cursor/.

Good fit Use it when placing new code, defining interfaces, adding MCP tools, or connecting the project to an API or transport.

Compare 6 cursor rules from other repositories ↓
Install with agentmods
npx agentmods add rules/kaminari-ad/mcp/mcp-architecture
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.

Clone the repo
git clone --depth 1 https://github.com/kaminari-ad/mcp

Made for: Cursor.

Wrote 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.

agentmods badge for mcp-architecture

README.md
[![agentmods](https://agentmods.dev/badge/rules/kaminari-ad/mcp/mcp-architecture/github.svg)](https://agentmods.dev/rules/kaminari-ad/mcp/mcp-architecture)
Your own site
<a href="https://agentmods.dev/rules/kaminari-ad/mcp/mcp-architecture"><img src="https://agentmods.dev/badge/rules/kaminari-ad/mcp/mcp-architecture/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.

agentmods 80×15 button for mcp-architecture

Your own site · 80×15
<a href="https://agentmods.dev/rules/kaminari-ad/mcp/mcp-architecture"><img src="https://agentmods.dev/badge/rules/kaminari-ad/mcp/mcp-architecture.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 0 Nothing until a file matches its globs; then the whole rule loads.
When invoked 1,348 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00000 $0.01348
Opus 5.5 $0.00000 $0.00539
Sonnet 5.5 $0.00000 $0.00270
Haiku 4.5 $0.00000 $0.00135

Measured today against content hash e1d14ac46e3f, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

mcp-architecture 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 today.

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.

.cursor/rules/mcp-architecture.mdc · 120 lines

How it starts

The opening of the file, as written. The whole thing — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Architecture Rules — mcp/

This project follows DDD (Domain-Driven Design) + Hexagonal Architecture (Ports & Adapters) with manual constructor injection. The structure mirrors the api repo, adapted for TypeScript.

Layers

src/
├── domain/          # Inner core — depends on nothing outward
│   ├── value-objects/
│   ├── ports/       # interfaces (1 file = 1 port)
│   └── services/    # pure decision policies (no I/O)
├── application/     # use cases
│   ├── tools/       # MCP tool definitions (1 file = 1 tool)
│   └── tool-registry.ts
├── infrastructure/  # adapters implementing ports
│   ├── api/
│   ├── logging/
│   ├── session/
│   ├── rate-limit/
│   ├── clock/
│   └── error-reporting/
├── presentation/    # transports — composition roots
│   ├── stdio/
│   └── http/
├── shared/          # generic utilities (Result, config schema)
└── bin.ts           # CLI entry; dispatches to a bootstrap

Dependency direction (enforced by dependency-cruiser)

presentation ──→ application ──→ domain
                                ↑
infrastructure ─────────────────┘   (implements ports)

shared ←── used by anyone, depends on nothing

Concretely:

  • domain/ imports from nowhere outward. Defines ports; doesn't know what implements them.
  • application/ imports from domain/ only. Tools depend on ApiGateway, Logger, Clock, etc. — never on concrete classes.
  • infrastructure/ implements domain/ports/*. Wires fetch/pino/etc. to satisfy a port. Stays out of application/.
  • presentation/ wires everything together. The *-bootstrap.ts files are the only place new HttpApiGateway(...) exists.
  • shared/ is generic utilities (Result, env-schema). No domain types live here.

ESLint enforces this with no-restricted-imports patterns. dependency-cruiser enforces it again with mcp/.dependency-cruiser.cjs. Two locks beats one.

Ports & Adapters

Every external interaction (HTTP, logging, time, randomness) goes through a port:

Read the full file on GitHub · 120 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. today First seen · 120 lines · 0 tokens per session scan A e1d14ac46e3f

Subscribe to this mod's changes

mcp-architecture is a cursor rule published in the GitHub repository kaminari-ad/mcp (0 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,348 tokens. 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 2026-11-05.