architecture-catalog CLAUDE.md

architecture-catalog CLAUDE.md is an instructions file for coding agents from ea-toolkit/architecture-catalog. It costs 2,891 tokens per session, scanned A, original, MIT.

Project instructions for an architecture catalogue: a Git-based collection of Markdown records and draw.io diagrams describing software architecture.

In plain words
What is it for?
They guide work on the registry, its relationships and diagrams, validation scripts, and the Astro website that displays the catalogue.
Why use it?
They explain where architecture data comes from and keep the catalogue, validation, and website views consistent when the model changes.

Instructions file

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/ea-toolkit/architecture-catalog/claude-md
Clone the repo
git clone --depth 1 https://github.com/ea-toolkit/architecture-catalog

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 architecture-catalog CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/ea-toolkit/architecture-catalog/claude-md.svg)](https://agentmods.dev/instructions/ea-toolkit/architecture-catalog/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/ea-toolkit/architecture-catalog/claude-md"><img src="https://agentmods.dev/badge/instructions/ea-toolkit/architecture-catalog/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 2,891 This file is loaded in full into every session.
When invoked 2,891 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 $0.02891 $0.02891
Opus 5 $0.01445 $0.01445
Sonnet 5 $0.00578 $0.00578
Haiku 4.5 $0.00289 $0.00289

Measured 4d ago against content hash 8998da6137a3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

architecture-catalog CLAUDE.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 4d 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.

CLAUDE.md · 237 lines

How it starts

The opening of the file, as written. The whole thing — 237 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

What is this project?

A lightweight, Git-native architecture catalog. It replaces monolithic tools like Archi with a simple repo structure: architecture elements are registered as Markdown files, diagrams are draw.io files, and Python scripts validate, extract, refresh, and generate dashboards.

The goal is to be an open-source, lightweight alternative for architects who want Git-friendly, AI-readable architecture models without learning a DSL or paying for enterprise tooling.

Architecture (how it works)

registry-mapping.yaml  ──→  registry-loader.ts  ──→  Astro pages
  (schema: types,             (reads .md files,        (static HTML
   fields, relationships)      resolves refs,           at build time)
                               builds graph)

Single source of truth: models/registry-mapping.yaml drives everything — the UI, validation, and data loading. Adding a new element type = add one entry to this YAML file + create a _template.md. Zero code changes.

Data pipeline: registry-mapping.yaml → registry-loader.ts (loadRegistry()) → registry.ts (bridge) → Astro pages. The loader builds an in-memory graph with elements, edges, and indexes (byType, byDomain, byLayer).

Key design: Framework-agnostic. ArchiMate alignment is optional, not required. Works with any architecture vocabulary. Layers are flat (1 to N), each with sub-folders for element types. No nested child layers.

Current State

Registry (registry-v2/)

  • 4-layer structure — flat layers, each with typed sub-folders
  • Template-driven — each element type has a _template.md with typed YAML frontmatter
  • Example domains: "Customer Management", "Billing & Payments", "Analytics & Insights" — 3 fictional B2B SaaS CRM domains with 71 elements across all 4 layers
  • Health: 71 healthy, 71 connected, 0 orphans, 0 broken refs, 84 pages

Catalog UI (catalog-ui/)

  • Astro 5 + React static site for browsing the registry
  • Schema-driven — all types, layers, and relationships derived from models/registry-mapping.yaml
  • ReactFlow graphs for visualizing element relationships (via @xyflow/react + dagre)
  • Diagram viewer supporting PlantUML, BPMN, and draw.io formats
  • CSS pattern: EventCatalog-inspired RGB variable pattern: --ec-page-bg: 255 255 255 used as rgb(var(--ec-page-bg))
  • 3-panel layout: icon bar (56px) + nested sidebar (315px) + main content

Read the full file on GitHub · 237 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. 4d ago First seen · 237 lines · 2,891 tokens per session scan A 8998da6137a3

Subscribe to this mod's changes

architecture-catalog CLAUDE.md is an instructions file published in the GitHub repository ea-toolkit/architecture-catalog (38 stars, last pushed 4mo ago), licensed MIT. It adds 2,891 tokens to every session, about $0.0145 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 2026-08-30.