erd-studio

A guide to the file format used by ERD Studio, a tool for designing entity-relationship diagrams and data models.

In plain words
What is it for?
Use it when working in .erd-studio/ to define reusable data models, connect them to domain diagrams, and maintain relationships and layouts.
Why use it?
It prevents edits from breaking the required relationship between shared YAML model definitions and JSON diagrams.

Skill for Claude CodeCodex

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 skills/liam-machine/erd-studio/erd-studio
Any agent
npx skills add liam-machine/erd-studio --skill erd-studio
Clone the repo
git clone --depth 1 https://github.com/liam-machine/erd-studio

Made for: Claude Code, Codex.

Per session 162 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,866 The whole file, excluding the scripts and references it only reads on demand.
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 $0.00162 $0.03866
Opus 5 $0.00081 $0.01933
Sonnet 5 $0.00032 $0.00773
Haiku 4.5 $0.00016 $0.00387

Measured 3d ago against content hash bc8e8486a6d0, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

erd-studio 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 3d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (enforce-skill.sh), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

test/fixtures/dbt-project/.claude/skills/erd-studio/SKILL.md · 328 lines

How it starts

The opening of the file, as written. The whole thing — 328 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ERD Studio — AI Data Modeling Guide

ERD Studio uses a central model store architecture. Model definitions are YAML files in .erd-studio/logical-models/. Domain JSON files reference models by name and define relationships and layout.

Architecture Overview

.erd-studio/
├── logical-models/           ← Central model definitions (YAML, one per model)
│   ├── dim_customer.yml
│   ├── dim_project.yml
│   └── fct_sale.yml
├── silver/
│   ├── customer-360.json     ← Domain file (model references + relationships + layout)
│   └── orders.json
└── gold/
    └── reporting.json

Key principle: Models are defined ONCE in logical-models/ and referenced from multiple domain files. Editing a model from any domain updates the shared definition.

Model Library (Sidebar)

The Model Library panel in the ERD Studio sidebar shows all YAML files in logical-models/. Use it to understand the difference between "model definition exists" and "model is referenced by a domain":

  • Referenced models show how many domains use them (e.g. "2 domains")
  • Orphaned models show a warning icon and "(unused)" — these exist as .yml files but are not in any domain's logical.models[] array

Important for AI agents: Before saying a model "already exists in the ERD", check whether it is referenced by the target domain's logical.models[] array — not just whether the .yml file exists. A model file in logical-models/ may be unused (orphaned) or only referenced by other domains.

Domain File Structure

File: .erd-studio/{layer}/{domain}.json

{
  "schemaVersion": 5,
  "domain": "customer-360",
  "layer": "silver",
  "description": "Customer domain — master data and transaction history",
  "modelFolder": "models/silver",
  "stubColumns": ["dim_project"],
  "logical": {
    "models": ["dim_customer", "fct_sale", "dim_product", "dim_project"],
    "relationships": []
  },
  "viewConfig": {}
}
Field Required Description
schemaVersion Yes Must be 5
domain Yes Domain slug (matches filename without .json)
layer Yes Layer name matching parent directory (e.g. silver, gold)
description No Human-readable domain description
modelFolder No Filter for "Add Existing Model" dialog (e.g. models/silver)
stubColumns No Model names whose physical-only columns are suppressed in sync comparison. Use for conformed dimensions and reference tables included only to anchor relationships — they define a few key columns (PK/NK) but not the full physical column set. Missing-column discrepancies are hidden; extra and type-mismatch discrepancies on defined columns still surface.
logical.models Yes Array of model name strings (references to logical-models/*.yml)
logical.relationships Yes Array of relationship objects
viewConfig Yes Root-level view settings. The extension auto-assigns positions for new models

Read the full file on GitHub · 328 lines

Files

What ships with it

2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 3d ago First seen · 328 lines · 162 tokens per session scan A bc8e8486a6d0

Subscribe to this mod's changes

erd-studio is a skill published in the GitHub repository liam-machine/erd-studio (5 stars, last pushed 2mo ago), licensed MIT. It adds 162 tokens to every session and 3,866 once invoked, about $0.0008 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.

Related

Other skills, from other repositories

warehouse-performance

Investigates the performance of an ENTIRE Fabric Data Warehouse (warehouse-wide, not a single query): surfaces long-running and frequent queries, reads SQL pool resource-pressure insights, audits statistics health, checks result-set caching, reviews and tunes SQL pool configuration, and produces a prioritized findings…

sdebruyn/fabric-dw-mcp-cli · 138 tokens

query-optimizer

Analyzes query performance on a Fabric Data Warehouse: captures the estimated execution plan, inspects query insights history, identifies costly operators and missing or stale statistics, inspects data-clustering columns, and proposes or applies optimizations. Use when the user asks to "optimize a query", "why is my…

sdebruyn/fabric-dw-mcp-cli · 103 tokens

dbt-setup

Scaffolds a complete dbt-fabric project for a Fabric Data Warehouse: generates profiles.yml, dbtproject.yml, a column-rich sources.yml (with data types for every table column, ready for dbt contracts), requirements.txt, and .gitignore, then writes them to the local filesystem. Use when the user asks to "set up dbt"…

sdebruyn/fabric-dw-mcp-cli · 115 tokens

dbt-data-transformation

Complete guide for dbt data transformation including models, tests, documentation, incremental builds, macros, packages, and production workflows.

manutej/luxor-claude-marketplace · 30 tokens

Convex Migrations

Schema migration strategies for evolving applications including adding new fields, backfilling data, removing deprecated fields, index migrations, and zero-downtime migration patterns.

powroom/flins · 35 tokens

omh-backend

This is a Hermes-native backend workflow skill.

rlaope/oh-my-hermes · 68 tokens