# Create Clear C4 Architecture Diagrams

Architecture discussions become confusing when diagrams mix detail levels or omit relationships. This skill creates structured C4 views and reviews them against consistent guidelines.

## Install

```bash
npx skillstore add consiliency/c4-modeling
```

## Metadata

- Status: approved
- Slug: consiliency-c4-modeling
- Skillstore revision: r1
- Version status: missing
- Tree hash: 718369fb833e552183ee59de19006256e2bc0c7822bf58796d55822e36ea0c00
- Author: Consiliency
- GitHub username: Consiliency
- License: MIT
- Repository: https://github.com/Consiliency/treesitter-chunker/tree/main/.ai-dev-kit/skills/c4-modeling
- Ref: 635f69fb8d2f4e6330ba47a4e5a0fb239c04d110
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: network, external\_commands
- Quality score: 79
- Quality tier: bronze
- Public page: https://skillstore.pages.dev/skills/consiliency-c4-modeling
- Manifest: https://skillstore.pages.dev/api/skills/consiliency-c4-modeling/manifest

## Capabilities

- Creates context, container, component, and selective code-level C4 views from supplied architecture details.
- Organizes people, systems, containers, components, boundaries, technologies, and labeled relationships.
- Produces Mermaid C4 syntax using documented elements, relationships, directions, boundaries, and styles.
- Guides transitions between C4 levels while preserving external systems and relevant context.
- Reviews diagrams for mixed abstractions, missing descriptions, unlabeled relationships, unclear boundaries, and excessive detail.

## Use Cases

- Plan a new system: Build context and container views before selecting services, data stores, and integration boundaries.
- Align technical stakeholders: Present consistent architecture views that separate business context from runtime and implementation details.
- Review an existing design: Check diagrams for mixed levels, missing relationships, unclear responsibilities, and unnecessary detail.

## Prompt Templates

### Create a context diagram

```
Create a Level 1 C4 context diagram for [system]. Include [users], [external systems], and labeled relationships. Ask for missing essential details.
```

### Expand into containers

```
Using this approved context view, create a Level 2 container diagram for [system]. Show applications, services, data stores, queues, technologies, and relationships.
```

### Model container components

```
Create a Level 3 component diagram for [container]. Include major modules, responsibilities, repositories, integrations, boundaries, and connections to other containers.
```

### Review and refine C4 views

```
Review these C4 diagrams for abstraction mixing, omissions, unclear labels, and excessive detail. Return prioritized findings and revised Mermaid diagrams for each level.
```

## Limitations

- Requires accurate architecture details from the user and does not discover deployed systems automatically.
- Provides detailed syntax examples for Mermaid, but not equivalent references for Structurizr or PlantUML.
- Offers conceptual Level 4 guidance without a dedicated code-diagram syntax reference.
- Does not render, validate, publish, or automatically synchronize diagrams with source code.

## Best Practices

- Start with a context diagram unless an approved higher-level view already exists.
- State the audience and purpose before selecting the C4 level and included details.
- Give every element a clear description and every relationship a meaningful label.

## Anti Patterns

- Do not create a detailed container or component view before establishing system context.
- Do not combine people, containers, components, classes, and functions in one diagram.
- Do not invent dependencies, technologies, or responsibilities when the source information is incomplete.

## Security Audit

- Audited at: 2026-07-13T15:42:37.641\+00:00
- Summary: All 31 static findings are false positives caused by Markdown code fences, inline code formatting, headings, links, and illustrative C4 relationships. The skill contains documentation only, with no executable commands, network operations, prompt injection, or data handling behavior.

## Stats

- Views: 212
- Downloads: 11
- Favorites: 0
- Popularity score: 0