# What Are the Four Dials That Shape draw.io Imports? > Master draw.io imports with four key dials Format, Size, Detail, and Audience. Control output media, dimensions, fidelity, and styling for editorial-quality diagrams. - Repository: [Cathryn Lavery/diagram-design](https://github.com/cathrynlavery/diagram-design) - Tags: deep-dive - Published: 2026-09-08 --- **The four dials that shape draw.io imports are Format, Size, Detail, and Audience—configurable parameters that determine output media type, dimensions, visual fidelity, and target styling when transforming `.drawio` files into editorial-quality diagrams.** When the **Diagram-Design** skill processes source files from the `cathrynlavery/diagram-design` repository, it does not merely copy the original graphic. Instead, the `import-drawio` slash command applies four transformational controls defined in [`commands/import-drawio.md`](https://github.com/cathrynlavery/diagram-design/blob/main/commands/import-drawio.md) to convert raw diagrams into production-ready assets. ## The Four Dials Explained Each dial corresponds to a specific command-line argument passed to the import utility. These settings dictate how [`drawio_extract.py`](https://github.com/cathrynlavery/diagram-design/blob/main/drawio_extract.py) (located at [`skills/diagram-design/scripts/drawio_extract.py`](https://github.com/cathrynlavery/diagram-design/blob/main/skills/diagram-design/scripts/drawio_extract.py)) parses the compressed `.drawio` payload and generates the structured intermediate representation. ### Format The **Format** dial controls the output media type. Available options include **HTML**, **SVG**, **PNG**, or a **HTML+PNG** combination. This determines whether the final asset renders as a web-ready vector, a raster image for slides, or a dual-format package for documentation systems. ### Size The **Size** dial defines the overall canvas dimensions and aspect ratio. Preset options include `slide-16x9` for presentation decks, `doc-inline` for embedded documentation figures, and `doc-full` for full-page technical diagrams. This ensures the output matches the destination container without manual resizing. ### Detail The **Detail** dial sets the level of visual fidelity. The three modes are: - **faithful**: Preserves every element from the source diagram - **balanced**: Applies typical editorial compression for clarity - **simplified**: Generates a high-level overview suitable for executive summaries ### Audience The **Audience** dial tailors styling and complexity for specific viewers. Options include `engineer` (technical notation emphasis), `mixed` (general technical audiences), and `executive` (simplified visual language for decision-makers). ## Configuring the Import Dials You invoke these parameters via the `/diagram-design:import-drawio` slash command or its programmatic equivalent. The dials are passed as arguments to override defaults. A basic import uses default values (HTML output, slide-16x9 size, balanced detail, mixed audience): ```bash /diagram-design:import-drawio platform.drawio ``` To generate a PNG for executive review with simplified detail: ```bash /diagram-design:import-drawio platform.drawio \ --format=png \ --size=slide-16x9 \ --detail=simplified \ --audience=executive ``` For technical documentation requiring faithful element preservation with dual-format output: ```bash /diagram-design:import-drawio platform.drawio \ --format=html+png \ --size=doc-inline \ --detail=faithful \ --audience=engineer ``` ## Technical Implementation According to the `cathrynlavery/diagram-design` source code, the extraction pipeline begins when [`drawio_extract.py`](https://github.com/cathrynlavery/diagram-design/blob/main/drawio_extract.py) decodes the compressed XML payload within the `.drawio` file. This script produces a structured intermediate representation (IR) that captures geometry, styling, and connectivity. The four dials then act as transformation filters applied to this IR before final rendering. The canonical reference for these controls resides in [`commands/import-drawio.md`](https://github.com/cathrynlavery/diagram-design/blob/main/commands/import-drawio.md), while detailed user-facing documentation appears in [`skills/diagram-design/references/import-drawio.md`](https://github.com/cathrynlavery/diagram-design/blob/main/skills/diagram-design/references/import-drawio.md). ## Summary - **Format** selects the output media type: HTML, SVG, PNG, or HTML+PNG - **Size** sets canvas dimensions using presets like `slide-16x9` or `doc-inline` - **Detail** controls fidelity through `faithful`, `balanced`, or `simplified` modes - **Audience** adjusts styling complexity for `engineer`, `mixed`, or `executive` viewers - The [`drawio_extract.py`](https://github.com/cathrynlavery/diagram-design/blob/main/drawio_extract.py) script parses source files before the dials transform the intermediate representation - Configuration occurs via command-line arguments to the `import-drawio` command ## Frequently Asked Questions ### What are the four dials that shape draw.io imports? The four dials are **Format** (output media type), **Size** (canvas dimensions), **Detail** (visual fidelity level), and **Audience** (target styling profile). These parameters control how the Diagram-Design skill transforms source `.drawio` files into final editorial assets. ### How do I export a draw.io diagram as both HTML and PNG? Use the **Format** dial with the `html+png` value. For example: `/diagram-design:import-drawio diagram.drawio --format=html+png`. This generates both a web-ready vector version and a raster image simultaneously. ### What is the difference between the faithful and simplified detail settings? The `faithful` setting preserves every element from the original diagram, maintaining maximum information density. The `simplified` setting filters out minor elements to create a high-level overview suitable for executive presentations or quick reference materials. ### Where does the extraction logic for draw.io files reside? The extraction logic is implemented in [`skills/diagram-design/scripts/drawio_extract.py`](https://github.com/cathrynlavery/diagram-design/blob/main/skills/diagram-design/scripts/drawio_extract.py) within the cathrynlavery/diagram-design repository. This script decompresses the `.drawio` payload and creates a structured intermediate representation that the four dials subsequently transform.