Nope

NOPE

NOPE ( Not anOther Pdf Exporter ) turns Obsidian notes into publication-ready PDFs through a Docker-based Pandoc/LaTeX pipeline. You write in Obsidian, keep your notes modular and export a polished PDF without installing a local TeX distribution.

What NOPE does

NOPE keeps the Obsidian authoring model intact and adds the parts that are usually painful to wire up by hand: PDF export, numbering, cross-references, glossaries, branding, citation handling and LaTeX-backed structured blocks.

The core authoring model is atomic. Chapters, theorems, tables, glossary entries, diagrams and other reusable content blocks live in their own notes and are embedded into a main document.

Under Construction: NOPE is still in early development. The Plugin is not stable yet and there may be breaking changes. Feedback and contributions are very welcome to help shape the direction of the plugin.

https://github.com/user-attachments/assets/93402f1f-b05f-48d8-879e-37bfac73c5cf

---

Install

Platform support: Only Linux and macOS are tested. Windows is experimental and currently untested — it may not work yet. Feedback and reports very welcome.

At a high level, the setup is simple:

1. Install Docker.
2. Install the NOPE plugin.
3. Export the active note to PDF.

The first export prepares the full toolchain, including the download and build of the Docker image. Alternatively you can build the image in the Settings or via the command palette. PDF export is a regular command from the Obsidian command palette.

Install the plugin

Community store (recommended):

1. In Obsidian: Settings → Community plugins → Browse.
2. Search for "Nope" and install.
3. Enable NOPE.

via BRAT:

1. Install BRAT.
2. Add beta plugin: MrIwan/obsidian-nope.
3. Enable NOPE.

Requires Docker. First export builds the image.

---

Recommended workflow

NOPE supports treating your vault like a set of reusable building blocks instead of one long monolithic document.

A typical workflow looks like this:

1. Create one short main document.
2. Put each chapter in its own note.
3. Put each reusable structured element in its own note, for example a theorem, table, equation, glossary entry, or Mermaid diagram. Some of these atomic notes are driven by a few frontmatter keys (latex-env, caption, …) — see Structured blocks with latex-env.
4. Embed these notes into the main document with normal Obsidian wikilink embeds.
5. Refer to embedded content with normal wikilinks.
6. Export the main document to PDF.

This keeps the source material maintainable, reusable and easy to rearrange.

---

Main document structure

A main document is usually short. It mostly contains frontmatter and embeds.

The main file acts as composition, while the actual content lives in reusable notes.

Example

---
title: "NOPE Example Document"
subtitle: "An example of the main document"
author: "Wanja Zemke"
date: 2026-06-01
toc: true
---

![[Chapter-Introduction]]
![[Chapter-Theory]]
![[Chapter-Results]]

---

Feature overview

Atomic notes

Atomic notes are the foundation of Obsidian. Instead of writing one huge export note, you split content into small units that each represent one concept.

Good candidates for atomic notes include:

• Chapters
• Theorems and definitions
• Tables
• Equations
• Glossary entries
• Mermaid diagrams
• Reusable image figures

This structure makes it easier to reuse content across multiple documents and reduces maintenance overhead when parts of a document change.

Embedded notes

NOPE uses normal Obsidian embed syntax to assemble documents.

Supported patterns:

• ![[Note]] embeds a full note.
• ![[Note#Heading]] embeds a section starting at a heading.
• ![[Note#^block-id]] embeds a block slice.
• ![[Base.base#View]] embeds an Obsidian base view — see Bases.

Auto heading shift

Embedded notes should start with a normal # Heading. NOPE automatically shifts heading depth based on where the note is embedded.

That means embedded content can keep a clean, local structure without forcing you to manually rewrite heading levels for every context.

latex-env notes (theorems, tables, math, Mermaid) should not contain headings, since their body is wrapped in a LaTeX environment and the title/caption come from frontmatter.

Cross-references from normal wikilinks

Normal wikilinks are turned into PDF references when the target is embedded in the exported document.

Typical behavior:

• [[Note]] becomes an automatic reference such as a theorem, table, figure, or equation reference.
• [[Note|Custom Text]] becomes a hyperlink with custom link text.
• If a target is not embedded, the link falls back gracefully to plain text.

This lets you keep using familiar Obsidian links while getting proper PDF cross-references.

Passive embeds

Passive embeds use the +[[...]] syntax. They behave like normal embeds during export, but remain visually compact in the Obsidian editor.

This is useful when a document contains many large chapter embeds and you want the source note to stay readable while still expanding everything in the final PDF.

---

Structured blocks with latex-env

Some notes are not just plain text blocks. NOPE can map a note to a LaTeX environment through frontmatter.

This allows structured content to stay author-friendly in Obsidian while being rendered correctly in the exported PDF.

Theorems, lemmas, definitions, proofs

Use latex-env to wrap a note in a theorem-like environment. The numbered family like theorem, lemma, definition, corollary, proposition and example (plus any custom amsthm environment) has its own counter, so a wikilink resolves to the typed name and number ("Theorem 1.1", "Lemma 1.2", "Definition 2.1"). proof, remark and note are unnumbered. A wikilink to them is a clickable hyperlink showing the note title instead of a number.

This is ideal for mathematical or technical writing where reusable theorem-style blocks should be referenced from the surrounding text.

A latex-env note contains no headings — its body is wrapped directly in the LaTeX environment and an optional bracket title comes from latex-short in the frontmatter.

Example

Theorem-Pythagoras.md:

---
latex-env: theorem
latex-short: Pythagoras
---

$a^2 + b^2 = c^2$

Embedded elsewhere:

![[Theorem-Pythagoras]]

From [[Theorem-Pythagoras]] we have ...

Tables

A note with latex-env: table becomes a numbered table in the PDF. The note must include a caption in frontmatter.

This makes tables first-class document objects with captions, numbering, references and list-of-tables support.

Layout: longtable (long vs. wide)

The optional longtable key (default false) lets you choose how a table is laid out — you pick what fits your data:

• longtable: false (default) — the table stays on one page and, if it is wider than the text, the whole table is scaled down until it fits (no column overflow). Best for wide, compact tables. It does not break across pages, so a table taller than one page would run off the bottom — use true for those.
• longtable: true — the table is set as a longtable and breaks across pages when needed, repeating the header row. Best for long tables. Downside: very wide columns with long words can overflow sideways into the next column.

Rule of thumb: wide → keep the default (false), long → true. A table cannot be both wide-and-scaled and multi-page at the same time. In false mode, wikilinks inside cells are rendered as-is rather than resolved to references.

Alignment: align

The optional align key (default center, or left) sets the table's horizontal placement. It works in both layouts.

---
latex-env: table
caption: "Long results table"
longtable: true
---

Example

Table-Results.md:

---
latex-env: table
caption: "Measurement results"
---

| x | y |
|---|---|
| 1 | 2 |
| 3 | 4 |

Embedded elsewhere:

![[Table-Results]]

As shown in [[Table-Results]], the measurements indicate ...

Equations and math environments

NOPE supports math-focused environments such as equation, align, gather, multline and alignat, including star variants.

This is useful when equations should be atomic, reusable and cross-referenceable instead of being buried inline inside long chapter notes.

Example

Equation-Energy.md:

---
latex-env: align
---

$$
E &= mc^2 \\
F &= ma
$$

Embedded elsewhere:

![[Equation-Energy]]

From [[Equation-Energy]] we see that ...

Mermaid diagrams

Mermaid diagrams can be authored as atomic notes with latex-env: mermaid. The note body contains the Mermaid code block, while frontmatter supplies metadata such as the caption.

On export, the diagram is rendered into an image, numbered like a figure and can be referenced from the surrounding text.

Optional frontmatter keys also let you control width and render resolution.

Example

Diagram-DataFlow.md:

---
latex-env: mermaid
caption: "Data flow from vault to PDF"
---
```mermaid
graph TD
  A[Vault] --> B[Pandoc]
  B --> C[PDF]
```

Embedded elsewhere:

![[Diagram-DataFlow]]

In [[Diagram-DataFlow]] we see the data flow from vault to PDF.

---

Images and figures

Image embeds become figures when used with a caption.

Supported width hints are percentages, pixel units and metric units (cm, mm). This makes it possible to reuse the same image at different sizes across different documents.

A reference with [[plot.png]] becomes a figure reference such as "Figure N" (localised to the document language, e.g. "Abbildung N" with lang: de).

Example

![[plot.png|My plot]]
![[plot.png|My plot|w=60%]]

---

Bases

NOPE can embed an Obsidian base view into a document with ![[Base.base#View]]. What it produces depends on the latex-env of the surrounding note:

• No latex-env (default) — every note the view returns is transcluded, in the view's order. Embed the base straight into a document to use it as a spine or table of contents.
• latex-env: table + caption — the view is rendered as a numbered table with exactly the columns, order and sort configured in that view. Table needs a caption, this lives in a small wrapper note whose body is just the base embed. Embed that wrapper with ![[…]]. longtable and align work like any other table.

The query (filters, formulas, sort, column selection) is evaluated by Obsidian's real Bases engine at export time — what the view shows is what you get. Add #View to pick a view. Without it the first view of the base is used. This requires the core Bases plugin to be enabled (Obsidian 1.10+).

Note — a base view briefly opens during export. To read a base with Obsidian's real engine, NOPE opens the referenced base view in a tab for a moment while it renders, then closes it again and returns focus to the preview or your current note. You may see it flash open and closed once per base during export. A base only produces data once Obsidian actually renders it, so it cannot be resolved fully invisibly.

Note — a _nope-temp-base.base file in your vault root. Obsidian can only query a base by mounting a real file, so NOPE keeps a single scratch base named _nope-temp-base.base at the vault root and overwrites it on every export. It is safe to ignore or delete — if you remove it, it is recreated on the next export. NOPE keeps one reused file rather than creating a new one each run so your vault doesn't fill up with temp bases.

Example

Transclude every note in a base (document spine), straight in the document:

## Chapters

![[Thesis-Chapters.base#Reading order]]

Render a base as a table — a wrapper note Project-Overview.md:

---
latex-env: table
caption: "Project overview"
---
![[Projects.base#Table]]

Embedded elsewhere:

![[Project-Overview]]

As shown in [[Project-Overview]], the project is on track.

---

Glossary and acronyms

Glossary entries and acronyms are defined as atomic notes with gls-* frontmatter.
When you link such a note with a normal wikilink like [[GLS]] or [[ACN]], NOPE automatically turns it into the correct glossary or acronym reference in the exported PDF.

Example

GLS.md:

---
gls-id: gls
gls-short: GLS
gls-long: Glossary
gls-description: A glossary entry is a note that defines a term. It can be referenced from the text and will appear in the generated glossary section of the PDF with the hyperlinked description.
gls-type: term
---

ACN.md:

---
gls-id: acn
gls-short: ACN
gls-long: Acronym
gls-description: ""
gls-type: acronym
---

---

Citations and bibliography

NOPE supports bibliography-driven citations through Pandoc. A document can point to a .bib file in the vault and optionally use a CSL style.

This makes the plugin suitable for academic or technical writing workflows that already rely on tools such as Zotero and Better BibTeX.

Example

Document frontmatter:

---
bibliography: refs.bib
csl: ieee.csl
---

In text just refer to the citation key:

As shown by [@smith2020], the result holds.

---

Obsidian-specific inline syntax

NOPE also preserves and transforms selected inline authoring patterns.

Supported examples include:

• %%comment%% for comments that do not appear in the PDF
• ==highlight== for highlighted text
• Obsidian callouts mapped to PDF callout styling

This helps preserve the convenience of writing in Obsidian without losing control over the final document output.

Example

This sentence has %%a hidden note%% that does not appear in the PDF.
This sentence has ==highlighted text==.

> [!note] Important
> This callout becomes a PDF callout.

---

Branding

Branding is handled through a dedicated branding note and regular frontmatter overrides. This keeps document styling configurable without hardcoding project-specific values into the content itself.

A branding note can control items such as:

• Title page behavior
• Logos
• Header and footer content
• Colors and template metadata
• Other Pandoc and Eisvogel options

The branding note is activated from the document frontmatter via nope-branding.

Logos in headers and footers

Logo paths can be managed through branding metadata. Header and footer logo slots are designed for easy per-project customization.

This is useful for customer-specific reports, institutional templates, or multi-brand documentation workflows.

Example

In a branding note:

---
header-logo: "[[logo-header.png]]"
footer-logo: "[[logo-2-footer.png]]"
---

Branding overrides per document

Defaults can come from the plugin configuration, while a branding note and the document frontmatter can override specific values.

This layered approach gives you a stable baseline and still allows document-level customization when needed.

Example

Document frontmatter overrides selected values from the branding note:

---
nope-branding: "[[Branding-Default]]"
title: "Q4 Report"
header-color: "#003366"
---

In this example, the document uses the Branding-Default note for general styling but overrides the title and header color for this specific report.

---

Custom LaTeX template

By default NOPE renders with the bundled Eisvogel template. Power users can supply their own .tex Pandoc template instead — selected via frontmatter on the document or the branding note:

---
nope-template: "[[my-template]]"
---

If unset, Eisvogel is always the fallback. When a custom template is used, NOPE shows a notice.

A custom template must pull in the packages and definitions NOPE depends on (tables, callouts, theorems, glossary, cross-reference names). These live in a clearly marked block inside the template:

%%% NOPE-IMPORTS — DO NOT CHANGE … %%%
…
%%% END NOPE-IMPORTS %%%

Keep that block verbatim. Besides the NOPE features, a template also needs a Pandoc baseline (image scaling, code highlighting, table-width macros, float placement) that Pandoc's default template would normally provide. The easiest start is therefore the shipped nope_minimal.tex — a no-overhead starter that already has the Pandoc baseline and the required slots ($body$, header-includes, title, table of contents, glossary). The command Create custom LaTeX template drops it into your vault; then point nope-template: at it and customize freely.

Editing .tex files inside Obsidian: Obsidian hides non-markdown files by default. Enable Settings → Files & Links → Detect all file extensions to see them and install the Custom File Extensions plugin to open and edit .tex files (it handles the .tex extension out of the box). Otherwise, right-click the file → Open in default app to edit it in your usual LaTeX editor.

---

Table of contents and generated lists

NOPE supports document metadata such as a table of contents, list of figures and list of tables.

That makes it practical for longer reports or formal deliverables where navigation and generated lists are expected.

---
toc: true
lof: true
lot: true
---

---

Commands

The plugin exposes several commands through the Obsidian command palette.

Export active note to PDF

This is the main command. It runs the document through the containerized export pipeline and writes the resulting PDF.

Open PDF preview / Sync PDF preview to cursor

Open PDF preview opens a live preview pane that re-renders the PDF as you edit. Sync PDF preview to cursor scrolls the preview to the part of the document your cursor is in.

Build docker image (with cache) / (no cache)

These commands build the Docker image used by the export pipeline. You typically need it for the first run and after environment-related changes. The cached variant reuses existing layers (faster). The no-cache variant rebuilds from scratch. The image also rebuilds itself automatically after a plugin update changes the pipeline.

Create branding template

This command creates a branding template note in the vault. It is a quick way to start a new brand configuration without copying old project files by hand.

Create custom LaTeX template

This command drops the nope_minimal.tex starter into the vault as a base for your own LaTeX template. Point a document at it via nope-template to use it instead of Eisvogel.

Add … (structured note scaffolds)

A set of commands — Add table, Add base table, Add theorem, Add lemma, Add definition, Add proof, Add mermaid diagram, Add equation, Add glossary term and Add abbreviation — each create a new atomic note next to the current one with the correct frontmatter (latex-env, required keys like caption, gls-*) and a ready-to-fill body, then insert the embed (![[…]]) or reference ([[…]]) at your cursor and open the note in a new tab. They are available while editing a note.

Remove docker image

This command removes the pipeline image. The next export can then rebuild the environment from scratch.

Cleanup build folder

This command clears the build folder used during export. It is mainly useful for maintenance and debugging.

---

AI skill integration

NOPE can install a skill into the vault, where the NOPE conventions are defined. The Install AI conventions skill command copies the skill into the vault. The goal is to help AI tools follow the plugin's writing conventions when generating or editing exportable notes.

This is particularly useful when AI is used for report drafting, technical writing, or vault-assisted authoring.

Example

With the skill installed, asking an AI assistant for "a new theorem note about the triangle inequality" yields an atomic note with proper latex-env: theorem frontmatter instead of an inline theorem buried in a chapter.

---

License

Copyright (C) 2026 Wanja Zemke

NOPE is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License v3.0 as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

It is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the full license text in LICENSE for details.

---

A note on AI assistance

Parts of this codebase were written with the help of Claude. Everything was reviewed, tested and understood before being committed.