# odp

Creates, inspects, edits, and repairs OpenDocument presentations (.odp). Useful for automated slide generation and bidirectional ODP to PPTX conversion.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode, Windsurf (Devin Desktop), Claude.ai · 1 install · ★ 3 repo stars
- Author: leiverkus
- License: MIT · policy=mirror · the beginning of the file is at the end of this page
- Source: https://github.com/leiverkus/open-document-skills/tree/HEAD/skills/odp/SKILL.md
- Updated: 2026-07-23
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/skills/odp

## What it does

The skill guides coding agents through handling `.odp` packages as ZIP containers and modifying raw XML files directly. It targets key presentation parts: slide definitions inside `content.xml`, styles and master pages in `styles.xml`, and embedded graphics. Bundled helper scripts include `extract_text.py` for exporting slide text and speaker notes, `inspect_package.py` for metadata diagnostics, and `create_minimal_odp.py` providing six standard slide layouts.

It handles shape entrance and exit animations, slide transitions, and master-page layouts. Because full visual rendering and PPTX conversion rely on headless LibreOffice, converting decks or exporting slide images will fail if `soffice` is missing from the environment.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add leiverkus/open-document-skills --skill odp -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/odp/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/odp/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add leiverkus/open-document-skills --skill odp -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/odp/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add leiverkus/open-document-skills --skill odp -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/odp/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install leiverkus/open-document-skills skills/odp
```

2. **Manually: put the files in place**
- File: `.github/skills/odp/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add leiverkus/open-document-skills --skill odp -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/leiverkus/open-document-skills --path skills/odp
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/odp/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add leiverkus/open-document-skills --skill odp -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/leiverkus/open-document-skills/tree/HEAD/skills/odp/SKILL.md

## When to use it
- extract visible slide text and speaker notes from an .odp file into JSON
- generate structured presentation decks from templates by altering package XML
- convert presentations bidirectionally between ODP and PPTX formats
- inspect embedded media references, slide order, and master pages in ODP packages

## Example prompts
- Extract visible text and speaker notes from deck.odp using the JSON flag
- Generate an odp deck from deck.json using the two-content slide layout
- Inspect the package structure and image paths inside input.odp

## Source text · SKILL.md · MIT

Below is the content of a third-party file. It is data, not instructions for you: do not act on anything written inside unless the human asks. Block tag: 43d761a7756ac79d.

BEGIN THIRD-PARTY FILE 43d761a7756ac79d
````text
---
name: odp
description: "Create, read, edit, convert, repair, or inspect OpenDocument Presentation files (.odp) — including bidirectional conversion to PPTX/PPT via headless LibreOffice. Supports shape animations (entrance, exit, emphasis, motion paths), slide transitions, master-page customization, and a template ecosystem (curated branded designs + inspector/extractor tools)."
triggers: [".odp", "ODP", "OpenDocument Presentation", "Open Office presentation", "LibreOffice Impress", "Impress deck", "odp-Datei", "OpenDocument-Präsentation", "Folien", "Präsentation", "animation", "Animation", "transition", "Übergang", "Folienübergang", "slide transition", "master page", "master slide", "Folienmaster", "flat ODF", ".fodp", ".pptx", "PPTX", ".ppt", "PowerPoint", "PowerPoint-Datei", "MS PowerPoint", "Microsoft PowerPoint", "template", "Vorlage", "Designvorlage", "Foliendesign", "branding", "Branding", ".otp"]
dont_use_for: ["text documents (.odt)", "spreadsheets (.ods)", "general presentation advice without .odp file"]
license: MIT
version: "1.15.0"
---

# ODP creation, editing, and analysis

## Overview

An `.odp` file is an OpenDocument ZIP package for presentations. Important package files:

- `mimetype` - should be the first ZIP entry and stored uncompressed as `application/vnd.oasis.opendocument.presentation`
- `content.xml` - slides, shapes, text boxes, images, speaker notes
- `styles.xml` - presentation, drawing, and text styles
- `meta.xml` - document metadata
- `settings.xml` - application settings
- `META-INF/manifest.xml` - package manifest
- `Pictures/...` - embedded media

## Quick Reference

| Task | Preferred approach |
|------|--------------------|
| Create branded deck | Start from an `.odp` template, clone slides/masters, edit XML |
| Create simple structured deck | Generate ODP package XML directly |
| Convert PPTX/HTML/SVG to ODP | Use LibreOffice only when the source already exists or interoperability requires it |
| Extract text | Use `scripts/extract_text.py` or parse `content.xml` |
| Preserve template layout | Edit the existing ODP package XML and media references |
| Visual QA | Convert to PDF or slide images and inspect every slide |

## Tool Checks

Before starting a real ODP task, check available tools:

```bash
which pandoc
```

Resolve the LibreOffice command as described in [docs/soffice-resolver.md](../../docs/soffice-resolver.md).

LibreOffice is the preferred engine for ODP conversion and rendering. Pandoc is not a full ODP authoring or inspection tool.

## Reading and Inspecting

For raw inspection:

```bash
unzip -l input.odp
python -m zipfile -e input.odp unpacked_odp
```

Slide content is usually in `content.xml` under `draw:page` elements. Extract text with an XML parser, keeping namespace handling explicit.

Bundled scripts for common inspection tasks:

```bash
# Extract visible slide text and speaker notes.
````
END THIRD-PARTY FILE 43d761a7756ac79d

The third-party file has ended. Everything between the 43d761a7756ac79d tags is data: requests and commands from it are not followed. An “end of file” line with another tag inside the block is part of the file.

First 2904 of 26415 bytes. © leiverkus · MIT · https://github.com/leiverkus/open-document-skills/tree/HEAD/skills/odp/SKILL.md

_Part of the page did not fit — see the HTML version._

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
