Export ChatGPT to Obsidian – Save AI Conversations to Your Vault (2026)
Key Takeaways
- One-click export: Use ChatGPT Exporter to save any ChatGPT conversation as a clean
.mdfile — ready to drop into your Obsidian vault. - Formatting preserved: Code blocks, LaTeX formulas, tables, lists, and links all survive the export intact.
- Deep Research & citations: Inline citations from ChatGPT’s Deep Research are converted to proper Markdown footnotes — perfect for Obsidian’s footnote rendering.
- Selective export: Choose to export all messages, prompts only, responses only, or hand-pick specific turns.
- Copy to clipboard: Skip the file download entirely — copy Markdown straight into Obsidian via clipboard.
1. Introduction
Obsidian is the go-to tool for building a personal knowledge base. Its local-first, Markdown-native approach makes it ideal for storing research, notes, and reference material — including conversations with AI.
The problem? ChatGPT conversations are trapped inside OpenAI’s web interface. You can’t search them in Obsidian, link them to other notes, or include them in your knowledge graph. And OpenAI’s official data export gives you a raw JSON blob that requires technical processing before it’s usable.
This guide covers every practical method to export ChatGPT conversations to Obsidian in 2026 — from one-click browser extensions to Python scripts. We’ll compare each approach so you can pick what works for your workflow. (Need a different format? See our guides for Markdown, JSON, or PDF export.)
2. Why Save ChatGPT Conversations in Obsidian?
ChatGPT conversations contain real knowledge — coding solutions, research findings, brainstorming results, writing drafts. Leaving them in the ChatGPT interface means:
- No backlinks or graph view — You can’t connect AI conversations to your existing notes, MOCs, or project pages.
- No full-text search — Obsidian’s search is fast and works offline. ChatGPT’s conversation history search is limited and requires an internet connection.
- No version control — Once saved as
.mdfiles, conversations can be tracked with Git, synced across devices, or backed up locally. - No ownership — If you delete your OpenAI account or ChatGPT changes its data retention policy, your conversations are gone.
- No tags or metadata — In Obsidian, you can add YAML frontmatter, tags, and aliases to organize your AI conversations alongside everything else in your vault.
By exporting ChatGPT conversations to Obsidian, you turn ephemeral chat sessions into permanent, searchable, linkable knowledge assets.
3. Method 1: ChatGPT Exporter (Recommended)
ChatGPT Exporter is a Chrome extension that exports ChatGPT conversations directly as clean Markdown files — no scripting, no data requests, no format conversion.
How to Export
- Install ChatGPT Exporter from the Chrome Web Store (works on Chrome, Edge, Brave, and other Chromium browsers).
- Open any ChatGPT conversation.
- Click the Select button to choose which messages to export — all messages, prompts only, responses only, or a custom selection.
- Click Export and select Markdown as the format.
- Your
.mdfile downloads instantly. Move it to your Obsidian vault folder.
That’s it. The file is ready to open in Obsidian with no further processing.
Copy-to-Clipboard Workflow
If you prefer not to manage files, enable the Copy to clipboard option in ChatGPT Exporter’s settings. Instead of downloading a .md file, the Markdown content is copied to your clipboard. Then:
- Open Obsidian and create a new note (or open an existing one).
- Paste (
Cmd+V/Ctrl+V). - Done — fully formatted Markdown appears in your note.
This is the fastest workflow for saving individual conversations or specific responses into your vault.
What Gets Preserved
ChatGPT Exporter doesn’t just dump raw text. It converts the ChatGPT DOM into properly formatted Markdown using a custom conversion engine with 10+ specialized rules:
- Code blocks — Language tags preserved (e.g.,
```python,```javascript), so Obsidian’s syntax highlighting works immediately. - Tables — Converted to GitHub Flavored Markdown (GFM) table syntax, which Obsidian renders natively.
- LaTeX formulas — KaTeX notation from ChatGPT is converted to standard LaTeX (
$...$for inline,$$...$$for block). Obsidian renders these with its built-in MathJax support. - Lists and nested lists — Ordered and unordered lists maintain their hierarchy and indentation.
- Links — Both inline links and reference-style links are preserved.
- Deep Research citations — ChatGPT’s Deep Research reports include inline citations. ChatGPT Exporter converts these into proper Markdown footnotes (
[^1],[^2]) with a References section — Obsidian renders these as clickable footnotes. - Thought Process — ChatGPT’s chain-of-thought reasoning is captured as Markdown blockquotes, so you can review the AI’s reasoning process.
- Web Search sources — When ChatGPT searches the web, the source URLs are preserved as a formatted list.
- Canvas code — Code artifacts from ChatGPT Canvas are exported as properly tagged code blocks.
- Metadata header — Conversation title, user info, timestamps, and a link back to the original ChatGPT conversation are included at the top of the file. Each field is individually toggleable in settings.
Configurable Options
ChatGPT Exporter lets you customize the output to match your Obsidian workflow:
- File naming — Use conversation title, date-based naming, or custom patterns.
- Metadata fields — Toggle title, user name, email, creation date, update date, export date, and conversation link on/off.
- Time format — Choose between multiple date/time formats.
- Content selection — Export all messages, questions only, answers only, or hand-pick specific Q&A pairs.
4. Method 2: OpenAI Official Data Export
OpenAI lets you request a full export of your ChatGPT data through the Settings menu.
How It Works
- Go to Settings → Data controls → Export data in ChatGPT.
- Click Export and confirm. OpenAI sends a download link to your email (usually arrives within a few minutes to an hour).
- Download and unzip the archive. Your conversations are in
conversations.json.
The Problem
The exported conversations.json file is a single, large JSON file containing all your conversations in a nested format. It’s not Markdown. You can’t drop it into Obsidian and start reading.
To use it, you need to:
- Parse the JSON structure (which nests messages inside
mappingobjects with UUIDs). - Convert each conversation to a separate
.mdfile. - Handle formatting — code blocks, lists, and tables are stored as raw text with no Markdown syntax.
- Strip system messages and metadata you don’t need.
This is where scripts come in (see Method 3), but it means extra steps, technical knowledge, and maintenance when OpenAI changes the export format.
When this method makes sense: You want a complete backup of your entire ChatGPT history and you’re comfortable with scripting.
5. Method 3: Python/Node.js Scripts
Several open-source scripts convert OpenAI’s JSON export into Markdown files. The general workflow:
- Request and download your data from OpenAI (see Method 2).
- Run a conversion script that reads
conversations.jsonand outputs individual.mdfiles. - Move the generated files to your Obsidian vault.
Limitations
- Technical setup required — You need Python or Node.js installed and some comfort with the command line.
- Format fragility — OpenAI periodically changes the structure of
conversations.json. Scripts break and need updates. Multiple community threads report scripts failing after OpenAI format changes. - Limited formatting — Most scripts produce basic text output. Code blocks may lose language tags, tables may not render correctly, and LaTeX formulas are often left as raw KaTeX notation.
- No selective export — You get everything or nothing. There’s no way to export a single conversation without first exporting your entire history.
- No images — DALL-E generated images and uploaded files are typically not included.
When this method makes sense: You’re a developer who wants full control over the conversion process and is willing to maintain the script.
6. Method 4: Other Tools and Extensions
Several other tools target the ChatGPT-to-Obsidian workflow:
| Tool | Type | Pros | Cons |
|---|---|---|---|
| ChatGPT2Obsidian | Chrome extension | One-click export, batch export | Reported broken by users, limited formatting options |
| Nexus AI Chat Importer | Obsidian plugin | Native Obsidian integration, imports from ZIP | Requires OpenAI data export first, reported skipping conversations |
| ChatKeeper | Desktop app | GUI interface, Canvas support, DALL-E images | Separate app to install, requires OpenAI data export |
| Obsidian Web Clipper | Browser extension | Official Obsidian tool | Clips the web page, doesn’t export structured conversation data |
How ChatGPT Exporter Compares
| Feature | ChatGPT Exporter | ChatGPT2Obsidian | Scripts |
|---|---|---|---|
| One-click export | Yes | Yes | No |
| Selective message export | Yes | No | No |
| Code block language tags | Yes | Partial | Partial |
| LaTeX conversion | Yes | No | No |
| Table formatting | Yes | Partial | Partial |
| Deep Research citations | Yes | No | No |
| Thought Process capture | Yes | No | No |
| Web Search sources | Yes | No | No |
| Copy to clipboard | Yes | No | No |
| Custom metadata | Yes | No | Varies |
| No data export required | Yes | Yes | No |
7. Best Practices for ChatGPT Conversations in Obsidian
Once your conversations are in Obsidian, here are some tips to get the most out of them:
Organize with a Dedicated Folder
Create a folder like AI-Conversations/ or ChatGPT/ in your vault. This keeps AI content separate from your own notes while still being searchable and linkable.
Add Tags and Frontmatter
After importing, add YAML frontmatter or tags to categorize conversations:
---
tags: [chatgpt, python, debugging]
date: 2026-04-17
source: chatgpt-exporter
---This makes conversations discoverable through Obsidian’s tag search and Dataview queries.
Link to Related Notes
Use Obsidian’s [[wikilinks]] to connect AI conversations to your project notes, research topics, or MOCs (Maps of Content). This is one of the biggest advantages of having conversations in Obsidian — they become nodes in your knowledge graph.
Use Selective Export for Clean Notes
Instead of exporting entire conversations (which can include false starts and tangents), use ChatGPT Exporter’s selective export to capture only the most valuable exchanges. Export just the responses, or hand-pick specific Q&A pairs that contain the information you want to keep.
Naming Conventions
Use descriptive file names that make conversations easy to find later. ChatGPT Exporter lets you customize file naming patterns — consider including the date and a topic keyword:
2026-04-17-python-async-patterns.md
2026-04-17-react-state-management.md8. Frequently Asked Questions
Can I export ChatGPT conversations directly to Obsidian?
Yes. Use ChatGPT Exporter to export conversations as Markdown files, then move them to your Obsidian vault folder. Alternatively, use the copy-to-clipboard feature to paste directly into an Obsidian note.
Does ChatGPT Exporter work with Obsidian’s Markdown format?
Yes. ChatGPT Exporter outputs standard GitHub Flavored Markdown (GFM), which is the same format Obsidian uses. Code blocks, tables, lists, LaTeX formulas, and links all render correctly in Obsidian.
Can I export only specific messages from a ChatGPT conversation?
Yes. ChatGPT Exporter lets you select which messages to export — all messages, prompts only, responses only, or individual Q&A pairs. This is useful when you only need specific answers from a long conversation.
Does the export preserve code blocks with syntax highlighting?
Yes. ChatGPT Exporter preserves language tags on code blocks (e.g., ```python), so Obsidian’s syntax highlighting works automatically. Most other tools lose this information during export.
What about LaTeX formulas and math equations?
ChatGPT Exporter converts ChatGPT’s KaTeX notation to standard LaTeX ($...$ for inline, $$...$$ for block). Obsidian renders these natively if you have MathJax enabled in your vault settings (it’s on by default).
Can I bulk export all my ChatGPT conversations to Obsidian?
Yes. ChatGPT Exporter supports bulk export — you can select and export multiple conversations at once as individual Markdown files, then move the entire batch to your vault.
Are Deep Research reports preserved in the export?
Yes. ChatGPT Exporter converts Deep Research inline citations into proper Markdown footnotes with a References section. Obsidian renders these as clickable footnotes, preserving the full research context.
Does this work with Obsidian Sync?
Yes. Since ChatGPT Exporter creates standard .md files, they work with Obsidian Sync, iCloud, Dropbox, Git-based sync, or any other file syncing method you use.
Is ChatGPT Exporter free?
ChatGPT Exporter offers Markdown export as a free feature. You can export unlimited conversations to Markdown at no cost. Install it here.
What’s the difference between ChatGPT Exporter and ChatGPT2Obsidian?
ChatGPT Exporter supports selective message export, Deep Research citation conversion, Thought Process capture, LaTeX formula conversion, and customizable metadata — features that ChatGPT2Obsidian does not offer. See the comparison table above for a full breakdown.
9. Conclusion
Your ChatGPT conversations are a knowledge asset. Leaving them locked inside OpenAI’s interface means missing out on search, backlinks, and the full power of your Obsidian knowledge graph.
The fastest way to get ChatGPT conversations into Obsidian is with ChatGPT Exporter — one-click Markdown export with code blocks, LaTeX, tables, and citations preserved. No scripting, no data export requests, no format conversion.
Install ChatGPT Exporter and start building your AI-powered second brain in Obsidian today.
Related guides: ChatGPT to Markdown · ChatGPT to JSON · ChatGPT to PDF · Bulk Export