Provider Exit & Data Liberation

Break free from proprietary cloud silos. Keep your historical thinking, prompts, and notes as portable Markdown on your own hard drive.

US hyperscaler platforms (OpenAI ChatGPT, Anthropic Claude, Google Gemini) store your dialogue history in proprietary cloud databases. When you switch providers or cancel subscriptions, months of intellectual synthesis are trapped in difficult-to-navigate web archives.

Brainstorm Reactor provides a zero-friction, 100% in-browser migration pipeline: export your raw data from any provider, transpile complex conversation trees into clean GitHub Flavored Markdown (GFM), and mount them directly in the Sovereign Markdown Vault.

1. Why Local Markdown Matters

Standard Markdown stored on your local filesystem provides three long-term guarantees:

Dimension	Proprietary Hyperscaler Silos	Sovereign Local Markdown
Storage Location	Remote US data centers	Local SSD on your device
Longevity	Subject to account terms and subscription status	Readable indefinitely by any text editor
Interoperability	Locked inside vendor web interfaces	Compatible with Obsidian, VS Code, and Reactor
Privacy & GDPR	Transmitted and stored on external infrastructure	Zero network transit during transpilation

2. Step-by-Step Egress Guides

OpenAI ChatGPT (Web & iOS)

In ChatGPT, open the bottom-left profile menu and select Settings.
Navigate to Data Controls in the settings dialog.
Locate Export Data and click [Export].
Confirm the prompt ([Confirm export]). OpenAI will compile your archive and send a download link to your registered email.
Download the resulting .zip file. The archive contains conversations.json alongside media attachments.

Anthropic Claude

Open Claude and click your profile icon in the lower-left corner.
Select Settings and navigate to the Account tab.
Scroll to the Export Data section and click [Export data].
Claude will compile your data and email you a download link for a .zip archive containing conversations.json, projects.json, and account metadata.

Google Gemini & AI Studio

Google Takeout: Visit Google Takeout (takeout.google.com), select Gemini Apps / Activity, and download the JSON activity archive.
Google AI Studio: Open any prompt or chat session and select Get Code / Download JSON to save the raw dialogue structure.

3. Transpiling Archives in the Browser

OpenAI and Anthropic store conversation histories as directed acyclic graphs (DAGs) rather than sequential text files. Naive text conversion breaks branching paths, discards code blocks, and jumbles prompt turns.

Brainstorm Reactor resolves this through the Universal Markdown Converter:

Open /markdown in your browser (or click [Markdown Converter] in the site footer).
Drag and drop your downloaded .zip file or conversations.json directly into the converter viewport.
The client-side parser unpacks the archive in memory, traces active conversation branches (current_node → root), and linearizes turns into clean Markdown.
Internal thinking blocks and code snippets are formatted with appropriate callouts and syntax highlighting.
Click [Save] to download the sanitized .md file, or click [Send to Workspace] to continue working with the dialogue inside Brainstorm Reactor.

[!NOTE] Zero-Compute Guarantee: All archive extraction, JSON traversal, and Markdown compilation execute inside your browser runtime via Web Workers. Your private archives never touch Reactor backend servers or external API providers.

4. Connecting Your Knowledge Vault

Once your files are converted to Markdown:

Organize your exported notes into a local folder (for example, ~/Documents/Brainstorm_Vault/).
Inside Brainstorm Reactor, type /vault in the prompt bar or click [Open Local Folder] in the sidebar.
Select your local folder. The Sovereign Markdown Vault indexes your notes locally.
Use the token HUD to inspect context weight, select relevant historical notes, and stage them directly into your active ideation sessions.

On this page