Why I built ChatVault to keep AI coding chats in one place
Hi! So I built ChatVault to solve a real problem: keeping useful AI coding conversations accessible when I switch tools or accounts.
If you use Claude Code, Codex or Cursor, you probably know the feeling. You remember discussing a bug, making a plan or finding a useful solution, but you cannot remember which tool or account holds that conversation. The work is somewhere on your computer, yet finding it interrupts everything.
I wanted one place to read those chats, search them, back them up and carry the context into another coding tool. That is what ChatVault is for.
Explore ChatVault on GitHub · Download for Windows
What is ChatVault?
ChatVault is a free, open source Windows desktop app for managing local AI coding conversations. It brings chat history from supported tools into one interface instead of making you look through separate apps and folders.
It supports Claude Code, Codex, Cursor, Gemini CLI, OpenCode and Cline. The goal is simple: make the conversations already stored on your PC easier to use, including local history associated with different accounts.
ChatVault is independent and is not affiliated with Anthropic, OpenAI, Anysphere or the other tool makers.
Why I built it
A coding conversation can contain much more than a prompt and an answer. There are decisions, failed approaches, error messages, code snippets and the explanation for why a particular fix worked. Losing track of that history means repeating work.
Switching coding tools adds another problem. You may want a different model to help, or you may hit a usage limit, but the next tool does not automatically know the conversation you just had. Rebuilding that context from memory takes time and leaves details out.
ChatVault gives that history a home outside any one tool's current chat list. It cannot recover files that have already been deleted, and it does not bypass usage limits. It helps you work with the local conversations that remain available.
Read your coding chats in one place
The chat reader lets you revisit conversations without hunting through raw files. You can follow the discussion, review code and look back at the steps taken during a task.
For me, the useful part is being able to return to an old explanation while working on something new. A fix from last week can be relevant again today, even if it came from another coding tool.
Search across tools and accounts
When all you remember is a function name, an error message or a topic, searching is much easier than opening chats one at a time. ChatVault lets you search supported local chat history across tools.
For example, you might search for a checkout bug, a deployment command or a specific API name. You can then open the matching conversation and read the surrounding context rather than relying on an isolated snippet.
Search depends on the chat files available on your computer. It is not a cloud archive of every conversation you have ever had.
Keep local backups and useful exports
ChatVault includes local backup and export features so you can preserve conversations and use them outside the app. Export formats include Markdown, HTML, plain text and JSON.
- Markdown: useful for project notes and readable handoff files.
- HTML: useful when you want a conversation you can open in a browser.
- Plain text: useful for a simple, portable copy.
- JSON: useful when you need structured data for your own workflow.
A backup is most valuable before you need it. If chat history matters to your work, keep a separate copy and check that you can open it.
Continue a conversation in another coding tool
This is one of the features I am most excited about. ChatVault can prepare a handoff so you can bring a conversation into another supported coding tool instead of explaining the task from scratch.
Imagine investigating a bug in Claude Code, then wanting Codex to take a look. The previous discussion may contain the error, attempted fixes and important constraints. Carrying that conversation over gives the next tool a better starting point.
A handoff transfers context. It does not transfer a model's internal state, guarantee identical reasoning or make a long conversation fit every context window. You should still check that the receiving tool has the right project and understands the next step.
Privacy: your chats stay on your computer
ChatVault reads local chat files and does not upload your conversations. You do not need to create a ChatVault account to use it. According to the project's documentation, GitHub is used to check for application updates, and the app does not touch your sign-in files.
Exporting a conversation creates a file that you control. Before sharing an export with another person or tool, check it for private information, credentials or project details you do not want to disclose.
How to get started
- Open the latest ChatVault release.
- Choose the Windows installer for a normal installation, the MSI for Windows Installer deployment, or the portable executable if you prefer to run it without installing.
- Launch ChatVault on a Windows 10 or Windows 11 PC running a 64-bit system.
- Browse the local conversations from your supported coding tools and try searching for an old task.
- Make a backup, export a conversation or try a handoff when you want to continue elsewhere.
The project uses the WebView2 runtime and is built with Tauri, Rust, SQLite, React and TypeScript. Developers can inspect the source and follow the build instructions in the repository.
Who is it for?
ChatVault is for developers who use more than one AI coding tool, switch between accounts, revisit old solutions or want a local copy of their coding history. It can also help when you want to turn a useful chat into project documentation.
If you only use one tool, search and backups may still be useful. If you move between several, the shared reader and handoff workflow are where the app becomes especially handy.
Try it and tell me what you think
ChatVault is free and released under the MIT license. You can download it, inspect the code, report a bug or contribute an improvement.
I would love to hear what is missing from your current workflow. Which coding tools do you switch between? What makes an old conversation difficult to find? And what would make a handoff more useful?
Visit ChatVault on GitHub. If it solves a problem for you, a star would mean a lot 🙂
The screenshots show sample conversations and demonstration data.
Comments
Post a Comment
Join the conversation using Blogger’s comment form.
Write a comment ↗