ip2a

ip2a / memorph

Public

在不同的编程助手之间无损切换会话

claude-code codex cursor deepseek-tui gemini
28
0
85% credibility
Found May 22, 2026 at 28 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Rust
AI Summary

Memorph is a session memory management tool that helps developers transfer their conversation history between different AI coding assistants (like Claude Code, Codex, OpenCode, DeepSeek, Kimi, Kiro, and Cursor). Instead of losing valuable context when switching tools, users can export sessions to files, import them into other tools, or switch directly with a single command. It provides both a web interface and a desktop app, plus command-line tools for power users. The project is open source, published on major package managers (npm, PyPI), and includes proper security measures like checksum verification during installation.

How It Works

1
💻 You're working on a coding project

You've been using an AI coding assistant like Claude Code to help build your project, and you've accumulated valuable conversation history.

2
🔄 You want to try a different AI tool

Maybe you're curious about Codex or OpenCode, but you don't want to lose all your conversation history and context from your current tool.

3
🎯 You use memorph to transfer your session

With one simple command, memorph reads your conversation history and creates a new session in your chosen AI tool, preserving all your context.

4
📋 You can also export and backup

Before switching, you might want to save your session as a file - memorph can export to readable formats like markdown or HTML for documentation.

5
Choose your interface
🌍
Web browser

Run 'memorph web' to open a clean web interface in your browser

🖥️
Desktop app

Install the desktop version for a native window experience

6
🔍 Find and manage all your sessions

Memorph scans all your AI tools and shows you every session across every tool, letting you search, rename, or delete them.

✨ Your context transfers seamlessly

The AI tool you switched to now has all your conversation history, so you can continue exactly where you left off without explaining anything.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 28 to 28 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is memorph?

Memorph is a session management tool built in Rust that lets you move AI coding conversations between Claude Code, Codex, and OpenCode without losing context. You get a CLI with commands like `list`, `switch`, `export`, and `import`, plus a web UI and desktop app. The tool handles session migration across providers, lets you export conversations as JSON or Markdown, and even includes workspace-aware filtering and shared session groups.

Why is it gaining traction?

The AI coding assistant landscape is fragmented. Developers often start projects in Claude, hit context limits, and want to continue in a different tool. Memorph solves that friction by treating session data as portable. The multi-platform distribution (npm, PyPI, crates.io, direct binary) and cross-platform Tauri desktop app mean you can run it anywhere. The comprehensive CLI paired with a clean web interface covers both power users and those who prefer GUIs.

Who should use this?

Developers working across multiple AI coding tools will get the most value. If you bounce between Claude Code for reasoning-heavy tasks and Codex for speed, or experiment with OpenCode for specific use cases, memorph prevents context loss when switching. Teams sharing session artifacts via the export/import flow will also benefit. Early-stage developers exploring which AI assistant fits their workflow will find it useful for comparison without committing exclusively to one tool.

Verdict

Memorph addresses a real pain point in the rapidly evolving AI coding space, with solid Rust performance and thoughtful cross-platform packaging. However, with 28 stars and a credibility score of 0.85%, the project is early and unproven at scale. The documentation and test coverage exist but lack the depth of mature tools. Worth trying if you actively use multiple AI coding agents, but evaluate carefully before relying on it for critical workflows.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.