ClaudePoint
The ClaudePoint MCP server allows you to generate instant snapshots (checkpoints) of your codebase, and lets Claude Code or Claude Desktop go wild with refactoring or adding new features, and if anything goes sideways.
You can use it as a version control system that helps you track changes, document work automatically, and recover from mistakes without interrupting your workflow.
Features
- π Global NPM install β Single setup for all projects
- π€ Native Claude integration β Direct CLI/GUI support
- π Auto-generated changelogs β Tracks every checkpoint/restore
- π AI-powered documentation β Claude logs its own changes
- π‘οΈ Emergency backups β Auto-safety net before restores
- β‘ Gitignore-aware snapshots β Skips junk files like
node_modules - π§Ή Smart cleanup β Keeps only recent checkpoints (configurable)
Use Cases
- Major Refactoring: You need to overhaul a core feature, like the authentication system. Before you start, you tell Claude, “Create a checkpoint before you start.” Then you can let the AI refactor everything. If the new implementation has issues, you can just say, “Restore the last checkpoint,” and you’re back to where you started, no harm done.
- Session Continuity: You finish up for the day in the middle of a complex task. The next morning, instead of trying to piece together where you left off, you can ask Claude, “Show me our development history.” ClaudePoint pulls up the changelog, and you (and Claude) have the full context of the last session’s work.
- AI-Driven Documentation: As you work, you can instruct Claude to log its actions. For instance, “Log that you just added the OAuth integration with a detailed changelog entry.” This builds a rich, real-time log of the project’s evolution without you having to stop and write documentation manually.
- Trying Out Different Ideas: You have a couple of ways you could implement a new feature. You can create a checkpoint, try the first approach, then restore and create another checkpoint to try the second. Itβs much faster than creating and managing multiple Git branches for quick experiments.
How to Use It
1. Install claudepoint.
npm install -g claudepoint2. Configure your Claude:
For Claude Code (CLI):
claude mcp add claudepoint claudepointFor Claude Desktop (GUI):
You’ll need to edit a JSON configuration file.
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
Add the following to the file:
{
"mcpServers": {
"claudepoint": {
"command": "claudepoint",
"args": []
}
}
}3. Once it’s set up, you can talk to Claude in natural language to manage checkpoints.
"Setup checkpoints for this project""Create a checkpoint called 'before-auth-refactor'""Show me the list of checkpoints""Restore the 'before-auth-refactor' checkpoint""Log that you just fixed the token expiration bug"
4. You can also use ClaudePoint directly from your terminal if you want manual control.
claudepoint setup– Initializes ClaudePoint in your project directory.claudepoint create --description "Initial setup"– Creates a new checkpoint.claudepoint list– Shows all checkpoints.claudepoint changelog– Displays the development history.claudepoint restore "checkpoint-name"– Restores a specific checkpoint.
5. When configured, Claude gets access to a set of tools it can use on your behalf, including create_checkpoint, list_checkpoints, restore_checkpoint, and set_changelog.
FAQs
Q: How is ClaudePoint different from Git?
A: They serve different purposes but work well together. ClaudePoint is for rapid, short-term experimentation and recovery within a coding session. Itβs your safety net for trying things out without cluttering your formal version control. Git is for long-term, permanent version history. The ideal workflow is: use ClaudePoint to experiment, and once you have a stable set of changes you’re happy with, commit them to Git.
Q: Do my checkpoints get saved in my Git repository?
A: No. ClaudePoint creates a .checkpoints folder in your project root to store its data, and this folder is automatically added to the ignore list. This separation keeps your Git history clean and focused on major, stable changes.
Q: Can I use this without Claude, just from my terminal?
A: Yes. All ClaudePoint features are available through direct CLI commands, so you can manually create, list, and restore checkpoints without interacting with the AI.
Q: What if a restore fails?
A: An emergency backup auto-creates before every restore. Find it in .checkpoints/ as emergency_backup_<timestamp>.
Q: How do I purge old checkpoints?
A: Set maxCheckpoints in config.json (default: 10). Old ones auto-delete.
Latest MCP Servers
Obsidian Web
Claude Peers
Memo
Featured MCP Servers
Claude Peers
Excalidraw
Claude Context Mode
FAQs
Q: What exactly is the Model Context Protocol (MCP)?
A: MCP is an open standard, like a common language, that lets AI applications (clients) and external data sources or tools (servers) talk to each other. It helps AI models get the context (data, instructions, tools) they need from outside systems to give more accurate and relevant responses. Think of it as a universal adapter for AI connections.
Q: How is MCP different from OpenAI's function calling or plugins?
A: While OpenAI's tools allow models to use specific external functions, MCP is a broader, open standard. It covers not just tool use, but also providing structured data (Resources) and instruction templates (Prompts) as context. Being an open standard means it's not tied to one company's models or platform. OpenAI has even started adopting MCP in its Agents SDK.
Q: Can I use MCP with frameworks like LangChain?
A: Yes, MCP is designed to complement frameworks like LangChain or LlamaIndex. Instead of relying solely on custom connectors within these frameworks, you can use MCP as a standardized bridge to connect to various tools and data sources. There's potential for interoperability, like converting MCP tools into LangChain tools.
Q: Why was MCP created? What problem does it solve?
A: It was created because large language models often lack real-time information and connecting them to external data/tools required custom, complex integrations for each pair. MCP solves this by providing a standard way to connect, reducing development time, complexity, and cost, and enabling better interoperability between different AI models and tools.
Q: Is MCP secure? What are the main risks?
A: Security is a major consideration. While MCP includes principles like user consent and control, risks exist. These include potential server compromises leading to token theft, indirect prompt injection attacks, excessive permissions, context data leakage, session hijacking, and vulnerabilities in server implementations. Implementing robust security measures like OAuth 2.1, TLS, strict permissions, and monitoring is crucial.
Q: Who is behind MCP?
A: MCP was initially developed and open-sourced by Anthropic. However, it's an open standard with active contributions from the community, including companies like Microsoft and VMware Tanzu who maintain official SDKs.



