claude-code-memory-bank
Health Warn
- License — License: MIT
- Description — Repository has a description
- Active repo — Last push 0 days ago
- Low visibility — Only 5 GitHub stars
Code Pass
- Code scan — Scanned 1 files during light audit, no dangerous patterns found
Permissions Pass
- Permissions — No dangerous permissions requested
No AI report is available for this listing yet.
Persistent project memory for Claude Code: context, decisions and progress across sessions
Claude Code Memory Bank
Persistent project memory for Claude Code: keep context, architectural decisions and progress across sessions.
Memory Bank is a plug-and-play skill for Copilot and Claude code agents, inspired by Roo's Code Memory Bank by GreatScottyMac. It enables efficient, grounded, and context-aware coding workflows by providing persistent memory and specialized modes for different tasks.
Features
- Easy Installation: Just clone this repo into your
.claude/skills/folder (global or local) and it works out of the box. - Multiple Modes:
architect: For high-level design and architecture promptscode: For code generation and completiondebug: For debugging assistanceask: For general Q&A and memory retrieval
- Token Efficient: Saves tokens by grounding responses in persistent memory.
- Seamless Integration: Works natively with Copilot and Claude code agents.
Installation
Option A: Plugin marketplace (recommended)
/plugin marketplace add Shivansh12t/claude-code-memory-bank
/plugin install memory-bank@claude-code-memory-bank
Option B: Manual clone
- Open your terminal and navigate to your
.claude/skills/directory (create it if it doesn't exist):mkdir -p ~/.claude/skills/ cd ~/.claude/skills/ - Clone the repository:
git clone https://github.com/Shivansh12t/claude-code-memory-bank memory-bank - That's it! The skill is now available for use.
Usage
You can invoke Memory Bank in any prompt using the following commands:
- Ask a question:
/memory-bank ask <your prompt> - Architectural guidance:
/memory-bank architect <your prompt> - Debugging help:
/memory-bank debug <your prompt> - Code generation:
/memory-bank code <your prompt>
How It Works
Memory Bank creates a memory-bank/ folder in your project with five markdown files, each with timestamped, append-only entries:
| File | Purpose |
|---|---|
productContext.md |
Project goals, features, architecture |
activeContext.md |
Current focus, recent changes, open questions |
progress.md |
Completed, current and next tasks |
decisionLog.md |
Architectural decisions and rationale |
systemPatterns.md |
Coding and architectural patterns |
Say UMB or Update Memory Bank at any time to sync the files with your current session.
Why Use Memory Bank?
- Grounded Responses: Keeps context and memory across sessions for more relevant answers.
- Saves Tokens: Reduces prompt size by storing and reusing information.
- Flexible: Supports multiple workflows and prompt types.
FAQ
How do I make Claude Code remember my project between sessions?
Install Memory Bank and run /memory-bank architect <prompt>. Context is stored in plain markdown files in your repo, so it is versioned and portable.
Is this the same as Roo Code's Memory Bank?
It is inspired by it, adapted as a Claude Code skill and plugin.
Does it work in an existing project?
Yes. If no memory-bank/ folder exists, it offers to create one, and reads projectBrief.md if present.
License
MIT License
Contributions and feedback welcome!
Reviews (0)
Sign in to leave a review.
Leave a reviewNo results found