github-starred/awesome-llm-apps
2
0
Fork 0
mirror of https://github.com/Shubhamsaboo/awesome-llm-apps.git synced 2026-07-18 10:48:20 -05:00
Code Issues 136 Packages Projects Releases Wiki Activity

[GH-ISSUE #446] Feature Request: add a git-like, traceable memory mcp tool for coding apps #12652

New Issue
Closed
opened 2026-06-21 05:19:00 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-06-21 05:19:00 -05:00
Owner
Copy Link

Originally created by @IRONICBo on GitHub (Jan 29, 2026).
Original GitHub issue: https://github.com/Shubhamsaboo/awesome-llm-apps/issues/446

I'd love to see a tutorial showing how to build LLM apps with MemoV context!

What is MemoV?

MemoV is an open-source tool that provides traceable, Git-powered version control for AI coding sessions. It creates a searchable memory of everything you've coded with prompts, responses, and code diffs - featuring automatic Skills learning from your successful patterns.

Integration Potential

MemoV would be perfect for several tutorial categories:

1. RAG with Coding Context

Build a RAG system that can answer questions about your entire coding history:

  • "What was the authentication logic I implemented last Tuesday?"
  • "Find all the bug fixes I made this week"
  • "How did I solve the database connection issue before?"

2. AI Agent with Skills Memory

Create agents that learn from your successful coding patterns:

  • Skills System: Auto-learns patterns like "JWT auth implementation" or "React state setup"
  • Personal coding assistant that knows your project's evolution
  • Debugging assistant that recalls similar issues you've fixed

3. Chat with Your Coding History

Similar to "Chat with PDF" but for your entire dev timeline:

  • Query weeks of coding sessions with natural language
  • Extract insights from any code you've modified
  • Cross-reference decisions across features with vibe_search

Technical Details

MemoV provides:

  • MCP Tools: snap, mem_history, mem_jump, mem_ui for version control
  • Skills System: AI-driven clustering with SQLite cache (GET /api/skills, POST /api/skills/refresh)
  • RAG Tools: vibe_search for semantic search, vibe_debug for multi-model debugging
  • Local-first: All data in .mem directory, zero pollution to .git

Resources

Would love to contribute this tutorial to awesome-llm-apps!

Originally created by @IRONICBo on GitHub (Jan 29, 2026). Original GitHub issue: https://github.com/Shubhamsaboo/awesome-llm-apps/issues/446 I'd love to see a tutorial showing how to build LLM apps with **MemoV** context! ## What is MemoV? [MemoV](https://github.com/memovai/memov) is an open-source tool that provides traceable, Git-powered version control for AI coding sessions. It creates a searchable memory of everything you've coded with prompts, responses, and code diffs - featuring automatic **Skills learning** from your successful patterns. ## Integration Potential MemoV would be perfect for several tutorial categories: ### 1. **RAG with Coding Context** Build a RAG system that can answer questions about your entire coding history: * "What was the authentication logic I implemented last Tuesday?" * "Find all the bug fixes I made this week" * "How did I solve the database connection issue before?" ### 2. **AI Agent with Skills Memory** Create agents that learn from your successful coding patterns: * **Skills System**: Auto-learns patterns like "JWT auth implementation" or "React state setup" * Personal coding assistant that knows your project's evolution * Debugging assistant that recalls similar issues you've fixed ### 3. **Chat with Your Coding History** Similar to "Chat with PDF" but for your entire dev timeline: * Query weeks of coding sessions with natural language * Extract insights from any code you've modified * Cross-reference decisions across features with `vibe_search` ## Technical Details MemoV provides: * **MCP Tools**: `snap`, `mem_history`, `mem_jump`, `mem_ui` for version control * **Skills System**: AI-driven clustering with SQLite cache (`GET /api/skills`, `POST /api/skills/refresh`) * **RAG Tools**: `vibe_search` for semantic search, `vibe_debug` for multi-model debugging * **Local-first**: All data in `.mem` directory, zero pollution to `.git` ## Resources * **GitHub**: https://github.com/memovai/memov * **Documentation**: https://docs.memov.ai * **Discord**: https://discord.gg/un54aD7Hug Would love to contribute this tutorial to awesome-llm-apps!
GiteaMirror commented 2026-06-21 05:19:02 -05:00
Author
Owner
Copy Link

@IRONICBo commented on GitHub (Jan 29, 2026):

Auto generate code skills in memory.

Image

Trace every interactions from llms, like inputs, outputs and context changes. Give the whole picture of actions.
Image

<!-- gh-comment-id:3817614500 --> @IRONICBo commented on GitHub (Jan 29, 2026): Auto generate code skills in memory. <img width="1909" height="838" alt="Image" src="https://github.com/user-attachments/assets/29857144-ce36-441a-adf3-9381a0b59af2" /> Trace every interactions from llms, like inputs, outputs and context changes. Give the whole picture of actions. <img width="2400" height="996" alt="Image" src="https://github.com/user-attachments/assets/e0ad4726-c1e4-4649-b7b5-7820f722385b" />
GiteaMirror commented 2026-06-21 05:19:03 -05:00
Author
Owner
Copy Link

@awesomekoder commented on GitHub (Feb 1, 2026):

Hey @IRONICBo! 👋 Thanks for bringing MemoV to our attention - the concept of Git-powered version control for AI coding sessions is really interesting!

Looking at how this could fit into our repo, I'd suggest focusing on one clear, focused tutorial rather than trying to cover all three ideas at once. Here's what I think would work best:

🎯 Recommended: "Chat with Your Coding History" Agent

This aligns perfectly with our existing patterns (like chat_with_pdf, chat_with_github) and would be super practical for developers.

Suggested Structure

mcp_ai_agents/
└── memov_coding_history_agent/
    ├── README.md
    ├── memov_agent.py
    └── requirements.txt

What the tutorial should include:

  1. Streamlit UI (we use Streamlit for most demos)
  2. Clear use case: Query your coding history with natural language
    • "What authentication patterns have I used?"
    • "Show me how I fixed that async bug last week"
    • "Find all database migrations I've done"
  3. MCP integration using MemoV's vibe_search and mem_history tools
  4. Simple setup - users should be able to run it in <5 minutes

README Style

Check out github_mcp_agent for our typical README format:

  • Features list with bullet points
  • Clear requirements & installation steps
  • Example queries section
  • Keep it focused and beginner-friendly

Key Tips

  • Keep the initial PR small and focused (~100-200 lines of code)
  • Include 3-5 practical example queries
  • Make sure it works with a fresh MemoV install
  • Add comments explaining the MCP tool calls

Would love to see a PR for this! Let me know if you have questions about the implementation. 🚀

<!-- gh-comment-id:3831638295 --> @awesomekoder commented on GitHub (Feb 1, 2026): Hey @IRONICBo! 👋 Thanks for bringing MemoV to our attention - the concept of Git-powered version control for AI coding sessions is really interesting! Looking at how this could fit into our repo, I'd suggest focusing on **one clear, focused tutorial** rather than trying to cover all three ideas at once. Here's what I think would work best: ## 🎯 Recommended: "Chat with Your Coding History" Agent This aligns perfectly with our existing patterns (like `chat_with_pdf`, `chat_with_github`) and would be super practical for developers. ### Suggested Structure ``` mcp_ai_agents/ └── memov_coding_history_agent/ ├── README.md ├── memov_agent.py └── requirements.txt ``` ### What the tutorial should include: 1. **Streamlit UI** (we use Streamlit for most demos) 2. **Clear use case**: Query your coding history with natural language - "What authentication patterns have I used?" - "Show me how I fixed that async bug last week" - "Find all database migrations I've done" 3. **MCP integration** using MemoV's `vibe_search` and `mem_history` tools 4. **Simple setup** - users should be able to run it in <5 minutes ### README Style Check out [`github_mcp_agent`](https://github.com/Shubhamsaboo/awesome-llm-apps/tree/main/mcp_ai_agents/github_mcp_agent) for our typical README format: - Features list with bullet points - Clear requirements & installation steps - Example queries section - Keep it focused and beginner-friendly ### Key Tips - Keep the initial PR small and focused (~100-200 lines of code) - Include 3-5 practical example queries - Make sure it works with a fresh MemoV install - Add comments explaining the MCP tool calls Would love to see a PR for this! Let me know if you have questions about the implementation. 🚀
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
bug
pull-request
Mirrored from GitHub Pull Request
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/awesome-llm-apps#12652
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?