Reading
A versionable personal AI development control plane that integrates skills, prompts, MCP tools, knowledge hubs, and workflows. It supports automatic generation of plans and outputs from conversation records, creating an evolvable AI-assisted development environment.
Section 01
A versionable personal AI development control plane that integrates skills, prompts, MCP tools, knowledge hubs, and workflows. It supports automatic generation of plans and outputs from conversation records, creating an evolvable AI-assisted development environment.
Section 02
Section 03
With the popularity of AI coding assistants like Cursor, Claude, and GitHub Copilot, 'Vibe Coding' has become a hot concept in the developer community. This development model emphasizes smooth collaboration with AI, allowing developers to focus on high-level intentions while delegating implementation details to AI. However, as the AI toolchain expands, a new problem emerges: How to effectively organize these tools? How to make AI understand your tech stack and preferences in every conversation? How to avoid repeating context explanations in each round of dialogue?
The Vibe Coding Stack project by spacebuilder13 is designed to address these organizational issues. It provides a personalized, versionable control plane for managing all elements of AI-assisted development.
Section 04
Vibe Coding Stack is a structured knowledge base and configuration collection, not a fixed template. Its core idea is: Your AI development environment should be an evolvable source of truth that continuously updates as you discover better patterns, rather than a disposable scaffold.
Main uses of the project include:
For each development session: Clone or submodule this repository, let Cursor/Claude read AGENTS.md and .cursor/rules, and automatically inherit default configurations.
For continuous evolution: Bump the VERSION, append to CHANGELOG.md, add new knowledge hubs and prompts, and record the best practices you've discovered.
Ultimate goal: Ingest a recorded conversation (or transcribed text) about someone's problem; the AI can produce a plan plus the first concrete output before the conversation ends—without you having to manually repeat context explanations.
Section 05
The project uses a clear layered structure, with each directory having a distinct responsibility:
Section 06
| Path | Role |
|---|---|
AGENTS.md |
Agent Entry: Priority, tech stack hierarchy, when to use what |
.cursor/rules/vibe-stack.mdc |
Cursor native rule file: Reinforces the same configuration |
These two files form the 'onboarding materials' for AI assistants. When the AI starts working, it first reads these files to understand the project's tech stack, priorities, and working methods.
Section 07
| Path | Role |
|---|---|
boards/ |
Architecture Board: Single source of truth, including ingestion v3 diagrams, blueprints, and local previews |
docs/github-landscape.md |
Your public GitHub work evolution date snapshot (context, not a rule) |
docs/mde-agent-flow.md |
Consumable architecture writing example (Markdown + Mermaid + conventions) |
This layer focuses on the documentation and visualization of system design. The architecture board provides a high-level view, the GitHub landscape records historical evolution, and the architecture document demonstrates how to write technical documents in a consumable way.
Section 08
| Path | Role |
|---|---|
hubs/ |
Knowledge Hubs: Checklists and conventions for Notebook LM, documents, design sources, Magic Patterns links, etc. |
notebooks/ |
Runnable narrative demos (Markdown -> Code -> Visualization) for explainable workflows |
prompts/ |
Reusable prompt shells (handover, plan prioritization, release cycle) |
The knowledge layer is the core asset of the project. Knowledge hubs are structured information repositories, notebooks combine documents with executable code, and prompts are validated interaction templates.
Keep going with more reads from the same topic.
Nornir MCP Server is an enterprise-level server based on the Model Context Protocol (MCP). It seamlessly integrates large language models (such as Claude) with the Nornir network automation framework, supporting natural language orchestration for multi-vendor network devices (Cisco, Arista, Juniper, etc.), and providing production-grade features like a dual-engine architecture (NAPALM + Netmiko), intelligent filtering, and a secure sandbox.
Bibliothèque Française LLM is a structured indexing and annotation project for French public domain literature designed specifically for large language models (LLMs). It integrates multiple authoritative sources such as DraCor, Common Corpus, and Wikisource, providing metadata indexing categorized by genre, author, and era, as well as in-depth annotations for dramatic texts (including characters, lines, stage directions, etc.). Its aim is to enable LLMs to efficiently read and understand classic French literary works.
Splinter is a minimalist, high-performance key-value (KV) and vector storage system enabling zero-latency inter-process communication via shared memory and atomic operations. With only 766 lines of core code, it supports millions of operations per second and 768-dimensional vector storage, offering a new architectural approach for local LLM inference and data-intensive applications.
Folkering OS is the world's first AI-native bare-metal operating system, entirely written in Rust no_std without relying on Linux, POSIX, or libc. It can generate commands from scratch, compile them into WASM, and run them in 10 seconds, achieving true self-evolution.