rag-context

Query Qdrant and code-index-mcp to retrieve architectural notes and code snippets.

2|Updated Oct 1, 2024
One-click install
npx skills add https://github.com/ZeiZel/dotfiles --skill rag-context-zeizel
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rag-context
Source: https://github.com/ZeiZel/dotfiles/tree/main/.claude/skills/rag-context
Command: npx skills add https://github.com/ZeiZel/dotfiles --skill rag-context-zeizel

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

RAG-context retrieval enables fast, scalable access to code and architectural knowledge in large projects by combining vector search with code-index summaries, reducing prompt length while preserving key insights.

Core Features & Use Cases

  • Hybrid search across code and architecture documents to surface relevant patterns, ADRs, and design decisions.
  • Auto-selects retrieval strategy (repomix vs RAG) based on repository size and Qdrant health to optimize results.
  • Supports agent spawning and context packing by providing pre-loaded context for complex tasks.

Quick Start

Run the rag-context skill to automatically configure and perform a context pull for your current project.

Frequently Asked Questions about rag-context

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I retrieve code context from large repositories without exceeding token budgets?

RAG-based code context retrieval uses vector search and code-index summaries to extract relevant architectural notes and snippets, reducing prompt length while preserving key insights for large codebases.

What is the best way to perform semantic search across code and architecture documents?

Semantic search across code and architecture documents is performed using a hybrid approach, querying Qdrant and code-index-mcp to surface relevant patterns, ADRs, and design decisions.

Does rag-context work with Qdrant and code-index-mcp for context retrieval?

Yes, rag-context interfaces directly with mcp__qdrant and mcp__code-index tools, performing discovery of project path, building indexes, and retrieving file summaries.

How does the skill choose between repomix and RAG strategies for repository indexing?

The retrieval strategy is auto-selected by evaluating repository size and the health of the Qdrant index, choosing repomix or RAG to optimize context retrieval results.

Can I use rag-context to provide pre-loaded context for agent spawning?

Yes, rag-context supports agent spawning and context packing by providing pre-loaded context, enabling complex tasks to execute with relevant architectural knowledge.

When should I not use RAG-based context retrieval for my codebase?

RAG-based context retrieval is specifically applied to large codebases where token budgets are a concern; smaller repositories may not require this hybrid search and indexing strategy.