skills.rest
Whaleylaw
WhaleylawCOMMUNITY
Explore all skills by this author →
GitHub

archon-rag-specialist

Index and retrieve semantically relevant document chunks for retrieval-augmented generation with Archon.

Updated Aug 23, 2026
Search Tags:#embeddings#rag#chunking#reranking#archon#semantic-search#vector-search
One-click install
npx skills add https://github.com/Whaleylaw/llm-lawyer --skill archon-rag-specialist
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: archon-rag-specialist
Source: https://github.com/Whaleylaw/llm-lawyer/tree/main/.claude/skills/archon-rag-specialist
Command: npx skills add https://github.com/Whaleylaw/llm-lawyer --skill archon-rag-specialist

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill streamlines implementing, operating, and debugging retrieval-augmented generation (RAG) with Archon to produce relevant, attributed context for LLMs and reduce irrelevant or stale search results.

Core Features & Use Cases

  • Indexing & Chunking: Best practices for chunk strategies, incremental and batch indexing, and parent document mapping to preserve provenance.
  • Embeddings & Storage: Guidance on choosing embedding models, generating embeddings, and storing vectors in a vector database.
  • Search, Filtering & Reranking: Vector, keyword, and hybrid search patterns, metadata filtering, reranking with cross-encoders, and deduplication.
  • Context Building & Debugging: Techniques for dynamic context windows, multi-query RAG, scoring analysis, and diagnosing embedding or chunking quality issues.
  • Use case: Index documentation and support articles to answer configuration questions with high-precision context for developer tools or support agents.

Quick Start

Use Archon to index your documentation with 500-character chunks, generate embeddings using a high-quality model, store vectors in your vector DB, and run a top-k vector search to retrieve context for a user query.

Frequently Asked Questions about archon-rag-specialist

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

FAQPage Schema
How do I implement retrieval-augmented generation with chunking and vector search?▼

Retrieval-augmented generation requires chunking documents, generating embeddings, storing vectors in a database, and running top-k vector search to retrieve semantically relevant context for LLM prompts.

What is the best way to chunk documents for semantic search?▼

The best way to chunk documents for semantic search involves applying best practices for chunk strategies, such as using 500-character chunks, and mapping parent documents to preserve provenance during indexing.

How does reranking improve retrieval-augmented generation results?▼

Reranking improves retrieval-augmented generation by applying cross-encoders to refine initial top-k vector search results, deduplicating near-identical entries, and ensuring only the most highly relevant chunks reach the LLM prompt.

Can I use hybrid search and metadata filtering for RAG indexing?▼

Yes, you can use hybrid search patterns combining vector and keyword search, apply metadata filtering to narrow results, and manage incremental or batch indexing for your document pipelines.

Why does my RAG pipeline return irrelevant or stale search results?▼

Irrelevant or stale search results in a RAG pipeline often stem from poor chunking quality, suboptimal embedding model selection, or index inconsistencies, which you can diagnose using scoring analysis and context debugging techniques.

Related Skills

Explore More Software Engineering Skills→
RepairYourTechRepairYourTech
6

lancedb

Implement LanceDB vector databases for RAG pipelines and semantic search.

Community
Advanced
jax2730jax2730

context-optimization

Optimize LLM context windows via compaction, masking, and partitioning.

Community
Advanced
aisa-groupaisa-group
90

rag-implementation

Build retrieval-augmented generation systems with embeddings, vector stores, and reranking.

Official
Intermediate
docutraydocutray
2

rag-research

Optimize document retrieval with RAG-based semantic search across indexed content.

Official
Intermediate