rag-architect

Use when building RAG systems, vector databases, or knowledge-grounded AI applications requiring semantic search, document retrieval, or context augmentation.

Install with Tessl CLI

npx tessl i github:jeffallan/claude-skills --skill rag-architect

What are skills?

Review — 48%

Does it follow best practices?

If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:

npx tessl skill review --optimize ./path/to/skill

Learn more

Validation — 11 / 11 Passed

Validation for skill structure

SKILL.md

Review

Evals

Evaluation results

95%

Document Ingestion Pipeline for Legal Knowledge Base

Document ingestion pipeline design

Criteria

Without context

With context

No default 512 chunk size

100%

Document-type-specific sizing

100%

Rationale for chunk size

100%

Source metadata on chunks

100%

Section/heading metadata

100%

Timestamp or indexed_at metadata

100%

Document preprocessing

60%

50%

Deduplication mechanism

100%

Idempotent re-run design

100%

Markdown-aware chunking

100%

Chunk index/position metadata

100%

Without context: $0.4861 · 2m 20s · 18 turns · 18 in / 7,817 out tokens

With context: $0.6574 · 2m 44s · 20 turns · 81 in / 8,900 out tokens

100%

Retrieval Pipeline Upgrade for Developer Documentation Search

Hybrid search and reranking pipeline

Criteria

Without context

With context

Hybrid search implemented

100%

RRF or equivalent fusion

100%

Retrieve-more then rerank fewer

100%

Reranking step present

100%

Not cosine-only

100%

Decoupled embedding model

100%

Empty result edge case

100%

Fusion weighting documented

100%

Design covers all four aspects

100%

Deduplication before rerank

100%

Without context: $0.3723 · 2m 5s · 10 turns · 11 in / 7,722 out tokens

With context: $0.5638 · 2m 24s · 18 turns · 265 in / 7,819 out tokens

99%

14%

RAG Architecture for Multi-Tenant Clinical Documentation Platform

RAG architecture design and evaluation plan

Criteria

Without context

With context

Ingestion pipeline diagram

100%

Retrieval pipeline diagram

100%

Vector DB recommendation with comparison

100%

Chunking parameters specified

71%

85%

Chunking rationale provided

100%

Retrieval metrics specified

100%

RAGAS or equivalent framework

100%

Multi-tenant isolation addressed

100%

Embedding versioning plan

100%

Decoupled embedding adapter

100%

Monitoring plan

62%

100%

Hybrid search in retrieval design

100%

Without context: $0.4328 · 2m 39s · 12 turns · 54 in / 9,192 out tokens

With context: $1.2683 · 4m 23s · 23 turns · 25,611 in / 15,382 out tokens

Evaluated: 3 days ago
Agent: Claude Code

Table of Contents

Document Ingestion Pipeline for Legal Knowledge Base Retrieval Pipeline Upgrade for Developer Documentation Search RAG Architecture for Multi-Tenant Clinical Documentation Platform

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.