rag-implementation

Build Retrieval-Augmented Generation (RAG) systems for LLM applications with vector databases and semantic search. Use when implementing knowledge-grounded AI, building document Q&A systems, or integrating LLMs with external knowledge bases.

Quality

—

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Quality

Content

72%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

An actionable, well-structured overview with executable code and clean one-level-deep progressive disclosure. It would benefit from trimming concept explanations Claude already knows and adding validation checkpoints to the RAG workflow.

Suggestions

Remove or tighten "Purpose" lines that define embeddings/vector databases — Claude already knows these concepts; keep only decision-relevant detail.

Add an explicit validation/retry checkpoint to the workflow (e.g., check retrieved context sufficiency before generating, and fall back to re-retrieval or state "I don't know" when context is empty).

Consider moving the embeddings model table and reranking method list into references/details.md to keep the overview leaner, since the quick-start already shows the core flow.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient (tables, tight bullet lists), but lines like "Purpose: Convert text to numerical vectors for similarity search" explain concepts Claude already knows and could be trimmed.	2 / 3
Actionability	Provides a fully executable LangGraph quick-start with real imports and a concrete two-node graph, plus specific named models, libraries, and methods throughout.	3 / 3
Workflow Clarity	The retrieve→generate sequence is clear in both prose and code, but there are no explicit validation checkpoints or error-recovery feedback loops for the pipeline.	2 / 3
Progressive Disclosure	SKILL.md is a well-organized overview with a single clearly signaled, one-level-deep reference to the real bundle file references/details.md for detailed patterns.	3 / 3
	Total	10 / 12 Passed

Description

90%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, third-person description that clearly defines a distinct niche and pairs concrete capabilities with explicit "Use when" triggers. The only weakness is that it leads with a single action verb rather than enumerating several concrete capabilities.

Dimension	Reasoning	Score
Specificity	Names the domain and concrete technologies ("Build Retrieval-Augmented Generation (RAG) systems... with vector databases and semantic search") but relies on a single main verb rather than a comprehensive list of multiple distinct actions.	2 / 3
Completeness	Explicitly states what it does (build RAG systems with vector DBs and semantic search) and when to use it via a clear "Use when..." clause, satisfying both requirements.	3 / 3
Trigger Term Quality	Includes natural terms users would say — "RAG", "document Q&A systems", "knowledge-grounded AI", "external knowledge bases" — giving good coverage of likely phrasings.	3 / 3
Distinctiveness Conflict Risk	The RAG niche with triggers like "document Q&A systems" and "external knowledge bases" is distinct and unlikely to fire for unrelated skills.	3 / 3
	Total	11 / 12 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: wshobson/agents
Commit: 5cc2549

Reviewed: 8 days ago

Table of Contents

Discovery Implementation Validation

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.