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.
74
66%
Does it follow best practices?
Impact
83%
2.07xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./plugins/llm-application-dev/skills/rag-implementation/SKILL.mdQuality
Discovery
89%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a solid skill description with strong trigger terms and completeness, clearly identifying both what the skill does and when to use it. Its main weakness is that the 'what' portion could be more specific about the concrete actions involved in building RAG systems (e.g., chunking, embedding, retrieval, prompt augmentation). Overall it performs well for skill selection purposes.
Suggestions
Add more specific concrete actions like 'chunk documents, generate embeddings, perform similarity search, augment prompts with retrieved context' to improve specificity.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (RAG systems) and some actions ('build', 'implementing', 'integrating'), but doesn't list multiple concrete specific actions like indexing documents, chunking text, embedding generation, similarity search, or prompt construction with retrieved context. | 2 / 3 |
Completeness | Clearly answers both 'what' (build RAG systems with vector databases and semantic search) and 'when' (explicit 'Use when' clause covering implementing knowledge-grounded AI, building document Q&A systems, or integrating LLMs with external knowledge bases). | 3 / 3 |
Trigger Term Quality | Good coverage of natural terms users would say: 'RAG', 'Retrieval-Augmented Generation', 'vector databases', 'semantic search', 'document Q&A', 'knowledge bases', 'LLM applications', 'knowledge-grounded AI'. These are terms users working in this space would naturally use. | 3 / 3 |
Distinctiveness Conflict Risk | RAG systems, vector databases, and semantic search form a clear niche that is unlikely to conflict with general coding skills or other AI-related skills. The combination of these specific terms creates a distinct trigger profile. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
42%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides highly actionable, executable code examples covering a comprehensive range of RAG patterns, which is its primary strength. However, it is severely over-long and verbose, treating SKILL.md as an exhaustive reference rather than a concise overview. It explains concepts Claude already knows, catalogs options without clear recommendations, and dumps all content into a single file without progressive disclosure or validation workflows.
Suggestions
Reduce the content by 60-70%: remove the catalog-style listings (vector DB options, embedding model tables, retrieval strategy descriptions) and keep only the Quick Start + 2-3 most important patterns with executable code.
Split into multiple files: move vector store configurations to VECTOR_STORES.md, advanced patterns to ADVANCED_PATTERNS.md, chunking strategies to CHUNKING.md, and evaluation to EVALUATION.md, with clear links from the main SKILL.md.
Add validation checkpoints: include steps to verify embeddings were indexed correctly (e.g., test query after indexing), handle retrieval failures, and validate RAG output quality before serving responses.
Remove explanatory text Claude already knows (e.g., 'Purpose: Store and retrieve document embeddings efficiently', 'Purpose: Convert text to numerical vectors') and trust Claude's existing knowledge of these concepts.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at ~400+ lines. Includes extensive catalog-style listings of vector databases, embedding models, retrieval strategies, and reranking methods that Claude already knows. The embedding model table with dimensions, the list of vector DB options with taglines, and explanations of what embeddings are ('Convert text to numerical vectors for similarity search') are all unnecessary padding. | 1 / 3 |
Actionability | The code examples are concrete, executable, and copy-paste ready. Every pattern includes complete, runnable Python code with proper imports, initialization, and usage. The LangGraph, LangChain, and vector store configurations are all specific and functional. | 3 / 3 |
Workflow Clarity | The Quick Start section shows a clear retrieve→generate pipeline, and advanced patterns show sequenced steps. However, there are no validation checkpoints—no guidance on verifying embeddings were stored correctly, no error handling for failed retrievals, no feedback loops for when retrieval quality is poor. For a system involving external services and data pipelines, this is a significant gap. | 2 / 3 |
Progressive Disclosure | This is a monolithic wall of content with no references to external files. Everything from quick start to advanced patterns, chunking strategies, vector store configs, optimization, prompt engineering, and evaluation is inlined. This should be split across multiple files (e.g., CHUNKING.md, VECTOR_STORES.md, ADVANCED_PATTERNS.md) with the SKILL.md serving as an overview with links. | 1 / 3 |
Total | 7 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (543 lines); consider splitting into references/ and linking | Warning |
Total | 10 / 11 Passed | |
- Repository
- wshobson/agents
- Commit
27a7ed9
- Reviewed
Table of Contents
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.