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.
67
59%
Does it follow best practices?
Impact
70%
2.12xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./tests/ext_conformance/artifacts/agents-wshobson/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 a clear 'Use when' clause that covers the key scenarios. Its main weakness is that the 'what' portion is somewhat high-level—it describes the category of system to build but doesn't enumerate specific concrete actions like chunking, embedding, indexing, or retrieval. Adding more granular capabilities would improve specificity.
Suggestions
Add specific concrete actions such as 'chunk documents, generate embeddings, index vectors, retrieve relevant passages, and rerank results' to improve specificity beyond the high-level 'build RAG systems' framing.
| 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 queries, retrieving relevant passages, or reranking results. | 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 a developer would naturally use when seeking this capability. | 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, generic LLM skills, or document processing skills. The trigger terms are specific to this particular architectural pattern. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
29%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is a comprehensive reference catalog for RAG implementation with excellent, executable code examples, but it severely violates conciseness by including encyclopedic coverage of options Claude already knows (vector DB comparisons, embedding model tables, multiple chunking strategies). It lacks workflow validation checkpoints critical for a multi-step pipeline and dumps all content into a single monolithic file rather than using progressive disclosure.
Suggestions
Reduce the content to a focused Quick Start + 1-2 advanced patterns, moving vector store configs, chunking strategies, and retrieval optimization into separate bundle files (e.g., VECTOR_STORES.md, CHUNKING.md, RETRIEVAL_PATTERNS.md)
Remove the catalog-style listings (vector DB options table, embedding models table, retrieval strategies list) that Claude already knows, keeping only the specific recommended choices with brief justification
Add explicit validation checkpoints to the workflow: verify document ingestion count, test retrieval quality with a sample query before building the full pipeline, validate chunk sizes produce expected results
Add a clear end-to-end workflow section with numbered steps and validation gates: 1. Load docs → 2. Chunk (verify count) → 3. Embed & store (verify index) → 4. Test retrieval (verify relevance) → 5. Build generation pipeline
| 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 embeddings table, vector store options list, and retrieval strategies overview are reference material that doesn't need to be in a skill file. Multiple near-duplicate patterns (4 chunking strategies, 4 vector store configs, 4 reranking approaches) bloat the content significantly. | 1 / 3 |
Actionability | All code examples are concrete, executable, and use real library APIs with proper imports. The Quick Start with LangGraph is a complete, runnable example, and advanced patterns provide copy-paste ready implementations with specific library calls and configurations. | 3 / 3 |
Workflow Clarity | Despite RAG being a multi-step pipeline (ingest → chunk → embed → store → retrieve → generate), there are no validation checkpoints anywhere. No guidance on verifying embeddings were stored correctly, no checks on retrieval quality before generation, no error handling in the indexing pipeline. The evaluation section exists but is disconnected from the build workflow. The 'Common Issues' section lists problems without actionable diagnostic steps. | 1 / 3 |
Progressive Disclosure | Monolithic wall of content with no bundle files to offload detail into. Vector store configurations, chunking strategies, and retrieval optimization patterns should each be in separate reference files. The Resources section links to external docs but doesn't organize the skill's own content across files. Everything is inline in one massive document. | 1 / 3 |
Total | 6 / 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 (571 lines); consider splitting into references/ and linking | Warning |
Total | 10 / 11 Passed | |
- Repository
- Dicklesworthstone/pi_agent_rust
- Commit
b09ec7f
- 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.