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.

2.12x

Quality

71%

Does it follow best practices?

Impact

70%

2.12x

Average score across 3 eval scenarios

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./tests/ext_conformance/artifacts/agents-wshobson/llm-application-dev/skills/rag-implementation/SKILL.md

The canonical home for this skill is rag-implementation in wshobson/agents

Quality

Content

53%Weight 40%Scale 1-5

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

The content is highly actionable with executable code throughout, but it is bloated for a SKILL.md overview: ~560 lines of inlined reference material that should be split into separate files, with no progressive disclosure or explicit validation checkpoints in the workflow. It explains concepts Claude already knows and would benefit from a leaner overview pointing to dedicated references.

Suggestions

Slim the overview to the Quick Start plus a 'When to use' and a brief component map, then move chunking strategies, vector store configurations, retrieval optimization, prompt engineering, and evaluation into separate one-level-deep reference files (e.g. references/chunking.md, references/vectorstores.md) linked from SKILL.md.

Remove explanatory 'Purpose' lines that restate concepts Claude already knows (e.g. 'Convert text to numerical vectors for similarity search', 'Store and retrieve document embeddings efficiently').

Add an explicit end-to-end indexing/retrieval workflow with validation checkpoints (e.g. verify chunk count and embedding dimensions after ingestion, validate retrieval recall on a sample query before deployment) so destructive/batch operations have feedback loops.

Dimension	Reasoning	Score
Conciseness	The body is noticeably verbose — it restates concepts Claude already knows (purpose sentences like 'Store and retrieve document embeddings efficiently', 'Convert text to numerical vectors for similarity search') and pads the SKILL.md with ~560 lines that largely belong in reference files rather than the overview.	2 / 5
Actionability	Guidance is fully executable and copy-paste ready: complete, runnable Python snippets for every pattern (Quick Start LangGraph graph, hybrid search, HyDE, chunking, vector store setup, reranking, structured output, evaluation), with concrete imports and parameter values.	5 / 5
Workflow Clarity	Steps are present within code (the StateGraph edges sequence retrieve->generate) but there is no explicit end-to-end workflow with validation checkpoints for destructive/batch operations like indexing or bulk ingestion, which caps this at 3 per the rubric.	3 / 5
Progressive Disclosure	The file is monolithic — chunking strategies, vector store configs, retrieval optimization, prompt engineering, and evaluation are all inlined in SKILL.md with no references/ or other bundle files, burying detail that clearly belongs in separate one-level-deep reference files.	2 / 5
	Total	12 / 20 Passed

Description

88%Weight 40%Scale 1-5

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

The description is strong: it states concrete capabilities and provides an explicit 'Use when' clause with natural trigger phrases, clearly answering both what the skill does and when to invoke it. Minor gains are possible from adding more trigger synonyms and acronym variants.

Dimension	Reasoning	Score
Specificity	Names the domain ('RAG systems for LLM applications') and multiple concrete actions ('vector databases and semantic search', 'knowledge-grounded AI', 'document Q&A systems', 'integrating LLMs with external knowledge bases'), giving comprehensive coverage of what the skill builds.	5 / 5
Completeness	Explicitly answers both 'what' ('Build RAG systems... with vector databases and semantic search') and 'when' ('Use when implementing knowledge-grounded AI, building document Q&A systems, or integrating LLMs with external knowledge bases') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Includes strong natural phrases users say ('RAG', 'document Q&A systems', 'knowledge-grounded AI', 'external knowledge bases') but lacks the explicit acronym variations and file/extension synonyms that would warrant a 5; a few common terms are missing.	4 / 5
Distinctiveness Conflict Risk	The RAG/knowledge-grounding niche is fairly distinct with clear triggers, but it has minor overlap risk with adjacent skills like semantic search or document processing that also touch embeddings and Q&A.	4 / 5
	Total	18 / 20 Passed

Validation

93%

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 — 15 / 16 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	15 / 16 Passed

Repository: Dicklesworthstone/pi_agent_rust
Path: tests/ext_conformance/artifacts/agents-wshobson/llm-application-dev/skills/rag-implementation/SKILL.md
Commit: 6453c0b

Reviewed: about 8 hours 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.