Tessl • The management layer for AI agent skills

SkillPublic

rag

Provides patterns to build Retrieval-Augmented Generation (RAG) systems for AI 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

73%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./plugins/developer-kit-ai/skills/rag/SKILL.md

The canonical home for this skill is giuseppe-trisciuoglio/developer-kit

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.

A content-rich skill with concrete, mostly-executable Java examples and a clear build sequence, but it is weighed down by duplicated sections, padded explanations, and a missing referenced asset. Workflow lacks explicit validation checkpoints.

Suggestions

Remove the duplicated 'When to Use' bullet list (it appears identically at lines 17-27 and 139-148) and consolidate 'Core Components', 'Quick Implementation', and 'Implementation Patterns' which restate the Steps and Examples sections.

Add explicit validation checkpoints between workflow steps (e.g., verify embeddings were stored, verify retrieval returns relevant results before building the pipeline) to lift workflow clarity.

Either add the missing assets/evaluation-metrics.java file or remove the broken reference to it in the Resources section.

Define or note the assumed variables (chatModel, embeddingModel, store, reranker) in code examples, or add imports, so examples are copy-paste ready.

Dimension	Reasoning	Score
Conciseness	Noticeably verbose with redundant padding: the entire 'When to Use' bullet list is duplicated verbatim, and 'Core Components', 'Quick Implementation', and 'Implementation Patterns' restate content already covered in the Steps and Examples. Also explains RAG basics Claude already knows. Not a 1 because the code blocks carry real content; not a 3 because the duplication and padding are substantial.	2 / 5
Actionability	Provides multiple concrete Java (LangChain4j) examples covering basic Q&A, metadata filtering, multi-source retrieval, chat memory, and the document processing pipeline. Not a 5 because several examples reference undefined variables (chatModel, embeddingModel, store, reranker) and lack imports, so they are not fully copy-paste ready; not a 3 because the guidance is genuinely executable beyond pseudocode.	4 / 5
Workflow Clarity	A six-step build sequence is present (choose vector DB, select embedding, implement processing, configure retrieval, build pipeline, evaluate), but there are no validation checkpoints between steps; Step 6 is a final phase rather than a checkpoint. Not a 4 because checkpoints are absent; not a 2 because the sequence itself is coherent and well-ordered.	3 / 5
Progressive Disclosure	References are clearly signaled one-level-deep in a Resources section with markdown links to real files (5 references/*.md, 2 of 3 assets exist), and the body is sectioned. Not a 5 because assets/evaluation-metrics.java is referenced but missing and substantial content duplicating the references is inlined; not a 3 because structure and signaling are otherwise good.	4 / 5
	Total	13 / 20 Passed

Description

83%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.

A strong, third-person description with an explicit 'Use when' clause and concrete capabilities. It clearly answers both what and when with natural trigger terms.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions (build RAG systems, vector databases, semantic search, document Q&A, integrate LLMs with external knowledge bases), though it omits specifics like chunking, embeddings, and reranking. Not a 5 because coverage has minor gaps; not a 3 because it goes beyond 1-2 actions.	4 / 5
Completeness	Explicitly states what it does ('Provides patterns to build RAG systems... with vector databases and semantic search') and when to use it ('Use when implementing knowledge-grounded AI, building document Q&A systems, or integrating LLMs with external knowledge bases') with concrete trigger phrases. Matches the anchor for both what and when.	5 / 5
Trigger Term Quality	Includes natural terms users would say (RAG, vector databases, semantic search, document Q&A, knowledge-grounded AI, LLMs) but misses common synonyms and extensions like 'embeddings', 'hallucination', or file types. Not a 5 because coverage lacks the comprehensive synonyms/extensions anchor; not a 3 because keyword set is solid.	4 / 5
Distinctiveness Conflict Risk	RAG is a recognizable niche with distinct triggers, but 'semantic search' and 'vector databases' could plausibly belong to sibling skills, creating minor overlap risk. Not a 5 due to that overlap risk; not a 3 because it is clearly more specific than a generic document/data skill.	4 / 5
	Total	17 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
referenced_paths_exist	Referenced path issues: 1 missing	Warning

	Total	14 / 16 Passed

Repository: fernandezbaptiste/developer-kit
Path: plugins/developer-kit-ai/skills/rag/SKILL.md
Revision: 810297c

Reviewed: 24 days ago

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.

rag

Table of Contents

Is this your skill?