Name: rag-engineer
Rating: 24 (1 reviews)
Author: administrakt0r

rag-engineer

I bridge the gap between raw documents and LLM understanding. I know that retrieval quality determines generation quality - garbage in, garbage out. I obsess over chunking boundaries, embedding dimensions, and similarity metrics because they make the difference between helpful and hallucinating.

Quality

14%

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/antigravity-bundle-agent-architect/skills/rag-engineer/SKILL.md

The canonical home for this skill is rag-engineer in administrakt0r/AI-Agents-Safe-Coding-Skills

Quality

Content

17%Scale 1-5

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

This skill reads more like a high-level role description or resume than actionable guidance for Claude. It lacks any executable code, concrete examples, or specific tool/library recommendations. The patterns are described in abstract bullet points inside misleadingly labeled code blocks, and the anti-patterns and sharp edges sections are incomplete or empty.

Suggestions

Replace abstract bullet-point 'code blocks' with actual executable code examples showing specific libraries (e.g., LangChain, LlamaIndex, chromadb) for chunking, embedding, and retrieval operations.

Complete the anti-patterns section with concrete examples of what to avoid and what to do instead, and fill in the truncated solutions in the sharp edges table.

Add a concrete end-to-end workflow with numbered steps, specific commands, and validation checkpoints (e.g., measuring retrieval recall before proceeding to generation).

Remove the capabilities list, requirements section, role description repetition, and tautological 'When to Use' section — these waste tokens without adding actionable value.

Dimension	Reasoning	Score
Conciseness	The skill repeats its description in the body, lists capabilities and requirements that Claude already knows, and includes vague anti-patterns with no content (just headers). The 'When to Use' section is a meaningless tautology. Significant padding throughout.	2 / 5
Actionability	Despite having code blocks, the content inside them is bullet-point descriptions, not executable code. There are no concrete commands, no actual code examples, no specific library usage, no API calls. Everything remains at the level of abstract advice (e.g., 'Use sentence boundaries, not token limits').	1 / 5
Workflow Clarity	The hierarchical retrieval pattern hints at a multi-step sequence (first pass, second pass), but steps are vague with no validation checkpoints, no error handling, and no concrete commands. The sharp edges table lists solutions but truncates them with colons and no actual content. No feedback loops exist.	2 / 5
Progressive Disclosure	Everything is in a single monolithic file with no references to supporting documents. The anti-patterns section has empty headers with no content. The sharp edges table promises solutions but doesn't deliver them. There are no bundle files and no external references despite the content being broad enough to warrant them.	2 / 5
	Total	7 / 20 Passed

Description

11%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.

This description is written in first person ('I bridge', 'I know', 'I obsess') which violates voice guidelines, and reads more like a personal manifesto than a functional skill description. It contains zero concrete actions, no 'when to use' guidance, and relies on abstract philosophical statements rather than specifying what the skill actually does. The technical terms present are buried in fluff rather than serving as effective trigger terms.

Suggestions

Replace the entire description with concrete actions: e.g., 'Chunks documents, generates embeddings, indexes content in vector stores, and performs similarity search for RAG pipelines.'

Add an explicit 'Use when...' clause with natural trigger terms: e.g., 'Use when the user needs to set up RAG, build a knowledge base, chunk documents, create embeddings, or query a vector database.'

Rewrite in third person voice ('Processes documents for retrieval') instead of first person ('I bridge the gap') to follow proper skill description conventions.

Dimension	Reasoning	Score
Specificity	The description contains no concrete actions whatsoever. It uses entirely abstract language like 'bridge the gap', 'obsess over chunking boundaries' without specifying what the skill actually does (e.g., chunk documents, generate embeddings, query vector stores).	1 / 5
Completeness	The 'what' is extremely vague (no concrete actions described) and there is no 'when' clause at all. The description reads as philosophical commentary rather than functional documentation.	1 / 5
Trigger Term Quality	Contains some domain-relevant terms like 'chunking', 'embedding dimensions', 'similarity metrics', and 'retrieval', but these are technical jargon rather than natural user phrases. Missing natural terms users would say like 'RAG', 'vector database', 'search', 'knowledge base', 'document indexing'.	2 / 5
Distinctiveness Conflict Risk	While it vaguely points toward RAG/retrieval-augmented generation, the description is so abstract that it could overlap with any document processing, embedding, or search-related skill. The lack of specific actions makes it hard to distinguish from other skills in the same domain.	2 / 5
	Total	6 / 20 Passed

Validation

90%

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 — 10 / 11 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	10 / 11 Passed

Repository: administrakt0r/AI-Agents-Safe-Coding-Skills
Path: plugins/antigravity-bundle-agent-architect/skills/rag-engineer/SKILL.md
Commit: 57c135b

Reviewed: 2 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.