embedding-strategies

Select and optimize embedding models for semantic search and RAG applications. Use when choosing embedding models, implementing chunking strategies, or optimizing embedding quality for specific domains.

Quality

82%

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

Quality

Content

65%

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

A content-rich, highly actionable reference with strong executable code, but it is a large monolithic file with generic instructional filler and no validation checkpoints or progressive file split to push conciseness and workflow clarity higher.

Suggestions

Tighten or remove the generic 'Instructions' and 'Do not use this skill when' sections, which add little that the description and code don't already convey.

Add explicit validation/verification steps to the embedding pipeline (e.g., check embedding dimensions and norms, verify chunk count vs. expected) to raise workflow clarity and satisfy the batch-operation feedback-loop guidance.

Move the longer templates (recursive splitter, DomainEmbeddingPipeline) into separate reference files linked from a concise overview to reduce inline volume and improve progressive disclosure.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient executable code with little concept-padding Claude already knows, but the generic "Instructions"/"Do not use this skill when" sections and very long templates (recursive_character_splitter, full DomainEmbeddingPipeline) add volume that could be tightened, sitting between anchors 2 and 3 rather than fully lean.	2 / 3
Actionability	It provides multiple copy-paste-ready, executable templates — OpenAI embeddings with batching, sentence-transformers LocalEmbedder/E5Embedder, four chunking strategies, and retrieval-quality metrics — with concrete function signatures and defaults.	3 / 3
Workflow Clarity	The pipeline diagram (Document → Chunking → Preprocessing → Embedding Model → Vector) implies a sequence, but there are no explicit validation/feedback checkpoints, and batch embedding operations lack the validate-fix-retry loop the scoring notes require for a 3.	2 / 3
Progressive Disclosure	No bundle files exist (references/scripts/assets absent) and everything — five templates, comparison table, best practices — is inline in one ~490-line SKILL.md; section headers give some organization, but content that could be split into separate references stays monolithic, matching the level-2 anchor.	2 / 3
	Total	9 / 12 Passed

Description

100%

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 description: third-person voice, concrete actions, explicit "Use when" triggers, and a clearly bounded embedding/RAG niche. It answers both what the skill does and when to use it without padding.

Dimension	Reasoning	Score
Specificity	The description lists multiple concrete actions — "Select and optimize embedding models", "implementing chunking strategies", and "optimizing embedding quality" — rather than vague language, matching the level-3 anchor.	3 / 3
Completeness	It explicitly answers both "what" (select and optimize embedding models for semantic search and RAG) and "when" via an explicit "Use when choosing embedding models, implementing chunking strategies, or optimizing embedding quality" trigger.	3 / 3
Trigger Term Quality	It surfaces natural terms users would actually say — "embedding models", "semantic search", "RAG", "chunking strategies" — giving good coverage rather than jargon-only phrasing.	3 / 3
Distinctiveness Conflict Risk	The embedding/RAG niche is clearly scoped with distinct triggers (embedding models, chunking, RAG), making it unlikely to fire for unrelated skills.	3 / 3
	Total	12 / 12 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: rmyndharis/antigravity-skills
Path: skills/embedding-strategies/SKILL.md
Commit: e63f7dd

Reviewed: about 3 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.