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.
75
66%
Does it follow best practices?
Impact
91%
1.65xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./plugins/llm-application-dev/skills/embedding-strategies/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 description that clearly defines its niche around embedding model selection and optimization for RAG/semantic search. It includes an explicit 'Use when' clause with good trigger terms. The main weakness is that the specific capabilities could be more granular—listing concrete actions like benchmarking models, configuring dimensions, or evaluating retrieval quality would strengthen it.
Suggestions
Add more specific concrete actions such as 'benchmark embedding models, configure chunk sizes and overlap, evaluate retrieval quality metrics' to improve specificity.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (embedding models, semantic search, RAG) and some actions (select, optimize, implement chunking strategies), but doesn't list multiple concrete specific actions like benchmarking, fine-tuning, dimension reduction, or specific model comparisons. | 2 / 3 |
Completeness | Clearly answers both 'what' (select and optimize embedding models for semantic search and RAG) and 'when' with an explicit 'Use when...' clause covering choosing embedding models, implementing chunking strategies, or optimizing embedding quality. | 3 / 3 |
Trigger Term Quality | Includes strong natural keywords users would say: 'embedding models', 'semantic search', 'RAG', 'chunking strategies', 'embedding quality', 'specific domains'. These cover the main terms a user would naturally use when needing this skill. | 3 / 3 |
Distinctiveness Conflict Risk | The focus on embedding models, chunking strategies, and embedding optimization for RAG/semantic search is a clear niche that is unlikely to conflict with general ML skills, vector database skills, or broader NLP skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
42%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides comprehensive, executable code templates for embedding workflows across multiple providers, which is its main strength. However, it is far too verbose for a SKILL.md — most of the code is standard library usage that Claude can generate on demand. The lack of progressive disclosure (everything in one file) and missing validation checkpoints in workflows significantly reduce its effectiveness as a concise, navigable skill reference.
Suggestions
Split the six code templates into separate reference files (e.g., VOYAGE.md, OPENAI.md, CHUNKING.md, EVALUATION.md) and keep only the model comparison table, pipeline diagram, and brief guidance in SKILL.md with links to each.
Reduce code templates to minimal differentiating snippets — e.g., just show the model-specific prefixes for BGE/E5, Matryoshka dimension parameter, and Voyage domain model names rather than full class implementations.
Add explicit validation steps to the pipeline workflow: verify embedding dimensions match expected values, check for empty/null embeddings, and validate chunk sizes before batch embedding.
Remove the do's/don'ts section or compress it to 2-3 critical non-obvious points — most items (batch requests, cache embeddings, normalize) are standard practices Claude already knows.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~400+ lines, with extensive code templates that cover multiple embedding providers, chunking strategies, evaluation metrics, and code-specific pipelines. Much of this is library boilerplate Claude already knows how to write. The model comparison table and do's/don'ts list add some value, but the bulk of the content is padded with standard Python patterns that don't need to be spelled out at this length. | 1 / 3 |
Actionability | The code templates are fully executable with concrete imports, function signatures, and complete implementations. Examples cover Voyage AI, OpenAI, local sentence-transformers, chunking strategies, and evaluation metrics — all copy-paste ready with real model names and API patterns. | 3 / 3 |
Workflow Clarity | The embedding pipeline diagram shows a high-level flow (Document → Chunking → Preprocessing → Embedding → Vector), and Template 5 shows a multi-step document processing pipeline. However, there are no explicit validation checkpoints — no steps to verify embedding quality before indexing, no error handling guidance for API failures, and no feedback loops for catching issues in the pipeline. | 2 / 3 |
Progressive Disclosure | All content is inlined in a single monolithic file with no references to external files. The six templates, evaluation code, and best practices could easily be split into separate reference files (e.g., CHUNKING.md, EVALUATION.md, MODELS.md) with a concise overview in the main skill. The current structure forces loading hundreds of lines of code templates regardless of the specific task. | 1 / 3 |
Total | 7 / 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 (601 lines); consider splitting into references/ and linking | Warning |
Total | 10 / 11 Passed | |
- Repository
- wshobson/agents
- Commit
27a7ed9
- 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.