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.
77
66%
Does it follow best practices?
Impact
100%
1.13xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./tests/ext_conformance/artifacts/agents-wshobson/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 relevant trigger terms and is distinctive enough to avoid conflicts. The main weakness is that the capability listing could be more specific—mentioning concrete actions like comparing model benchmarks, configuring vector dimensions, or evaluating retrieval quality would strengthen it.
Suggestions
Add more specific concrete actions such as 'compare model benchmarks, configure vector dimensions, evaluate retrieval quality, select tokenizers' 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 applications') and when ('Use when choosing embedding models, implementing chunking strategies, or optimizing embedding quality for specific domains') with explicit trigger guidance. | 3 / 3 |
Trigger Term Quality | Includes strong natural keywords users would say: 'embedding models', 'semantic search', 'RAG', 'chunking strategies', 'embedding quality', 'domains'. These cover the main terms a user would naturally use when seeking help with embeddings. | 3 / 3 |
Distinctiveness Conflict Risk | The focus on embedding models, chunking strategies, and embedding quality for RAG/semantic search is a clear niche that is unlikely to conflict with general ML skills, search skills, or other NLP-related 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 highly actionable, executable code templates covering a broad range of embedding use cases, which is its primary strength. However, it is severely over-long and monolithic—all content is crammed into a single file with no progressive disclosure, making it a poor use of context window. It also lacks validation checkpoints in its workflows and includes some unnecessary explanatory content that Claude doesn't need.
Suggestions
Split the six code templates into separate bundle files (e.g., voyage_embeddings.py, openai_embeddings.py, chunking.py, evaluation.py) and reference them from a concise overview in SKILL.md
Add explicit validation steps to the embedding pipeline workflow, such as verifying embedding dimensions, checking for empty/truncated inputs, and handling API rate limits or failures
Remove the Do's/Don'ts section or reduce it to 2-3 non-obvious, domain-specific tips—most items are general knowledge Claude already has
Trim the SKILL.md to a concise overview (~50-80 lines) with a model comparison table, one short quick-start example, and clear links to detailed template files
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~400+ lines, with massive code templates that could be split into separate files. It includes unnecessary explanations (e.g., do's/don'ts that Claude already knows like 'garbage in, garbage out'), and the sheer volume of code templates—six full implementations—bloats the context window significantly. Much of this belongs in referenced bundle files. | 1 / 3 |
Actionability | The code templates are fully executable with concrete implementations for Voyage AI, OpenAI, local embeddings, chunking strategies, domain pipelines, and evaluation metrics. They include proper imports, type hints, and are copy-paste ready. | 3 / 3 |
Workflow Clarity | The embedding pipeline diagram provides a high-level sequence, and Template 5 shows a document processing workflow, but there are no explicit validation checkpoints or error recovery steps. For a pipeline involving batch embedding operations, missing verification steps (e.g., checking embedding dimensions, validating API responses, handling rate limits) is a notable gap. | 2 / 3 |
Progressive Disclosure | All six templates are inlined in a single monolithic file with no bundle files to offload detailed implementations. The chunking strategies, evaluation code, and domain-specific pipelines should each be in separate referenced files. The skill reads as a wall of code rather than an overview with navigation to details. | 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 (609 lines); consider splitting into references/ and linking | Warning |
Total | 10 / 11 Passed | |
- Repository
- Dicklesworthstone/pi_agent_rust
- Commit
bbc5ade
- 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.