sentence-transformers

Framework for state-of-the-art sentence, text, and image embeddings. Provides 5000+ pre-trained models for semantic similarity, clustering, and retrieval. Supports multilingual, domain-specific, and multimodal models. Use for generating embeddings for RAG, semantic search, or similarity tasks. Best for production embedding generation.

Quality

76%

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 ./skills/rag/sentence-transformers/SKILL.md

Quality

Content

65%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.

The content excels at actionability with comprehensive executable examples, but is weighed down by redundancy, missing validation checkpoints for batch/fine-tuning operations, and a references/ bundle that exists but is never referenced from the body.

Suggestions

Link references/models.md from the 'Popular models' and 'Model selection guide' sections and move the duplicated model/performance tables there to reduce inlined redundancy and improve progressive disclosure.

Add validation checkpoints to batch encoding and fine-tuning workflows (e.g., verify embedding shape/norms after encode; evaluate the fine-tuned model on a held-out set before saving).

Trim promotional metrics and the duplicated model tables to tighten conciseness; keep one authoritative model reference and point to it.

Dimension	Reasoning	Score
Conciseness	Mostly efficient executable reference, but it is long with redundant content — model info appears in both 'Popular models', 'Model selection guide', and 'Performance' tables, plus promotional metrics ('15,700+ GitHub stars') that add little actionable value.	3 / 5
Actionability	Fully executable, copy-paste-ready code spans installation, encoding, semantic search, batch encoding, fine-tuning, and LangChain/LlamaIndex integration, covering the common cases.	5 / 5
Workflow Clarity	Fine-tuning and batch-encoding sequences are present, but batch/epoch operations lack validation checkpoints (no embedding-shape/norm verification, no fine-tuned-model evaluation), capping workflow clarity at 3 per the batch-operation rule.	3 / 5
Progressive Disclosure	Headers provide structure, but large model/performance tables and integration code are inlined, and the existing references/models.md is never linked from the body, leaving references un-signaled and content that should be separate inline.	3 / 5
	Total	14 / 20 Passed

Description

88%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.

The description is strong: it states concrete capabilities and provides explicit trigger guidance with natural terms. Minor gaps in synonym coverage and some overlap with adjacent embedding skills keep trigger-term quality and distinctiveness at 4 rather than 5.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete capabilities across categories — 'semantic similarity, clustering, and retrieval', 'multilingual, domain-specific, and multimodal models', and 'generating embeddings for RAG, semantic search, or similarity tasks' — matching the comprehensive-coverage anchor.	5 / 5
Completeness	Clearly answers 'what' ('Framework for state-of-the-art sentence, text, and image embeddings') and explicitly answers 'when' ('Use for generating embeddings for RAG, semantic search, or similarity tasks').	5 / 5
Trigger Term Quality	Strong natural terms a user would say ('embeddings', 'semantic search', 'RAG', 'similarity', 'clustering') but missing common synonyms/extensions like 'vector', 'cosine similarity', 'Hugging Face', or file extensions.	4 / 5
Distinctiveness Conflict Risk	The sentence-transformers embedding niche is mostly distinct, but overlap risk remains with general embedding skills (OpenAI Embeddings, Cohere Embed, Instructor are even named as alternatives).	4 / 5
	Total	18 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

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

	Total	15 / 16 Passed

Repository: OpenLAIR/dr-claw
Path: skills/rag/sentence-transformers/SKILL.md
Commit: 4b35756

Reviewed: 1 day 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.