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.

Install with Tessl CLI

npx tessl i github:wshobson/agents --skill embedding-strategies

What are skills?

Review — 77%

Does it follow best practices?

If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:

npx tessl skill review --optimize ./path/to/skill

Learn more

Validation — 10 / 11 Passed

Validation for skill structure

SKILL.md

Review

Evals

Evaluation results

100%

25%

Multi-Domain Document Embedding Service

Voyage AI model selection

Criteria

Without context

With context

Voyage AI library

100%

Legal model

100%

Code model

100%

General/default model

100%

API key from env

100%

Separate model instances

100%

Document embedding method

100%

Query embedding method

100%

Without context: $0.3491 · 4m 35s · 18 turns · 121 in / 4,998 out tokens

With context: $0.6587 · 6m 45s · 25 turns · 430 in / 6,689 out tokens

100%

10%

Offline Research Paper Search Engine

Local model prefixes and chunking

Criteria

Without context

With context

Local embedding library

100%

Recommended model name

100%

Normalized embeddings

100%

Query prefix applied

100%

Document prefix (E5) or no-prefix (BGE)

100%

Chunking with overlap

100%

Metadata stored

100%

Cosine similarity ranking

100%

Without context: $0.8039 · 8m 46s · 25 turns · 201 in / 12,793 out tokens

With context: $0.9980 · 8m 56s · 31 turns · 431 in / 13,035 out tokens

100%

Embedding Model Evaluation Toolkit

OpenAI batching and quality evaluation

Criteria

Without context

With context

OpenAI model selection

100%

Batching loop

100%

Batch size 100

100%

Matryoshka dimensions param

100%

Precision@k metric

100%

Recall@k metric

100%

MRR metric

100%

NDCG@k metric

100%

Without context: $1.1627 · 10m · 37 turns · 289 in / 15,169 out tokens

With context: $1.0525 · 10m 5s · 36 turns · 557 in / 12,251 out tokens

Evaluated: about 19 hours ago
Agent: Claude Code

Table of Contents

Multi-Domain Document Embedding Service Offline Research Paper Search Engine Embedding Model Evaluation Toolkit

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.