Embedding Strategies

From the author

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.

— the description by Seth Hobson

AutoGineer summary

Written by AutoGineer from the skill’s own files, not by the author

This skill provides a framework for choosing and implementing embedding models within RAG pipelines. It covers model selection based on specific use cases, such as code, legal, or financial domains, and offers best practices for chunking, normalization, and pipeline management. The skill includes technical templates for integrating Voyage AI, OpenAI, and local sentence-transformer models.

What it does

Compares various embedding models based on dimensions, token limits, and domain suitability.
Provides implementation templates for Voyage AI, OpenAI, and local models.
Outlines best practices for chunking, batching, and caching embeddings.
Explains the importance of preprocessing and metadata usage in vector search.
Demonstrates how to handle dimension reduction using Matryoshka embeddings.
Offers guidance on matching models to specific content types like code or multilingual text.

When to use it

When selecting an embedding model for a new RAG application.
When optimizing existing chunking strategies for better retrieval.
When needing to implement or switch between different embedding providers.
When fine-tuning or adapting embedding strategies for specialized domains.
When troubleshooting issues related to vector space compatibility or token limits.

Advantages

Provides clear criteria for matching models to specific application needs.
Includes ready-to-use code templates for major embedding providers.
Establishes a structured approach to the embedding pipeline.
Highlights critical pitfalls like mixing incompatible models or ignoring token limits.

Usage examples

“Which embedding model should I use for a legal document RAG system?”: The skill identifies voyage-law-2 as a specialized model for legal documents with 1024 dimensions.
“How do I implement OpenAI embeddings with dimension reduction?”: The skill provides a Python template demonstrating the use of the dimensions parameter for Matryoshka dimensionality reduction.
“What are the best practices for chunking text for embeddings?”: The skill advises preserving semantic boundaries and warns against over-chunking to avoid losing context.
“How do I use local models for embeddings?”: The skill provides a LocalEmbedder class template using sentence-transformers for local deployment.

Install

Download the skill as a ZIP:
https://autogineer.com/agent-skills/wshobson/embedding-strategies/download

Unzip it into the cross-client skills folder of your project:

mkdir -p .agents/skills && unzip embedding-strategies.zip -d .agents/skills

Documentation by Seth Hobson

Guide to selecting and optimizing embedding models for vector search applications.

When to Use This Skill

Choosing embedding models for RAG
Optimizing chunking strategies
Fine-tuning embeddings for domains
Comparing embedding model performance
Reducing embedding dimensions
Handling multilingual content

Core Concepts

1. Embedding Model Comparison (2026)

Model	Dimensions	Max Tokens	Best For
voyage-3-large	1024	32000	Claude apps (Anthropic recommended)
voyage-3	1024	32000	Claude apps, cost-effective
voyage-code-3	1024	32000	Code search
voyage-finance-2	1024	32000	Financial documents
voyage-law-2	1024	32000	Legal documents
text-embedding-3-large	3072	8191	OpenAI apps, high accuracy
text-embedding-3-small	1536	8191	OpenAI apps, cost-effective
bge-large-en-v1.5	1024	512	Open source, local deployment
all-MiniLM-L6-v2	384	256	Fast, lightweight
multilingual-e5-large	1024	512	Multi-language

2. Embedding Pipeline

Document → Chunking → Preprocessing → Embedding Model → Vector
                ↓
        [Overlap, Size]  [Clean, Normalize]  [API/Local]

Templates and detailed worked examples

Full template library and detailed worked examples live in references/details.md. Read that file when you need the concrete templates.

Best Practices

Do's

Match model to use case: Code vs prose vs multilingual
Chunk thoughtfully: Preserve semantic boundaries
Normalize embeddings: For cosine similarity search
Batch requests: More efficient than one-by-one
Cache embeddings: Avoid recomputing for static content
Use Voyage AI for Claude apps: Recommended by Anthropic

Don'ts

Don't ignore token limits: Truncation loses information
Don't mix embedding models: Incompatible vector spaces
Don't skip preprocessing: Garbage in, garbage out
Don't over-chunk: Lose important context
Don't forget metadata: Essential for filtering and debugging

Files

ATTRIBUTION.mdAdded by AutoGineer998 B
LICENSELicence file added by AutoGineer1.0 KiB
SKILL.md2.8 KiB
references/details.md16.4 KiB

ATTRIBUTION.md

Attribution

This folder is an unmodified copy of the Agent Skill "embedding-strategies", distributed by AutoGineer (https://autogineer.com/agent-skills).

Source: https://github.com/wshobson/agents/tree/4bde3168b43c0655c4d00ea2726e631719908a6d/plugins/llm-application-dev/skills/embedding-strategies
Commit: 4bde3168b43c0655c4d00ea2726e631719908a6d
Created by: Seth Hobson (https://github.com/wshobson/agents/tree/4bde3168b43c0655c4d00ea2726e631719908a6d/plugins/llm-application-dev/skills/embedding-strategies)
Publisher: Seth Hobson (https://github.com/wshobson)
Licence: MIT License (https://github.com/wshobson/agents/blob/4bde3168b43c0655c4d00ea2726e631719908a6d/LICENSE)
Licence evidence: repository licence LICENSE
Retrieved: 2026-11-07

The skill's own files are unmodified copies. AutoGineer is not affiliated with or endorsed by the author or the publisher.

Files added by AutoGineer

ATTRIBUTION.md: this file, written by AutoGineer
LICENSE: copied from LICENSE

Requirements and compatibility

Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta

“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.

Security

No issues found by our static scan· No issues found

A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.

Claude API
AnthropicOfficial
Comprehensive reference for integrating Anthropic's Claude API, SDKs, and Managed Agents into your applications.
Data and AIClaude CodeClaude appsOpenAI Codex+14 more
Apache-2.0Medium riskHas scriptsGitHub stars of the source repository188.6k
Transformers
K-Dense· Community
Integrate Hugging Face Transformers for model loading, inference, and fine-tuning across NLP, vision, and audio tasks.
Data and AIClaude CodeClaude appsOpenAI Codex+14 more
MITLow riskGitHub stars of the source repository56.4k
Open Notebook
K-Dense· Community
Integrate with the self-hosted Open Notebook platform to automate research ingestion, context-aware chat, and AI-driven analysis.
Data and AIClaude CodeClaude appsOpenAI Codex+14 more
MITLow riskHas scriptsGitHub stars of the source repository56.4k
Vision Sft
Seth Hobson· Community
A structured workflow for supervised fine-tuning of vision-language models, focusing on adapter configuration and silent failure prevention.
Data and AIClaude CodeClaude appsOpenAI Codex+14 more
MITGitHub stars of the source repository42.3k
LLM Evaluation
Seth Hobson· Community
A comprehensive framework for evaluating LLM application performance using automated metrics, human feedback, and LLM-as-a-judge techniques.
Data and AIClaude CodeClaude appsOpenAI Codex+14 more
MITGitHub stars of the source repository42.3k
Lora Qlora Recipes
Seth Hobson· Community
Provides configuration recipes and best-practice hyperparameters for LoRA and QLoRA supervised fine-tuning.
Data and AIClaude CodeClaude appsOpenAI Codex+14 more
MITGitHub stars of the source repository42.3k

Download ZIP

The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download embedding-strategies.zip