How do I reduce the size of the generated vectors?

You can use the output_dimensionality parameter to truncate embeddings to a smaller size (e.g., 256), which saves storage and improves search speed.

Which model does this skill use for embeddings?

This skill utilizes the gemini-embedding-001 model via the Google Generative Language API for reliable and high-dimensional text representations.

What are Task Types in Gemini Embeddings?

Task types like RETRIEVAL_DOCUMENT or SEMANTIC_SIMILARITY tell the model how the vectors will be used, optimizing the mathematical representation for that specific use case.

What authentication is required?

You must have a valid Google AI Studio API key set in your environment as GOOGLE_API_KEY to use this skill.

Can I process multiple documents at once?

Yes, the skill supports batch embedding calls, allowing you to convert multiple text strings into vectors in a single, efficient API request.

Gemini Embeddings Integration

Name: Gemini Embeddings Integration
Author: legacybridge-tech

bylegacybridge-tech

•

データサイエンスとML

Generates high-performance text embeddings via the Google Gemini API for semantic search, RAG, and data classification tasks.

This skill empowers Claude to interact directly with the Google Gemini Embedding API, enabling the conversion of raw text into dense vector representations. By supporting specific task types like retrieval query, document indexing, and semantic similarity, it allows for more accurate search results and efficient data processing within AI workflows. It is particularly useful for developers building Retrieval-Augmented Generation (RAG) systems, recommendation engines, or automated categorization tools where high-quality, context-aware text representation is critical for performance.

主な機能

01Optimized workflow for semantic similarity and clustering

02Single and batch text embedding generation via REST

03Adjustable output dimensionality for vector storage optimization

043 GitHub stars

05Support for specialized task types like RETRIEVAL_QUERY and CLASSIFICATION

06Direct integration with the gemini-embedding-001 model

ユースケース

01Building RAG pipelines to enhance Claude's knowledge with local documents

02Automating document classification and grouping for large datasets

03Developing semantic search engines that understand user intent beyond keywords

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add legacybridge-tech/claude-plugins gemini-embed

For use in Claude.ai and ChatGPT

Download Skill