Does this skill support custom output dimensions?

Yes, it utilizes Matryoshka Representation Learning, allowing you to specify dimensions between 128 and 3072 to balance accuracy and storage costs.

Can I use these embeddings in serverless environments like Cloudflare Workers?

Absolutely. The skill includes specific Fetch API patterns designed for edge environments where the full Node.js SDK might not be available.

How are API rate limits handled in this skill?

The skill provides implementation patterns for batching requests and utilizing delay mechanisms to stay within the 100 RPM limit of the Google AI Studio free tier.

What is the maximum token limit for Gemini embeddings?

The gemini-embedding-001 model has an input limit of 2,048 tokens per individual text snippet.

Which task type should I use for a RAG system?

You should use RETRIEVAL_QUERY for the user's search input and RETRIEVAL_DOCUMENT for the content you are indexing in your vector database.

Google Gemini Embeddings

Name: Google Gemini Embeddings
Author: jezweb

byjezweb

•

117

•

数据科学与机器学习

Implements high-performance text embeddings for RAG, semantic search, and document clustering using the Google Gemini API.

This skill provides a complete production-ready framework for integrating Google's gemini-embedding-001 model into AI applications. It enables developers to generate high-dimensional vectors (ranging from 768 to 3072) for advanced retrieval-augmented generation (RAG) systems and semantic similarity tasks. With built-in support for Cloudflare Vectorize, optimized task types, and efficient batch processing, it simplifies the complexity of vector-based data modeling while ensuring performance through Matryoshka Representation Learning.

主要功能

01Seamless integration patterns for Cloudflare Vectorize and Edge environments

02Task-specific optimization for retrieval, clustering, and fact verification

03117 GitHub stars

04High-efficiency batch processing with rate-limit and token management

05Production-ready error handling for dimension mismatches and API limits

06Full support for gemini-embedding-001 with flexible dimensionality (128-3072)

使用场景

01Building robust RAG systems for custom knowledge bases and documentation

02Automating document clustering and topic modeling for large-scale text analysis

03Developing semantic search engines that understand intent beyond keyword matching

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add jezweb/claude-skills google-gemini-embeddings

For use in Claude.ai and ChatGPT

Download Skill