How do I resolve a dimension mismatch error?

The skill includes a troubleshooting mode that queries your embedding provider directly to detect actual dimensions and compares them against your config, offering a one-click fix.

Why do I need to re-index after changing models?

Embedding models produce vectors with specific dimensions and semantic mappings. Because these mappings are unique to each model, old index data is mathematically incompatible with a new model and must be rebuilt.

Does this skill support local-first development?

Yes, it provides deep integration with local providers like Ollama and LM Studio, including the ability to pull models via Docker or validate local endpoints.

Can I use different models for different project folders?

Yes, the skill detects GrepAI workspace mode and allows you to apply unique provider and model settings to specific workspaces without affecting your global configuration.

GrepAI Embedder Config

Name: GrepAI Embedder Config
Author: jugrajsingh

byjugrajsingh

0•

Data Science & ML

Configures and troubleshoots embedding providers and models for the GrepAI semantic search tool.

GrepAI Embedder Config is a specialized Claude Code skill designed to manage the technical lifecycle of embedding configurations within the GrepAI ecosystem. It simplifies the complex process of switching between local providers like Ollama and LM Studio or cloud services like OpenAI by automatically handling vector dimensions, API endpoints, and workspace-specific settings. The skill is particularly useful for developers who need to optimize search performance, manage API costs, or resolve synchronization issues, as it automates the critical re-indexing workflows required whenever an underlying model is changed.

Key Features

01Multi-provider support for Ollama, OpenAI, and LM Studio

02Diagnostic troubleshooting for connectivity and dimension mismatches

03Interactive model comparison with speed, quality, and pricing references

04Automatic detection of local vs. workspace configuration modes

05Automated re-indexing workflows to prevent vector incompatibility

060 GitHub stars

Use Cases

01Configuring specific embedding models for isolated project workspaces

02Troubleshooting indexing failures caused by dimension mismatches in config files

03Switching from local Ollama embeddings to OpenAI for higher search accuracy

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add jugrajsingh/skillgarden embedder-config

For use in Claude.ai and ChatGPT

Download Skill