echo

Perform text embeddings, semantic search, and generation with local Ollama models.

1|Updated Apr 11, 2026
One-click install
npx skills add https://github.com/AIWander/echo --skill echo-aiwander
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: echo
Source: https://github.com/AIWander/echo/tree/main/skills/echo
Command: npx skills add https://github.com/AIWander/echo --skill echo-aiwander

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

echo solves the problem of needing a local AI toolkit that is not tied to a specific model, allowing for the use of any local model for tasks like embeddings, semantic search, generation, summarization, pattern matching, and HTML fetch+summarize.

Core Features & Use Cases

  • Model-Agnostic: Supports any local Ollama model.
  • Local-first: All processing happens on the user's machine, ensuring data privacy.
  • Features: Embeddings, semantic search, generation, fetching, pattern matching, heuristics, scoring, error learning, and system health checks.
  • Use Case: Use echo to summarize web pages, analyze text for insights, or build a local similarity classifier without the need for cloud services.

Quick Start

Start by checking the available Ollama models and their health with 'echo:health'.

Frequently Asked Questions about echo

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run semantic search and text analysis locally without sending data to the cloud?

Local-first text analysis processes embeddings and semantic search entirely on your machine. By utilizing local Ollama models, it ensures data privacy while performing pattern matching and generation tasks without relying on cloud services.

Can I use any local Ollama model for generating embeddings and summarizing text?

Yes, this AI toolkit is model-agnostic and supports any local Ollama model. You must specify the desired Ollama model as a parameter for each text generation, embedding, or summarization task to execute the processing locally.

What is the best way to analyze web pages and extract insights locally using Claude Desktop?

Fetching and summarizing HTML content locally is the best way to analyze web pages. The toolkit fetches web data and uses local Ollama models to summarize the content, extract insights, and perform pattern matching directly within Claude Desktop.

Do I need to install specific dependencies or cloud services to build a local similarity classifier?

No external cloud services or specific dependencies are required. You can build a local similarity classifier by leveraging local Ollama models for embeddings and semantic search, keeping all data processing strictly on your machine.

How do I check if my local Ollama models are available and ready for text processing tasks?

You can perform a system health check to verify available Ollama models. Running the health check command assesses model readiness and system status before executing any embeddings, generation, or semantic search tasks.

Why does local text generation require passing the model name as a parameter for each task?

Passing the model name as a parameter is required because the toolkit is model-agnostic. This design ensures the system knows exactly which local Ollama model to route your specific text analysis, embedding, or generation task to.