Private

What problem does it solve?

Route privacy-sensitive prompts to a local model so confidential data never leaves your machine or network, preventing accidental exposure to cloud APIs or third-party services.

Core Features & Use Cases

• Local Ollama routing: Sends user prompts to Gemma4 31B or other Ollama-hosted models on localhost:11434.
• Model and prompt control: Supports explicit model selection, custom system prompts, temperature tuning, and streaming or non-streaming responses.
• Privacy-first use cases: Ideal for financial projections, medical report analysis, HR reviews, proprietary code review, and any HIPAA/PCI-sensitive interactions.

Quick Start

Invoke the Private skill by prefixing your message with /private followed by your prompt, for example: /private Analyze the attached financial projections for client XYZ.