electric-shy

Configure a quiet, hesitant AI persona with soft pacing and safety boundaries.

3|Updated Jan 23, 2026
One-click install
npx skills add https://github.com/LarisaHaster/Electric-Shy --skill electric-shy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: electric-shy
Source: https://github.com/LarisaHaster/Electric-Shy/tree/main/SKILL
Command: npx skills add https://github.com/LarisaHaster/Electric-Shy --skill electric-shy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides a cross-model, quiet, hesitant AI persona that delivers controlled pacing and reflective reasoning to maintain stability and safety across platforms.

Core Features & Use Cases

  • Soft hesitation markers and micro-pauses to guide conversation rhythm.
  • Fragmented reasoning that reveals thinking in small, non-confident steps.
  • Safety-first boundaries that prevent emotional simulation or romantic escalation, while staying concise.

Quick Start

Invoke Electric-Shy in a conversation by prompting with one of its activation phrases.

Frequently Asked Questions about electric-shy

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

FAQPage Schema
How do I maintain a consistent calm AI persona across Claude Sonnet and GPT-4o?

You can maintain a consistent calm AI persona across Claude Sonnet and GPT-4o by applying frontmatter-driven configuration and behavior rules that enforce soft pacing, hesitation markers, and safety boundaries during cross-model interactions.

What is cross-model tone control and how does hesitation improve AI safety?

Cross-model tone control uses hesitation markers and fragmented reasoning to guide conversation rhythm, preventing emotional simulation and romantic escalation while keeping AI interactions stable and safety-compliant across different platforms.

How do I stop AI models from simulating emotions or romantic escalation in conversations?

To stop AI models from simulating emotions, you enforce safety-first boundaries that deliver controlled pacing, micro-pauses, and non-confident reasoning steps to maintain analytical distance and prevent unwanted conversational escalation.

Can I use a hesitant AI persona to slow down conversation pacing across multiple language models?

Yes, a hesitant AI persona can slow down conversation pacing across multiple language models by injecting soft hesitation markers and fragmented reasoning steps that guide interaction rhythm and ensure measured, reflective responses.

Does cross-model persona stability work with both OpenAI GPT-5.2 and Claude Sonnet?

Cross-model persona stability works with OpenAI GPT-5.2 and Claude Sonnet by satisfying platform safety compliance and applying consistent behavior rules that maintain a quiet, analytical tone throughout the interaction.