changshu-arnold

Mimic Changshu Arnold's speech patterns using 27 language rules and a 6-dimensional cognitive model.

7|Updated Apr 15, 2026
One-click install
npx skills add https://github.com/factnn/changshu-arnold-skill --skill changshu-arnold
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: changshu-arnold
Source: https://github.com/factnn/changshu-arnold-skill/tree/main
Command: npx skills add https://github.com/factnn/changshu-arnold-skill --skill changshu-arnold

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires requests, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill solves the challenge of generating authentic Changshu Arnold-style responses without manual prompt engineering. It provides a comprehensive linguistic and cognitive model distilled from 269 livestream transcripts, enabling Claude to mimic Arnold's unique speech patterns, contradictions, catchphrases, and behavioral quirks for entertainment and educational purposes.

Core Features & Use Cases

  • Three-Tier Priority Output: Automatically selects between P0 quote replication for triggered memes, P1 knowledge-driven responses for biographical questions, and P2 autonomous style generation for casual chat.
  • Linguistic Pattern Engine: Implements 27 language rules including confirmation bias catchphrases, 30-second self-contradictions, typo generation systems, and semantic drift patterns.
  • Cognitive Modeling: Uses a 6-dimensional cognitive model to drive authentic speech patterns, emotional triggers, and behavioral responses.
  • Use Case: When a user asks about "head shape" or mentions "sugar-free cola", the Skill instantly replicates Arnold's original livestream quotes without rewriting or summarizing.

Quick Start

Activate the changshu-arnold skill by mentioning any trigger word like "常熟阿诺", "诺神", "诺言诺语", "那我问你", "三卡车", "生米冲碳", "申气", "脑袋尖尖", or "无糖可乐" in your conversation with Claude to instantly transform the AI into Arnold's speaking style.

Frequently Asked Questions about changshu-arnold

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

FAQPage Schema
How do I make Claude roleplay a specific persona using livestream transcripts?

To make Claude roleplay a specific persona, this Skill applies a distilled linguistic model from 269 livestream transcripts, integrating 27 language rules and a 6-dimensional cognitive model to generate authentic speech patterns and behavioral quirks.

What are trigger words to activate persona impersonation in Claude?

Persona impersonation activates instantly when you mention trigger words like "常熟阿诺", "诺神", "那我问你", "脑袋尖尖", or "无糖可乐" in your conversation, transforming Claude into the target persona without manual prompt engineering.

How does keyword-triggered quote replication work for entertainment chatbots?

Keyword-triggered quote replication works through a three-tier priority output architecture where P0 instantly replicates original livestream quotes when specific meme keywords are detected, avoiding rewriting or summarizing for authentic entertainment responses.

Can I use linguistic modeling for educational scenarios requiring authentic mimicry?

Yes, linguistic modeling supports educational scenarios requiring authentic mimicry by applying a cognitive model that drives behavioral responses, semantic drift patterns, and confirmation bias catchphrases distilled from transcript data.

What linguistic patterns are needed to impersonate a cyber persona in AI chat?

Impersonating a cyber persona requires 27 language rules including typo generation systems, 30-second self-contradictions, and semantic drift patterns, combined with a 6-dimensional cognitive model to drive authentic speech and emotional triggers.

Does persona roleplay require manual prompt engineering for each conversation?

No, persona roleplay eliminates manual prompt engineering by automatically applying a pre-distilled linguistic and cognitive model, automatically selecting between quote replication, knowledge-driven responses, and autonomous style generation based on context.