smith-guidance

Community

Ethical, structured steering for AI agents.

Authortianjianjiang
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Coordinating AI agent interactions in a safe, effective, and auditable manner requires a clear governance framework. This Skill provides a core steering approach built on the HHH framework (Helpful, Honest, Harmless), an exploration-before-implementation workflow, and anti-sycophancy rules to guide behavior across tasks and disagreements.

Core Features & Use Cases

  • HHH framework: Helps ensure Helpful, Honest, and Harmless agent interactions.
  • Exploration-Before-Implementation: Promotes reading, clarifying, proposing, and implementing with guardrails.
  • Anti-Sycophancy: Encourages evidence-based dissent and proactive risk mitigation.

Quick Start

Apply the HHH-guided exploration-before-implementation workflow to steer all agent interactions.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: smith-guidance
Download link: https://github.com/tianjianjiang/smith/archive/main.zip#smith-guidance

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 223,000+ vetted skills library on demand.