taste

Evaluate artifacts against eight cross-domain anchors in audit or anchor mode.

5|2|Updated Nov 17, 2025
One-click install
npx skills add https://github.com/OutlineDriven/odin-gemini-cli-extension --skill taste-outlinedriven
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: taste
Source: https://github.com/OutlineDriven/odin-gemini-cli-extension/tree/main/skills/taste
Command: npx skills add https://github.com/OutlineDriven/odin-gemini-cli-extension --skill taste-outlinedriven

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill helps teams enforce deliberate, domain-spanning judgment on artifacts instead of defaulting to centroid-AI guidance.

Core Features & Use Cases

  • Modes: audit mode performs multi-anchor evaluation and reports top fixes; anchor mode loads the charter/anchors into context for consistent responses.
  • Cross-domain anchors: applies eight portable anchors (Clarity, Hierarchy, Intent, Coherence, Restraint, Generosity, Honesty, One strong moment) across prose, code, and decisions.
  • Persistent register: loads and maintains a registered set of imperatives until changed or reset, enabling consistent responses across interactions.

Quick Start

Provide a concise, step-by-step prompt to audit an artifact using the anchors and generate top-3 fixes.

Frequently Asked Questions about taste

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

FAQPage Schema
How do I evaluate artifacts to avoid default AI convergence?

Cross-domain artifact evaluation applies eight portable anchors—Clarity, Hierarchy, Intent, Coherence, Restraint, Generosity, Honesty, and One strong moment—across prose, code, design, and decisions to enforce deliberate, domain-spanning judgment.

How do I audit an artifact and generate top fixes?

To audit an artifact, provide a concise prompt to trigger audit mode. The system evaluates the artifact against the eight anchors and generates the top-3 fixes to improve clarity, hierarchy, intent, coherence, restraint, generosity, honesty, and strong moments.

Can I maintain consistent evaluation criteria across multiple interactions?

You can maintain consistent evaluation criteria using a persistent register. It loads and maintains a registered set of imperatives until changed or reset, ensuring consistent cross-domain artifact judgment across multiple interactions.

Does cross-domain artifact evaluation work for both prose and code?

Yes, cross-domain artifact evaluation works for prose, code, design, and decisions. It applies eight portable anchors to each artifact type, enabling deterministic evaluation and avoiding centroid-AI default convergence in dynamic workflows.

What is anchor mode and when should I use it?

Anchor mode loads the two-sided charter and eight anchors into context for consistent responses. Use anchor mode when you need to apply deliberate, domain-spanning judgment to artifacts during dynamic workflows instead of auditing them for top fixes.