skills.rest
Peleke
PelekeCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

ax-rubric

Score and rewrite tool descriptions against the AX Description Rubric.

Updated Mar 7, 2026
Search Tags:#documentation#prompt engineering#developer experience#ai agent#rubric#tool description
One-click install
npx skills add https://github.com/Peleke/hunter --skill ax-rubric
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ax-rubric
Source: https://github.com/Peleke/hunter/tree/main/skills/ax-rubric
Command: npx skills add https://github.com/Peleke/hunter --skill ax-rubric

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill ensures that tool descriptions are clear, concise, and informative for AI agents, preventing tools from being overlooked due to poor descriptions.

Core Features & Use Cases

  • Automated Scoring: Evaluates tool descriptions against a five-criterion rubric (Output shape, Cost signal, Trigger clarity, Specificity, Differentiation).
  • Failure Analysis: Identifies specific shortcomings in descriptions and explains their impact on agent discoverability.
  • Automated Rewriting: Generates improved descriptions that meet the 5/5 scoring criteria, adhering to principles like leading with the problem and stating costs.
  • Use Case: When developing new tools for an AI agent, use this Skill to review and refine their descriptions before deployment, ensuring they are easily discoverable and usable by the agent.

Quick Start

Use the ax-rubric skill to score and rewrite the following tool description: "This tool helps you find information."

Frequently Asked Questions about ax-rubric

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

FAQPage Schema
How do I write better tool descriptions for AI agents?▼

Tool descriptions for AI agents are evaluated against a five-criterion rubric covering output shape, cost signal, trigger clarity, specificity, and differentiation, then rewritten to maximize discoverability and usability.

Why does my AI agent overlook or fail to call external tools?▼

AI agents overlook external tools when descriptions lack clear trigger conditions, cost signals, or output shapes, causing poor discoverability. Scoring descriptions against a structured rubric identifies these shortcomings and generates improved rewrites.

What is the best way to score tool descriptions for prompt engineering?▼

Scoring tool descriptions for prompt engineering involves analyzing them for output shape, cost signal, trigger clarity, specificity, and differentiation, providing actionable scores and rewrites based on a structured rubric.

How do I rewrite API descriptions to improve developer experience for agents?▼

Rewriting API descriptions to improve agent developer experience involves leading with the problem, stating costs clearly, and ensuring explicit output shapes. Automated rewriting generates descriptions meeting 5/5 scoring criteria for agent interaction.

What makes a good tool description for an AI agent?▼

A good tool description for an AI agent leads with the problem, states costs, and ensures high specificity and differentiation. Scoring against a five-criterion rubric verifies the description meets 5/5 criteria for optimal discoverability.

Related Skills

Explore More Software Engineering Skills→
H2OSLabsH2OSLabs

neuron-analyze-api

Analyze API documentation formats and output structured JSON for neuron-ui page generation.

Official
Advanced
jax2730jax2730

tool-design

Design agent tool APIs with clear descriptions and unambiguous parameters.

Community
Advanced
benmaierbenmaier

Create data science agents

Design and validate data-science agent prompts and parallel-agent configurations.

Community
Advanced
slgoodrichslgoodrich
121

usability-frameworks

Plan and evaluate usability testing for user interfaces using Nielsen's heuristics.

Community
Basic