skill-up

skill-up

A tool for evaluating and improving AI agent skills.

Agent SkillDevOpen source
Type
Agent Skill
Open source
Yes
GitHub Stars
★ 1.0k
Source
skill-github

Overview

skill-up is a tool designed to evaluate and enhance AI agent capabilities. It defines evaluation environments, engines, models, and test cases through declarative YAML configuration files and supports multiple evaluation strategies. The tool generates structured reports and iteratively improves by automatically fixing or expanding test cases. Ideal for developers who need continuous optimization of AI agent skills. Integration options include installing the skill-upper plugin via platforms such as Codex or Claude Code.

Capabilities

  • ▪Declarative evaluation configuration
  • ▪Multi-engine support
  • ▪Flexible evaluation strategies
  • ▪Structured report output
  • ▪Automatic repair and expansion of test cases

Use cases

Assessing the quality of AI agent skillsAutomatically repairing and expanding test casesGenerating structured reportsSkill evaluation in continuous integration

Setup

Requires: API KeyNode 环境
npx skills add https://github.com/alibaba/skill-up/tree/main/skills/skill-upper -g -a codex -y

This information was compiled by AI from public sources and may contain inaccuracies — please refer to the source.

FAQ

How do I get started with skill-up?

By installing the skill-upper plugin and running the evaluation command.

Which engines are supported?

Supports Qoder CLI, Claude Code, Codex, and others.

Related skills