antongulin

Opencode Skill Creator — Testing skill for Claude Code

Testing community

OpenCode skill for creating, testing, and optimizing other OpenCode skills.

How to install Opencode Skill Creator

This entry records only its repository, not the path inside it, so there is no exact command to give. Open antongulin/opencode-skill-creator and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Opencode Skill Creator does

OpenCode skill for creating, testing, and optimizing other OpenCode skills. Adapted from Anthropic's skill-creator for Claude Code.

Alternatives in Testing

Webapp Testing — Test local web applications using Playwright for UI verification and debugging 94.1k ★
Fix Issue — by metabase - Addresses GitHub issues by taking issue number as parameter, analyzing context, implementing sol 46.5k ★
Playwright — claude-plugins-official Browser automation, E2E testing, screenshots 29.4k ★

README

opencode-skill-creator

**Create, test, and optimize OpenCode skills — from first draft to production-grade.**

[![npm](https://img.shields.io/npm/v/opencode-skill-creator)](https://www.npmjs.com/package/opencode-skill-creator) [![License: Apache 2.0](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0) [![GitHub stars](https://img.shields.io/github/stars/antongulin/opencode-skill-creator?style=social)](https://github.com/antongulin/opencode-skill-creator/stargazers)

A **skill + plugin** for [OpenCode](https://opencode.ai) that brings eval-driven development to AI agent skills — based on Anthropic's official [skill-creator](https://github.com/anthropics/skills/tree/main/skills/skill-creator) for Claude Code, ported to TypeScript and adapted for OpenCode's plugin architecture.

[Install](#install) · [What it does](#what-it-does) · [Plugin tools](#plugin-tools) · [Usage](#usage) · [Architecture](#architecture)

Why this exists

Creating AI agent skills is guesswork. You write a skill, test it manually, maybe tweak the description, and hope it triggers correctly. There's no systematic way to measure whether a skill works or to track improvements across iterations.

opencode-skill-creator fixes this with **eval-driven development for skills**:

Test — Auto-generate eval test sets and measure trigger accuracy
Optimize — Iteratively improve skill descriptions with a train/test split
Benchmark — Quantitative comparison across iterations with variance analysis
Review — Built-in visual eval viewer for human-in-the-loop feedback
Install — Deploy validated skills to project or global config

Based on Anthropic's proven methodology. Free for everyone. Works with any model OpenCode supports.