What problem does it solve?
Creating effective AI skills requires more than just writing prompts—it demands systematic testing, measurement, and iterative refinement to ensure reliable performance across diverse user queries. This skill eliminates the guesswork by providing a complete workflow for skill development, evaluation, and optimization.
Core Features & Use Cases
- End-to-End Skill Creation: Guides you from capturing user intent to packaging a distributable .skill file, including draft writing, test case generation, and iterative improvement.
- Evaluation Harness: Spawns parallel test runs with and without the skill, grades outputs against assertions, and aggregates results into benchmark statistics with variance analysis.
- Description Optimization: Runs an automated loop that tests trigger accuracy across realistic queries and iteratively improves the skill's description for better GLM invocation.
- Use Case: A developer building a PDF processing skill can use this to draft the skill, create eval cases for invoice extraction, run benchmarks comparing with/without the skill, and optimize the description so it triggers correctly when users mention "invoices" or "PDF forms".
Quick Start
Use the skill-creator skill to build a new skill by telling it what you want the skill to do, then let it guide you through drafting test cases, running evaluations, and iterating on improvements.
