Skill Creator

syahiidkamil/software-engineer-ai-agent-atlas

This is meta tooling for building and refining skills themselves. It walks you through the full cycle: drafting a skill from a rough idea, generating test prompts, running quantitative evals with variance analysis, reviewing results both qualitatively and quantitatively, then iterating based on what breaks. It can also optimize skill descriptions to improve triggering accuracy, which matters since Claude tends to undertrigger skills. The approach is pragmatic about user technical level and flexible about process. If you want formal benchmarking, it'll set that up. If you just want to vibe and iterate, it does that too. Essentially a skill workshop that handles the tedious parts of the create, test, measure, revise loop.

GitHub stars: 298
last updated: Jun 11, 2026

View on GitHub

Install

$ npx -y skills add syahiidkamil/software-engineer-ai-agent-atlas --skill skill-creator --agent claude-code

Installs into .claude/skills of the current project.

Files

Skill Creator

A skill for creating new skills and iteratively improving them.

At a high level, the process of creating a skill goes like this:

Decide what you want the skill to do and roughly how it should do it
Write a draft of the skill
Create a few test prompts and run claude-with-access-to-the-skill on them
Help the user evaluate the results both qualitatively and quantitatively
- While the runs happen in the background, draft some quantitative evals if there aren't any (if there are some, you can either use as is or modify if you feel something needs to change about them). Then explain them to the user (or if they already existed, explain the ones that already exist)
- Use the eval-viewer/generate_review.py script to show the user the results for them to look at, and also let them look at the quantitative metrics
Rewrite the skill based on feedback from the user's evaluation of the results (and also if there are any glaring flaws that become apparent from the quantitative benchmarks)
Repeat until you're satisfied
Expand the test set and try again at larger scale