hook-factory

I was curious about how you're structuring the hook lifecycle—managing the event emission, dependency resolution, and cleanup in a single factory pattern feels like it could get complex fast. How are you handling cases where hooks need to fire in a specific order or have interdependencies?

Links:

• GitHub
• Your agentic skill in the SkillzWave marketplace

The TL;DR

You're at 69/100, solidly in D territory. This is based on Anthropic's skill quality rubric—think of it as how well your skill balances token efficiency, clarity, and practical value. Your strongest area is Ease of Use (19/25)—the metadata and workflow structure are solid. The biggest drag is Progressive Disclosure Architecture (20/30)—you've got a 462-line SKILL.md that's trying to do too much in one place.

What's Working Well

• Clear modes and templates: Three distinct usage patterns (auto-format, test-runner, git-auto-add) with 10 templates give users concrete options to work with
• Solid metadata structure: Your frontmatter is valid, name conventions are correct, and the skill is technically discoverable
• Practical examples: You've included actual Q&A interactions showing how the skill works in practice, which helps users understand the flow

The Big One: Missing Reference Architecture

Here's what's holding you back the most: Your SKILL.md claims an examples/ directory with 10 reference folders (auto-format-python/, git-auto-add/, test-runner/, etc.), but these files don't actually exist. This breaks Progressive Disclosure—you're promising layered reference material that isn't there, and it costs you 8 points.

Why it matters: A 462-line monolithic file is hard to navigate and token-expensive. Progressive Disclosure means building a pyramid: th...