rebasing-branch

I took a look at your rebasing-branch skill and wanted to share some thoughts.

Links:

• GitHub
• Your agentic skill in the SkillzWave marketplace

The TL;DR

You're at 64/100, which puts you in D territory (needs work). The grading is based on Anthropic's best practices for agentic skills. Your Ease of Use is pretty strong at 22/25—the metadata and navigation are solid—but Spec Compliance is completely missing at 0/15, which is a massive anchor dragging everything down.

What's Working Well

• Clear phase structure - Your 8 numbered phases with headers and objectives make it easy to follow the workflow, even in a 318-line file
• Solid trigger definitions - "rebase branch", "rebase on main", "rebase onto" are specific and discoverablePHASE, good scope boundaries vs. other skills
• Comprehensive error handling - You're thinking through edge cases like mainline protection and state preservation, which shows you understand real Git complexity
• Strong validation gates - Each phase has clear success/failure criteria with explicit next steps

The Big One: Invalid Frontmatter

Here's the thing—your frontmatter isn't parsing, which is why you're getting zero points on Spec Compliance. This isn't just a formatting issue; it breaks discoverability in skill registries.

Your current frontmatter is missing the closing --- and has a malformed description field:

---
name: rebasing-branch
description: Standard workflow for all rebase operations...

Here's what it should look like:

---
name: rebasing-branch
description: Performs rebasing branch operations. Use when asked to "rebasing branch", "run rebasing branch", or similar rebase requests.
---

This one fix gets you +5 points immediately. But there's more—your description needs to include actual trigger phrases like "rebasing branch help" or "rebase my feature" so agents can recognize when to invoke you. Right now it's descriptive but doesn't give trigger clarity.

Other Things Worth Fixing

1. Extract error handling to a reference file - You've got 25+ lines of repeated error patterns across phases. Move that to references/common-errors.md and you'll shed weight while improving your Progressive Disclosure Architecture score. Currently 22/30, this could push you to 25+.

2. Add concrete input/output examples - Your "When to Use" section lists triggers but doesn't show an actual example like: Input: "rebase my feature branch onto develop" → Extracts: target=develop, preserve_dates=false. This would bump your Utility score (currently 17/20).

3. Tighten validation gate formatting - Phase 1's IF/ELSE blocks are a bit verbose. A simple table (| Condition | Action | Message |) would be cleaner and save ~8 lines. Small win on Writing Style (8/10).

Quick Wins

• Fix frontmatter (+5 points) - Validate YAML, add trigger phrases to description
• Extract error patterns (+3 points) - Move to references/ directory
• Add input examples (+1-2 points) - Show what the skill looks like in action
• Total potential: Get to 75-80 with these three changes

Checkout your skill here: SkillzWave.ai | SpillWave We have an agentic skill installer that install skills in 14+ coding agent platforms. Check out this guide on how to improve your agentic skills.