impact-analyzer

I took a look at your impact-analyzer skill and wanted to share some thoughts.

Links:

• GitHub
• Your agentic skill in the SkillzWave marketplace

The TL;DR

You're at 75/100, which is solid C-grade territory. This is based on Anthropic's skill best practices rubric. Your strongest area is Spec Compliance (13/15) — the metadata and YAML are dialed in. Weakest area is Utility (14/20) — the skill doesn't quite deliver enough concrete value or guidance for users trying to actually use it.

What's Working Well

• Tight spec compliance. Your YAML frontmatter is valid, naming conventions are correct (hyphen-case), and the description includes good trigger phrases like "what will break if I change X" — that's solid discoverability.
• Clean, concise structure. At 44 lines, there's zero fluff. Each section earns its place, which is the opposite of most skills I review.
• Good flexibility in scope. Offering impact analysis across files, APIs, components, and models gives users real degrees of freedom instead of a one-size-fits-all tool.

The Big One: Missing Sample Output

This is holding you back the most. You list output types (impact summary, risk level assessment, breaking changes, etc.) but never show what that actually looks like. A user reading your skill has no concrete picture of what they're getting.

Fix: Add a "Sample Output" section showing a real, formatted example — something like:

## Sample Output

**Target:** UserAuthService class

**Impact Summary:**
- 3 direct dependents
- 12 transitive dependents
- Affects login flow, password reset, session management

**Risk Level:** 🔴 HIGH
- Breaking changes: `authenticate()` signature change
- Migration effort: 2-3 hours per dependent

This alone could bump you +2-3 points in Utility.

Other Things Worth Fixing

1. Add a references directory. Complex scenarios like interpreting risk levels or migration patterns deserve their own docs. Create references/risk-levels.md explaining red/yellow/green criteria and references/migration-patterns.md with common strategies. (+3 points potential)

2. Error handling guidance is missing. What happens if the target isn't found? If the command fails? Add a step 4: "If target not found, suggest similar names or ask user to clarify scope." (+2 points)

3. Redundant trigger documentation. Your triggers are listed in metadata, "When to Use" section, AND "Example Triggers" section. Pick the strongest two places and consolidate — saves tokens and confusion. (+1 point)

Quick Wins

• Add concrete sample output (biggest bang for buck)
• Create references/ directory for edge cases and strategies
• Add error handling step to instructions
• Consolidate trigger documentation to eliminate overlap

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.