spanish-language-tutor

I took a look at your spanish-language-tutor skill and wanted to share some thoughts.

Links:

• GitHub
• Your skill in the SkillzWave marketplace

The TL;DR

You're at 88/100, solid B-grade territory. This is based on Anthropic's skill best practices rubric. Your strongest area is Ease of Use (24/25) – the workflow and navigation are really clean. The biggest drag is unvalidated legacy content that's tanking your Utility score (16/20) and creating uncertainty about what actually works.

What's Working Well

• Progressive Disclosure nailed it – Your modular structure with _toc.md and layered modules is exactly how PDA should work. Clear hierarchy without burying users in detail upfront.
• Workflow clarity is chef's kiss – The research checklist with numbered steps and validation checkpoints in modules/research-checklist.md is how you actually help users avoid bad info.
• Metadata and discoverability – Name follows conventions, trigger terms are explicit in the core-guidance module, terminology stays consistent across files.

The Big One: Unvalidated Legacy Content

This is your main bottleneck. Your skill is marked with:

⚠️ Legacy template awaiting research upgrade

This creates a trust problem. Users can't tell which sections have been validated and which haven't. You're leaving 5+ points on the table here.

The fix: Complete the research checklist you've already built. Go through each module (grammar, conversation, dialects, etc.), validate the examples against current Spanish standards, and mark confidence levels. Update the status from "low confidence" to "medium" or "high" with specific notes on what's been validated. This isn't about rewriting – it's about auditing what you have.

Other Things Worth Fixing

1. Add trigger phrases to the description – Currently reads as generic. Change from "Comprehensive Spanish language expert..." to include explicit triggers like: "Use when asked to 'spanish language tutor', 'teach me spanish grammar', or 'spanish conversation practice'." That's +2 points on Spec Compliance and helps discoverability.

2. Zero examples or input/output pairs – You're at 0/3 on this criterion. Add one concrete example: show a user prompt ("Explain the subjunctive mood") and a sample response. Makes the skill immediately tangible.

3. Voice consistency – You've got mostly imperative instructions but slip into "You are an expert" phrasing in places. Pick one voice and stick with it throughout.

Quick Wins

• Complete legacy validation → +5 points, kills the biggest red flag
• Add trigger phrases → +2 points, easier discovery
• Add 1-2 examples → +3 points, proves utility
• Standardize voice → +1-2 points, cleaner writing

Bang for your buck: Hit the validation checklist first (biggest impact), add trigger phrases (quick win), then examples. That gets you into A territory pretty easily.

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.