Skip to content

Feedback on your spanish-cooking skill #12

@RichardHightower

Description

@RichardHightower

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

Links:

The TL;DR

You're at 87/100, which puts you solidly in B territory — good fundamentals with real polish in your architecture. This is based on Anthropic's best practices for skill design. Your strongest area is Progressive Disclosure Architecture (28/30) — you've nailed the token efficiency and layering. The weakest spot is Spec Compliance (11/15), mainly because your description is missing trigger phrases that help Claude know when to use this skill.

What's Working Well

  • Clean module structure — Your _toc.md navigation is crystal clear, and you've kept the main SKILL.md concise while stashing the detailed recipes and techniques in separate modules. That's exactly how PDA should work.
  • Solid utility — The research checklist and example responses with input/output pairs show you actually thought about how someone would use this. The comprehensive tapas-to-paella range gives real practical value.
  • Consistent voice — Your writing stays instructional and objective throughout. No marketing fluff, just straightforward guidance on Spanish cooking techniques.

The Big One: Legacy Status & Low Confidence

This is holding you back more than anything else. Your skill is marked as "⚠️ Legacy template awaiting research upgrade" with low confidence. That's a 5-point hit right there, and it signals to users that you haven't validated your sources recently.

Here's the fix: Work through your modules/research-checklist.md systematically. Capture authoritative sources with 2024-2025 dates (not just legacy recipes). Update the status to at least medium confidence once you've verified key techniques like proper paella timing, regional tapas variations, or modern Spanish cooking trends. This alone could push you to 92/100.

Other Things Worth Fixing

  1. Add trigger phrases to your description — Right now it's "Comprehensive expertise in Spanish cuisine..." but it needs action words. Change it to: "Performs spanish cooking operations. Use when asked to 'spanish cooking', 'run spanish cooking', 'spanish recipe help', or 'paella techniques'." That's an easy +2 points.

  2. Name convention mismatch — Your folder is spanish-cooking but your skill name in frontmatter doesn't match cleanly. Align these for better discoverability (+1 point).

  3. Workflow clarity in research-checklist.md — The research checklist is good, but the workflow for actually upgrading confidence could be more explicit. Walk through: (1) Run checklist, (2) Verify 3+ sources, (3) Update status field. Right now it feels a bit abstract.

Quick Wins

  • Add trigger phrases to frontmatter description (+2 points)
  • Run through research checklist and upgrade from "legacy" to "medium confidence" (+5 points)
  • Capture fresh 2024-2025 sources in known-gaps module (+3 points)
  • Fix name convention alignment (+1 point)

These four moves could get you to 98/100 — basically production-ready.


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.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions