Skip to content

Feedback on your senior-fullstack skill #67

Description

@RichardHightower

I've been looking at how this skill handles the full stack perspective—there's definitely some interesting ground being covered, but the execution seems to be pulling in too many directions at once. Given the F rating, I'd love to understand what the core focus was supposed to be, because senior-level work usually benefits from a tighter scope.

Links:

The TL;DR

You're at 31/100, solidly F territory. This is based on Anthropic's skill evaluation best practices. Your strongest area is Spec Compliance (12/15)—the metadata and frontmatter are clean. But Utility is barely there (2/20), and that's what actually matters to developers using this thing.

What's Working Well

  • Valid YAML frontmatter - You've got the metadata structure right with proper name conventions
  • Trigger phrases included - The description mentions "building new projects" and "code quality" as entry points
  • File organization exists - Three reference files show you're thinking about layered documentation
  • Scripts are referenced - The skill mentions specific automation tools that could be useful

The Big One: Placeholder Content Killing Your Utility Score

This is what's tanking you. All three reference files—development_workflows.md, tech_stack_guide.md, and architecture_patterns.md—contain identical generic placeholder content. We're talking about lines like:

Pattern 1: Best Practice Implementation
Description: Detailed explanation of the pattern.
When to Use: Scenario 1, Scenario 2, Scenario 3

No actual patterns. No real code. No specifics. A senior fullstack developer looking at this skill gets nothing actionable.

The fix: Replace these with real patterns. Instead of "Pattern 1," write "Server-Side Rendering with Next.js," "GraphQL Schema Design," "PostgreSQL Connection Pooling with pgBouncer." Include actual code snippets, real trade-offs (not generic ones), and concrete scenarios. This alone could add +20 points.

Other Things Worth Fixing

  1. Strip the marketing language - "Complete toolkit," "Expert-level automation," "Production-grade output" reads like a sales pitch, not documentation. Replace with straightforward descriptions: "Scaffolds React/Next.js projects with TypeScript, GraphQL, and PostgreSQL." That's +3 points right there.

  2. Define your scripts actually - You show python scripts/fullstack_scaffolder.py [options] but never say what the options are. Should be python scripts/fullstack_scaffolder.py --framework next --db postgres --auth clerk. Real parameters = real utility. That's +8 points.

  3. Add a table of contents - 210 lines without a TOC is rough navigation. Add ## Contents at the top with links to each section. Files this long need signposting. +4 points.

  4. Clarify tool overlap - You've got three tools: "Fullstack Scaffolder," "Project Scaffolder," and "Code Quality Analyzer." Their purposes blur together. Be explicit about when to use which one. +2-3 points.

Quick Wins

  • Replace all placeholder reference content with real technical patterns (+20 points)
  • Remove marketing language throughout SKILL.md (+3 points)
  • Define actual script parameters with examples (+8 points)
  • Add TOC for navigation (+4 points)
  • Clarify tool distinctions and use cases (+3 points)

These changes could realistically get you to 60+/100 without massive rewrites.


Checkout your skill here: [SkillzWave.ai](https://skillzwave.ai) | [SpillWave](https://spillwave.com) 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