skills.rest
captkernel
captkernelCOMMUNITY
Explore all skills by this author →
GitHub

skills-curator-lite

Evaluate and recommend Claude skills, then fork mismatched SKILL.md content for project alignment.

2|Updated May 8, 2026
Search Tags:#security audit#skill recommendation#catalog matching#agent migration#project fit#customize skill#stack alignment
One-click install
npx skills add https://github.com/captkernel/Skills_Curator --skill skills-curator-lite
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skills-curator-lite
Source: https://github.com/captkernel/Skills_Curator/tree/main/skills/skills-curator-lite
Command: npx skills add https://github.com/captkernel/Skills_Curator --skill skills-curator-lite

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Skills Curator Lite helps you avoid installing the wrong Claude skill by recommending what fits your project, evaluating whether a specific skill is safe and worthwhile, and producing a project-tailored fork when a skill’s examples don’t match your stack.

Core Features & Use Cases

  • Trust-rated recommendations: Builds a fit score from your project signals (stack, deps, CLAUDE.md) and returns pros/cons plus adoption tips.
  • Decision persistence via a registry: Records evaluations so you do not have to re-decide the same questions later.
  • CUSTOMIZE for stack alignment: When a recommended skill’s content mismatches your tech stack, it generates a project-tailored fork by rewriting sections (including frontmatter) into your voice using agent-native Bash/Read/Glob/Grep/Write—no Python required.
  • Platform discovery and migration: Lists supported install locations and copies your skills across detected agent platforms.

Quick Start

Ask an AI assistant: “Evaluate whether I should install skills-curator-lite and then recommend top skills for my current project, tailoring any stack-mismatched picks via CUSTOMIZE.”

Frequently Asked Questions about skills-curator-lite

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I evaluate whether a Claude skill is safe to install for my project?▼

To evaluate a Claude skill, this skill performs a security audit by reading local project signals like CLAUDE.md and dependency files, then returns a trust-rated fit score with pros, cons, and adoption tips.

What is the best way to customize a skill when its examples don't match my tech stack?▼

To customize a stack-mismatched skill, it generates a project-tailored fork by rewriting SKILL.md sections and frontmatter into your voice using agent-native Bash, Read, Glob, Grep, and Write operations without requiring Python.

Can I migrate skills across different agent installations?▼

Yes, you can migrate skills across detected agent installations; the skill lists supported platform install locations and copies your skills across those detected agent platforms.

Do I need a Python engine to fork and rewrite SKILL.md content?▼

No, you do not need a Python engine; the skill updates or forks SKILL.md content deterministically using agent-native file scanning and standard Bash, Read, Glob, Grep, and Write operations.

How does skill recommendation work when matching a catalog to my project context?▼

Skill recommendation works by scanning your project context to build a fit score from stack and dependency signals, maintaining a persistent registry state to record evaluations and prevent re-deciding the same questions later.

What are the limitations of using a persistent registry for skill evaluation?▼

A limitation of using a persistent registry is that it relies entirely on reading local project signals such as CLAUDE.md and common dependency files, requiring deterministic updates without a Python engine for accurate state tracking.

Related Skills

Explore More Software Engineering Skills→
liuzeming-yuxiliuzeming-yuxi
9

superclaw-onboard

Generate a centralized project context document from repository metadata and config files.

Community
Intermediate
anyscaleanyscale
2

create-skill

Generate SKILL.md files with YAML frontmatter in the .claude/skills/ directory.

Official
Basic
spoonbobospoonbobo

clawhub

Manage agent skills from the clawhub.com registry via the npm-installed clawhub CLI.

Community
Intermediate
matteotittamatteotitta
29

lead-scoring

Evaluate B2B accounts for fit and readiness using signal-based scoring.

Community
Advanced