improve-skills

Audit and improve skill files and metadata across a repository.

3|Updated Apr 5, 2026
One-click install
npx skills add https://github.com/dvy1987/agent-loom --skill improve-skills-dvy1987
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: improve-skills
Source: https://github.com/dvy1987/agent-loom/tree/main/.agents/skills/improve-skills
Command: npx skills add https://github.com/dvy1987/agent-loom --skill improve-skills-dvy1987

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill automates a systematic improvement pass over a skills repository to find, fix, and strengthen individual Skill files so the library stays current, accurate, and well-linked.

Core Features & Use Cases

  • Automated Validation & Scoring: Runs validate-skills to surface structural gaps and produce a prioritized work queue.
  • Per-Skill Improvement Cycle: Prunes stale content, fixes structural gaps, performs link checks, runs research with security gates, rewrites content, re-scores, and enforces size policies.
  • Governance & Reporting: Generates Impact Reports, cross-link repair triggers, commit-ready messages, and a final library summary for audits and maintainers.
  • Use Case: Maintainers managing a large skill library can run this to rapidly upgrade low-scoring skills, add missing metadata, and ensure safe use of external research.

Quick Start

Run an improvement pass across the repository to audit, repair structural gaps, rewrite skills where needed, and produce the improvement summary report.

Frequently Asked Questions about improve-skills

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

FAQPage Schema
How do I audit and upgrade a large skill library systematically?

Auditing a skill library involves running automated validation to surface structural gaps, pruning stale content per skill, fixing metadata, enforcing size policies, and producing commit-ready artifacts with an impact report for maintainers.

What is the best way to fix structural gaps and broken cross-links in skill files?

Fixing structural gaps and cross-links involves running validation to identify issues, triggering cross-link repair, rewriting content, and enforcing validation security gates to ensure commit-ready artifacts for maintainers.

How does automated skill pruning and research-driven rewriting work?

Automated skill pruning works by running validation checks, removing outdated content, applying research-driven rewrites, bumping manifest versions, and re-scoring the skill to ensure the repository remains current and accurate.

Can I use automated validation and scoring to manage skill file sizes and metadata?

Yes, automated validation and scoring manage skill file sizes and metadata by enforcing size policies, fixing structural gaps, bumping manifest versions, and generating a prioritized work queue for systematic repository improvement.

What are the limitations of running a full-scope skill library improvement pass?

Limitations of a full-scope improvement pass include the necessity of enforcing security gates for external research and the requirement to bump manifest versions, meaning incomplete security validation will block commit-ready artifact generation.