Skill Library Lint and Overlap Report
Lint every skill in a repo at once: frontmatter, name-folder match, weak descriptions, overlapping triggers, broken refs, plus a CI script.
Before & after
Same prompt, same agent — the only difference is the skill.
I can help you put together a skills status report. You'd want to inventory what skills your team has, assess proficiency levels, identify gaps compared to your goals, and probably flag which areas need training or hiring. The format could be a spreadsheet or a narrative report depending on your audience. What specific areas are you concerned about?
SKILLS STATUS REPORT: Engineering Team INVENTORY SUMMARY Total skills tracked: 23 Fully proficient (tier 3): 8 Partially proficient (tier 2): 10 Novice (tier 1): 5 Gap areas: 3 DETAILED BREAKDOWN WEB BACKEND Node.js + Express: tier 3 (3 engineers, 5+ years avg) PostgreSQL: tier 3 (all 5 team members) Docker & orchestration: tier 2 (2 engineers proficient, 3 learning) System design: tier 2 (team discusses tradeoffs, lacks formal architecture reviews) FRONTEND React: tier 3 (all 5 engineers proficient) CSS/styling systems: tier 2 (component-level good, design-system integration weak) Testing (Jest/Cypress): tier 1 (basic skills, low test coverage culture) OPERATIONS GitHub Actions CI/CD: tier 2 (setup working, optimization gaps) Monitoring (Datadog): tier 1 (basic dashboards, alert strategy underdeveloped) Incident response: tier 1 (no formal runbook program) CRITICAL GAPS 1. Load testing / performance optimization (needed for 10x scale) 2. Security hardening (audit findings unaddressed) 3. Go language (required for new microservice planned Q3) RECOMMENDATIONS → Hire or train 1 Go engineer before Q3 milestone → Implement formal testing culture (3-month initiative) → Schedule security audit follow-up workshop
About this skill
Once a skills folder grows past a handful, drift sets in: broken frontmatter, names that do not match folders, thin descriptions, and skills that compete for the same requests. This lint checks every skill the same way and can leave a validator in CI.
What it does
- Validates frontmatter, name rules, name-folder match, description length, and a when-to-use signal
- Finds duplicate names, broken file references, orphan files, and oversized bodies
- Maps pairs of skills whose descriptions overlap and recommends narrowing or merging
- Reports git-based age without inventing usage data, and does not fail skills for missing non-standard fields like version
- Optionally writes a dependency-light validation script and CI step
What you get
A one-screen status table (fail, warn, pass), an overlap list, rewrite-or-remove candidates, and an optional CI check.
Best for
Teams and plugin authors maintaining a shared library of skills.
SKILL.md preview
---
name: skill-library-lint
description: Lint and inventory every skill in a repository or skills directory at once - frontmatter validity, name rules and name-folder match, description length and trigger quality, duplicate names, pairs of skills whose descriptions overlap and will compete, oversized bodies, broken file references, and staleness from git history - and produce a one-screen status table plus an optional validation script for CI. Use when the user says "audit all our skills", "lint the skills folder", "status report on our skill library", "which skills overlap?", "set up a skills check in CI", "clean up our .claude/skills", or "we have 40 skills, which ones need work?".
---
… (sign up to view the full skill)More ai & agents skills
View all AI & Agents skills →New Skill Scaffolder
Create a new Claude Code skill correctly: right mechanism, valid name and folder, a triggering description, and a smoke test.
Self-Verifying Agent Loop: Plan, Check, Retry, Close
Turns a goal into a task plan with machine-run checks, executes with retry caps, escalates, and refuses to close until verified.
Skill Test Runner (Trigger and A/B)
Test a skill in fresh sessions: ready-to-run prompts, commands, and rubric for trigger recall, false triggers, and blind A/B.
Chief AI Officer Plan Review
Evidence-first review of one AI plan: eval gaps, cost per successful task, model and vendor risk, prompt injection, and EU AI Act screen.