ms365-tenant-manager

I just ran the grading rubric on your ms365-tenant-manager skill and noticed you're handling some genuinely complex multi-tenant orchestration—the architecture scores well, but the Progressive Disclosure could use some attention at 72/100. Have you considered breaking down the higher-level concepts earlier so developers don't hit a wall when they're trying to understand the tenant isolation strategy?

Links:

• GitHub
• Your agentic skill in the SkillzWave marketplace

The TL;DR

You're at 72/100, solid C-grade territory. This is based on Anthropic's best practices for skill design. Your Utility section is genuinely strong (15/20)—the PowerShell automation and security guidance addresses real gaps in M365 admin tooling. But Progressive Disclosure and Ease of Use are dragging you down (18/30 and 18/25). The biggest issue? You've got two files saying almost the exact same thing, which burns through developer tokens without adding depth.

What's Working Well

• Utility hits the mark. You're solving actual M365 problems—user management at scale, security policies, tenant configuration. That's real value, not fluff.
• Quality checklist in the pre-commit section is a nice touch. The validation approach with -WhatIf parameters shows you understand safe operations.
• Frontmatter is clean. Your YAML is valid, naming convention is correct (hyphen-case), and you've got the required fields locked down.
• Grep-friendly structure. Your skill is easy to parse and index—solid infrastructure underneath.

The Big One: Content Duplication Killing Your PDA Score

Here's what's happening: HOW_TO_USE.md duplicates about 70% of SKILL.md instead of providing reference depth. Sections like "What You'll Get," "Prerequisites," and "Safety & Best Practices" just repeat what's already in the main file. This violates the layered architecture principle—you should go one level deep, not sideways.

The fix: Consolidate SKILL.md to 100-120 lines (overview + quick start), then use HOW_TO_USE.md for deep dives: complete script templates, troubleshooting workflows, step-by-step tenant migration examples. Create separate reference files like references/security-policies.md and references/powershell-templates.md for the gnarly stuff. You'll reclaim ~5 points here and make the skill way more scannable.

Other Things Worth Fixing

1. Missing trigger phrases in metadata. Your description doesn't include terms developers actually search for—"M365 tenant", "Office 365 admin", "Azure AD users", "Global Administrator". Add these explicitly to boost discoverability. (+2 points)

2. No Table of Contents. At 197 lines with 10+ sections, SKILL.md needs a TOC. It's just navigation scaffolding, but it signals that you respect the reader's time. (+1-2 points)

3. Second-person voice scattered in HOW_TO_USE.md. Lines like "you'll get" and "you'll receive" pull away from the instructional tone. Flip these to imperative: "Get X output" instead of "you'll get X output." (+1 point)

4. No actual script templates or code examples. You mention PowerShell modules but don't show sample scripts or configuration samples. Even one concrete example (user bulk import, Conditional Access policy setup) would boost confidence in implementation. (+2-3 points)

Quick Wins

• Add trigger phrases to frontmatter description ("M365 tenant manager", "Office 365 admin", etc.) — 2 points
• Deduplicate the two files — restructure as overview + deep reference — 5 points
• Add TOC to SKILL.md — 1-2 points
• Include 1-2 actual PowerShell script templates — 2-3 points
• Fix second-person voice in HOW_TO_USE.md — 1 point

You're looking at 11-13 additional points with these fixes, which puts you solidly in B territory (83-85/100). The core skill is sound; it just needs tighter structure and clearer discoverability.

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.