scrum-master-agent

I took a look at your scrum-master-agent skill—the 68-point score reflects some solid foundational work, but there's a gap between what the skill could do and what it actually delivers. You've got the bones of something useful for agile teams, but the Progressive Disclosure architecture needs some rethinking to make it genuinely intuitive for developers who just want to run it without reading a manual.

Links:

• GitHub
• Your agentic skill in the SkillzWave marketplace

The TL;DR

You're at 68/100, which puts you in D territory. This is based on Anthropic's Progressive Disclosure Architecture standards—basically, how efficiently your skill uses developer tokens while remaining discoverable and usable. Your Utility scores strong at 15/20 (the skill actually solves real Scrum Master problems), but Writing Style drags you down hard at 4/10 (marketing language everywhere). The real issue isn't what the skill does—it's how it presents what it does.

What's Working Well

• Utility is genuinely solid. Your sprint calculations, capacity planning, and multi-tool integration (Linear, Jira, GitHub) address real workflow gaps. The skill solves concrete problems.
• Your examples in HOW_TO_USE.md are strong. Sample data files and concrete workflows show you've thought about real usage patterns.
• Clear scope boundaries. The skill knows what it does and doesn't do—that's worth +1 modifier bonus and saves developers from guessing.

The Big One: Marketing Language is Killing Your Objectivity Score

Your SKILL.md reads like sales copy instead of technical documentation. Phrases like "production-ready," "intelligent," "comprehensive," and "ultra-lightweight" aren't describing what your skill does—they're selling it. Line 8, line 15, line 22, line 335... it's pervasive. A developer scanning your skill is looking for triggers and capabilities, not marketing adjectives.

Fix: Rewrite your frontmatter description to be factual and trigger-focused:

• Current: "Comprehensive Scrum Master assistant designed for SaaS startups..."
• Better: "Analyze sprint velocity, burndown charts, and capacity planning. Use with 'sprint planning', 'daily standup', 'retrospective', 'burndown analysis', or attach Linear/Jira/GitHub exports."

This alone could push you +2 points on Writing Style and +2 on Spec Compliance (trigger phrases).

Other Things Worth Fixing

1. SKILL.md is 405 lines—way too long. Your Progressive Disclosure score is 18/30 because developers hit a wall of text. SKILL.md should be a 100-150 line overview with "See README.md and HOW_TO_USE.md for details" references. Move the detailed calculations and best practices sections to your reference docs. This could gain you +3 points on PDA.

2. No Table of Contents anywhere. All three files (SKILL.md, README.md, HOW_TO_USE.md) exceed 300 lines each, but none have TOCs. Add one right after frontmatter—takes 5 minutes, gains +2 points on Navigation Signals.

3. Missing an invocation section at the top. Your "How to Use" is buried at line 169. Put a quick Invocation section within the first 50 lines showing exactly how to invoke this (@scrum-master-agent pattern, with examples). Right now developers have to hunt for it. Gains +2 on Discoverability.

4. Terminology is inconsistent. You mix "data source" with "tool," "standup" with "daily standup"—minor, but it adds friction. Pick one term per concept and stick with it throughout all three docs.

Quick Wins

• Strip marketing language from SKILL.md—rewrite as technical fact. (+2-3 points)
• Cut SKILL.md in half—move details to README. (+3 points)
• Add trigger phrases to your description field. (+2 points)
• Add TOCs to all files >100 lines. (+2 points)

Those four changes alone could push you from 68 to 77+. The skill already works—now make it discoverable.

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.