New Skill: wp-plugin-directory-guidelines

[davidperezgar]

Summary

Adds the wp-plugin-directory-guidelines skill — a structured review checklist for AI assistants evaluating WordPress plugins against the Detailed Plugin Guidelines and GPL licensing requirements.

Fixes #21

---

What's included

SKILL.md — structured as actionable review steps, not a reference doc:

• Each of the 18 guidelines reformulated with Check / Violation signals / Verdict / Fix structure
• Guideline 5 (Trialware) includes code pattern examples, review questions, and a pre-submission checklist
• Guideline 17 (Trademarks) cross-references the Naming Rules section for slug/name validation
• GPL compliance deep-dive: version summary, valid headers, accepted/rejected license tables sourced from Plugin Check's License_Utils trait, common violation patterns, and key concepts (distribution, derivative works, source availability)

README.md — added wp-plugin-directory-guidelines to the Available Skills table

---

Design decisions

• Check/Violation/Verdict/Fix per guideline — replaces encyclopedic summaries with reviewer-ready instructions; each finding cites a specific guideline number
• Guideline 18 marked informational — no code or readme check required; avoids false positives
• License data from Plugin Check — accepted licenses match the is_license_gpl_compatible() regex in plugin-check, not a hand-curated list
• GPLv1 omitted — WordPress requires GPLv2 or later; v1 is not relevant for plugin review
• No scripts — purely Markdown-based knowledge skill per repo conventions

---

Test plan

• SKILL.md frontmatter parses correctly (name, description, compatibility)
• All references/ links in SKILL.md resolve to existing files
• Skill structure matches repo conventions (compare with wp-project-triage)
• Accepted licenses table matches current License_Utils.php in plugin-check
• All 18 guidelines match the current published version of the Detailed Plugin Guidelines

[github-actions]

The following accounts have interacted with this PR and/or linked issues. I will continue to update these lists as activity occurs. You can also manually ask me to refresh this list by adding the props-bot label.

If you're merging code through a pull request on GitHub, copy and paste the following into the bottom of the merge commit message.

Co-authored-by: davidperezgar <davidperez@git.wordpress.org>
Co-authored-by: justlevine <justlevine@git.wordpress.org>
Co-authored-by: obenland <obenland@git.wordpress.org>

To understand the WordPress project's expectations around crediting contributors, please review the Contributor Attribution page in the Core Handbook.

[davidperezgar]

Summary

• Expanded the skill to cover trialware/freemium compliance with a new “Trialware & Upsell Checks” section under Guideline 5.
• Added review questions, checklist, and common violation patterns to help detect trialware behavior.
• Updated “When to use” and description to include upsell/licensing flow review.
• Restored required compatibility frontmatter for eval harness.

Testing

• Not run (please run node eval/harness/run.mjs).

[justlevine]

Thanks for submitting this @davidperezgar 🙌

We should do our best to align with Skills best practices, particularly around progressive disclosure. For example, at most a skill should have 5K tokens, whereas this in its current form is around 8500-10,000 tokens depending on the model.

I left some more specific feedback inline on the PR, but if you havn't yet, I recommend you read both that Claude doc and the best practices from here: https://agentskills.io/skill-creation/best-practices .

[obenland]

I can recommend using the skill-creator skill to help with this.

In general, is the intent for this to be an all-encompassing plugin review skill, or more focussed on just guidelines and license? I assume there's more to a review? But I'm not sure.

[davidperezgar]

Hello, thanks for your suggestions. I'll take care of them.

[davidperezgar]

@obenland and @justlevine all changes have been done. Could you check again? Thanks.

[davidperezgar]

Thanks for all suggestions @obenland . I've fixed, and added more examples in the rest of guidelines.

[justlevine]

This skill looks much closer to the recommended best practices and skills that have been evaled for effectiveness, and resembles what I would expect a well-authored skill to look like. 🚀

We're still unable to return home, so I cannot eval them myself. looking forward to seeing the results.