loyalty-specialist

JeriCommerce Claude Code Plugins

Official plugin marketplace for Claude Code by JeriCommerce.

Installation

Add this marketplace to your Claude Code:

/plugin marketplace add JeriCommerce/claude-plugins

Then install individual plugins:

/plugin install loyalty-specialist@jericommerce-plugins

Available Plugins

Plugin	Description	Version
loyalty-specialist	Jeri - Generate professional PDF loyalty program performance reports by querying Metabase and analyzing program data	202603.10.1
bug-reporter	Jeri - Report bugs and customer issues to Linear through guided conversational flow with screenshots/video support	202602.19.0
qa-specialist	Jeri - QA testing assistant — verify features, test endpoints, inspect logs, and document findings in Linear	202602.21.0
feature-request	Jeri - Capture customer feature requests as user stories in Linear with customer association	202603.11.0
loyalty-template-creator	Jeri - Create CDN loyalty page templates for Shopify theme extensions by replicating a reference website's style	202603.10.0
niche-research	Jeri - Generate pSEO niche research reports from Ahrefs data, ready to copy into the JeriCommerce blog repo	202603.12.0
optimize-skills	Audit and restructure Claude Code skills into modular folder structures for better consistency and maintainability	202603.21.0
communication	Jeri - On-brand content creation: marketing copy, cold emails, social posts, VoC research. EN/ES/IT	202603.30.0

Updating

To get the latest plugins after changes are pushed:

/plugin marketplace update

---

Creating a New Plugin

1. Directory structure

Create a new folder under plugins/ with this structure:

plugins/my-plugin/
├── .claude-plugin/
│   └── plugin.json        # Required: plugin metadata
├── skills/                 # Optional: auto-invoked by Claude when relevant
│   └── my-skill/
│       ├── SKILL.md        # Skill definition (frontmatter + instructions)
│       └── references/     # Optional: supporting files the skill can read
│           └── data.md
├── commands/               # Optional: user-invoked via /my-plugin:command-name
│   └── my-command.md
├── agents/                 # Optional: specialized subagents with isolated context
│   └── my-agent.md
└── README.md               # Recommended: plugin documentation

All directories are optional except .claude-plugin/. Include only what your plugin needs.

2. Plugin manifest (plugin.json)

Create .claude-plugin/plugin.json with your plugin's metadata:

{
  "name": "my-plugin",
  "description": "Short description of what the plugin does",
  "version": "1.0.0",
  "author": {
    "name": "JeriCommerce"
  },
  "keywords": ["relevant", "search", "terms"]
}

Field	Required	Description
name	Yes	Unique identifier in kebab-case. Becomes the namespace for commands (/my-plugin:command)
description	No	Shown in the plugin manager when browsing plugins
version	No	Semantic versioning (MAJOR.MINOR.PATCH). Bump this on every change or users won't see updates
author	No	Attribution. { "name": "...", "email": "..." }
keywords	No	Tags for plugin discovery

3. Skills (auto-invoked by Claude)

Skills are loaded automatically when Claude determines they're relevant to the current task. Create a folder inside skills/ with a SKILL.md file.

skills/my-skill/SKILL.md:

---
name: my-skill
description: >
  Describe WHEN Claude should use this skill. Include trigger phrases
  and scenarios. Example: "Use when the user asks about X, works on Y,
  or mentions Z."
---

# My Skill

Instructions for Claude when this skill is activated.

## Reference Files

You can point Claude to supporting files in the same directory:

Read `${CLAUDE_PLUGIN_ROOT}/skills/my-skill/references/data.md` for details.

Frontmatter fields:

Field	Required	Description
name	Yes	Skill identifier
description	Yes	Tells Claude when to activate the skill. Be specific with trigger phrases
version	No	Skill version
disable-model-invocation	No	Set to true if only the user should invoke it (not Claude automatically)

Tips:

• Use ${CLAUDE_PLUGIN_ROOT} to reference files within your plugin directory
• Put large reference data in references/ subdirectory to keep SKILL.md focused
• The description field is critical — Claude reads it to decide when to use the skill

4. Commands (user-invoked via slash command)