Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
Noibu avatar

Noibu

  • 5 repo stars
  • Updated July 13, 2026
  • Noibu/ai-plugin

Connects Claude to your Noibu store data — session analytics, error diagnosis, checkout analysis, product performance, and more.

About

noibu is a Claude Code skill in the AI & Agent Building category. Connects Claude to your Noibu store data — session analytics, error diagnosis, checkout analysis, product performance, and more.

  • noibu
  • AI & Agent Building
  • AI-coding skill

Noibu by the numbers

  • Data as of Jul 21, 2026 (Skillselion catalog sync)
/plugin marketplace add Noibu/ai-plugin
/plugin install noibu@noibu-plugins

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
repo stars5
Last updatedJuly 13, 2026
RepositoryNoibu/ai-plugin

What it does

Connects Claude to your Noibu store data — session analytics, error diagnosis, checkout analysis, product performance, and more.

README.md

Noibu AI Plugin

Official Noibu plugin for AI clients. Access Noibu products direclty from your AI coding tool.

Description

Built for ecommerce, the Noibu plugin bridges the gap between customer experience and revenue by connecting Claude directly to your store's session, error, and conversion data through Noibu — and to the marketing, support, and commerce platforms that put insight into motion. Go beyond analysis: surface what's costing you revenue, take action across your stack, build workflows to automate work end-to-end.

Installation

Claude Desktop (manual)

  1. Download the Noibu plugin: https://github.com/Noibu/ai-plugin/releases/latest/download/noibu.zip
  2. Go to Customize
  3. Click Create pluginUpload plugin
  4. Upload the noibu.zip file

Claude Code

Install directly from the marketplace — no download required:

/plugin marketplace add Noibu/ai-plugin
/plugin install noibu@noibu-plugins
/reload-plugins

Releases & distribution

This repo (Noibu/ai-plugin, public) is the source of truth and feeds two channels:

  • Claude Code — installs from this repo as a marketplace (Noibu/ai-plugin).
  • Cowork — served from the private mirror Noibu/ai-plugin-sync.

Every push to main runs .github/workflows/release.yml, which:

  1. Bumps the patch version (from the latest v* git tag),
  2. Stamps it into noibu.zip only (the committed plugin.json carries no version),
  3. Tags + creates a GitHub release with the versioned noibu.zip,
  4. Force-pushes main to the private ai-plugin-sync mirror.

Versioning is by commit SHA, deliberately. main is branch-protected, so the workflow cannot push a version-bump commit back to it. Instead, plugin.json omits the version field: the marketplace then versions the plugin by commit SHA, so every merge to main is a new version and Claude Code auto-updates with no write to main. The semver tag/release exists only for the downloadable noibu.zip (Claude Desktop + the public directory). Do not add a version field back to plugin.json — it would re-freeze marketplace updates.

To distribute to all Claude Code users automatically, deploy a managed-settings.json via MDM with extraKnownMarketplaces (pointing at Noibu/ai-plugin, autoUpdate: true) and enabledPlugins: { "noibu@noibu-plugins": true }.

How to develop

  1. Clone and install the plugin
  2. Authenticate with Noibu

Documentation

https://help.noibu.com/articles/7434601323-installing-the-noibu-plugin-for-claude

Example Use Cases

Example 1:

Spot and act on your top revenue opportunities on autopilot: Every Monday, surface my highest-impact opportunities across acquisition, products, and experience, and help me take action on each.

Example 2:

Optimize your checkout: Find where shoppers are dropping off, the payment and delivery methods they're using, and the technical issues hurting conversion, then tackle the top fixes.

Example 3:

Resolve technical issues at the source: What errors and performance issues are costing my store the most revenue, and help me resolve them.

Example 4:

Uncover and resolve shopper friction: Surface where real shoppers are struggling, through session replays and revenue heatmaps, and act on the highest friction.

Example 5:

Turn product views into sales: Find which products and collections are getting views but no sales, diagnose why they aren't converting, and execute on the biggest opportunities.

Example 6: Shift marketing spend with site behaviour: Surface which campaigns are sending high-quality traffic, identify how to lift upper-funnel conversion, and reallocate budget accordingly.

License

MIT

Related skills

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.