Boxem MCP for Openclaw

Boxem MCP connects Openclaw Skills to Amazon seller operations, analytics, automation, knowledge, and notifications through an SSE-based MCP server.

boxem
v1.3.0
Sep 26, 2026
0
44
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install boxem

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install boxem using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Boxem MCP?

Boxem MCP is a Model Context Protocol integration for Amazon sellers who need programmatic access to product data, listing workflows, SKU analytics, supplier information, and daily performance statistics. It gives Openclaw Skills a structured way to interact with the Boxem platform through remote tools instead of relying on manual dashboard operations.

Beyond e-commerce data, the skill supports Lua code execution, proxied HTTP requests, persistent memory, knowledge-base lookup, email notifications, and recurring background AI jobs. This combination makes Boxem useful for seller research, operational automation, repricing analysis, listing creation, and repeatable marketplace workflows.

Boxem MCP Use Cases

  • Research Amazon products by ASIN, including product names and sales rank.
  • Check listing restrictions and retrieve ungate URLs before sourcing or listing products.
  • Review SKU-level item analytics to inform repricing decisions.
  • Create new Amazon listings through the Boxem platform.
  • Retrieve daily analytics for date ranges of up to 30 days per call.
  • Run Lua scripts or fetch web resources with the hosted execution environment.
  • Send proxied GET, POST, PUT, or DELETE HTTP requests from an agent workflow.
  • Schedule recurring background AI jobs with access to supported tools.
  • Look up FBA, WFS, e-commerce, Boxem, and selling guidance from the knowledge base.
  • Preserve user preferences and conversation context with file-based memory.
  • Notify users by email and in-app messages when an automated workflow completes.

How Boxem MCP Works

  1. Configure an MCP client with the Boxem server URL and SSE transport.
  2. Discover the exposed tool definitions using the client or mcporter schema command.
  3. Invoke Amazon seller tools for catalog research, ungating checks, SKU analytics, supplier lookup, or listing creation.
  4. Request daily analytics within the supported maximum range of 30 days per call.
  5. Extend workflows with Lua execution, HTTP proxy requests, knowledge lookup, or persistent memory.
  6. Create interval jobs when an operation should run repeatedly in the background, then monitor or remove those jobs by UUID.
  7. Deliver workflow results through the MCP response, email notifications, or in-app notifications.
  8. Apply operational safeguards: repriceAmazonItems is exposed but not currently implemented, and Lua execution has a 30-second timeout.

Boxem MCP Setup

1. Add the Boxem MCP server with mcporter

mcporter config add boxem --url https://new.boxem.com/backend/mcp
mcporter list boxem --schema

2. Configure an MCP client directly

Add the following server definition to the client configuration:

{
  "boxem": {
    "url": "https://new.boxem.com/backend/mcp",
    "transport": "sse"
  }
}

3. Verify connectivity

Use the client schema or tool-listing command to confirm that Boxem tools are available. The server endpoint is https://new.boxem.com/backend/mcp, and the required transport is Server-Sent Events (SSE).

4. Start with read-oriented operations

Test catalog lookup, daily analytics, knowledge lookup, or supplier retrieval before running listing creation, HTTP requests, code execution, or background jobs. Confirm authentication and platform permissions according to the Boxem environment.

Boxem MCP Data Schema & Taxonomy

Connection metadata

Field Value
Skill name boxem
Server URL https://new.boxem.com/backend/mcp
Transport SSE (Server-Sent Events)
Version 1.0.0

Functional data domains

  • Amazon catalog data: ASINs, product names, and sales ranks.
  • Listing controls: restriction checks, ungate URLs, and listing creation requests.
  • Seller analytics: SKU-level item analytics and daily statistics over date ranges up to 30 days per call.
  • Automation: Lua source or file paths, proxied HTTP requests, and interval-job UUIDs with token usage details.
  • Knowledge: an indexed collection of files covering FBA, WFS, e-commerce, Boxem, and selling advice.
  • Suppliers: seller supplier records returned by the Boxem platform.
  • Memory: persistent, file-based user preferences and conversational context.
  • Notifications: email and in-app notification events.

The skill does not specify a local output directory or a fixed client-side file format. Knowledge files, memory files, Lua files, analytics records, and job metadata are handled through the connected Boxem tools and the MCP client configuration.

Boxem MCP Advanced Features

  • SSE-based MCP connectivity for agent-compatible remote tool invocation.
  • Lua execution through executeCode, including website retrieval with fetch().
  • Lua file execution through executeLuaFile for reusable automation scripts.
  • Proxied HTTP support for GET, POST, PUT, and DELETE requests.
  • Recurring background AI jobs with tool access through createIntervalJob.
  • Interval-job lifecycle management with listing, token-usage visibility, UUID tracking, and deletion.
  • Persistent file-based memory for preferences and context across conversations.
  • Indexed knowledge-base access with a general index and targeted file lookup.
  • Email and in-app notifications for workflow results.
  • Amazon-specific workflows spanning catalog research, ungating, analytics, supplier discovery, and listing creation.
  • Operational constraint awareness: Lua execution times out after 30 seconds, analytics calls support up to 30 days, and repriceAmazonItems is currently not implemented.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*