Skip to content
MCP server

Runwaystock

Reorder quantities, stockout dates and dead stock from your sales and stock CSVs

Runs locallyStatus and history
Install
Pricing
Open source
Platforms
CLI
Runs
Locally
Version
0.2.0
Listing updated
today

Similar MCP servers

About Runwaystock

Runwaystock is an open-source MCP server for e-commerce. It runs locally on your machine and is started by your AI client.

Runwaystock does inventory arithmetic on two spreadsheets you already have: a sales export and a stock count. For each SKU it estimates daily sales over a window (90 days by default, and at least 28 for reliable figures), applies a safety-stock and reorder-point formula, and returns a status. Orders round up to any minimum order quantity in the stock file, and a per-SKU lead time overrides the default.

Statuses are order now, order soon (reaching the reorder point within 7 days), ok, overstocked (over 180 days of cover) and no recent sales, with the money tied up in unsold stock totalled. Lead time, cover days, window and service level (90, 95 or 99) are adjustable. It reads common Shopify-style column names but does not connect to any store platform.

It ships as a single Go binary with no account, API key or network access. Through MCP, results (SKUs and quantities) reach your AI model; the command-line version stays entirely local. The README lists its limits: no seasonality or promotions, lost sales while out of stock understate demand, and it has been tested on made-up data so far. It is MIT licensed.

Install

Run in a terminal

claude mcp add runwaystock -- npx -y runwaystock

Add to claude_desktop_config.json

{
  "mcpServers": {
    "runwaystock": {
      "command": "npx",
      "args": [
        "-y",
        "runwaystock"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "runwaystock": {
      "command": "npx",
      "args": [
        "-y",
        "runwaystock"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "runwaystock": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "runwaystock"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "runwaystock": {
      "command": "npx",
      "args": [
        "-y",
        "runwaystock"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "runwaystock": {
      "command": "npx",
      "args": [
        "-y",
        "runwaystock"
      ]
    }
  }
}

Using several servers? Build one config file with the MCP config generator.

Tools this server provides (2)

Taken from the project documentation; the running server may differ.

inventory_check
Returns each SKU's status and order quantity, most urgent first.
inventory_inspect_csv
Shows detected columns and row counts without row contents.

What it does

Reorder points
Applies standard safety-stock and reorder formulas to every SKU.
Stockout dates
Estimates when each product runs out at current sales pace.
Dead stock value
Totals the money tied up in products with no recent sales.
Tunable assumptions
Lead time, cover days, window and service level can be changed.
CSV inspection
Shows detected columns and row counts without exposing row contents.
Read-only
It cannot place orders.

Good for

  • Decide what to reorder this week and in what quantity
  • See which products will run out first
  • Find overstocked items
  • Estimate cash locked in unsold stock

Pricing

Open source

MIT licensed; no account or API key

Runwaystock FAQ

Does it connect to Shopify or Amazon?

No. It reads CSV exports and recognises common column names, but does not connect to those platforms.

How much sales history is needed?

At least 28 days, or it warns the figures are rough; the default window is 90 days.

Does my data leave my machine?

The CLI is fully local. Through MCP, SKUs and quantities returned by inventory_check reach whichever AI model you use.

What are the main limits?

It does not see seasonality or promotions, assumes steady demand and a fixed lead time, and reads CSV only up to 50 MB.

Is Runwaystock free?

Yes. Runwaystock is open source, so you can use and self-host it for free.

How do I add Runwaystock to Claude Code?

Run this command in your terminal: claude mcp add runwaystock -- npx -y runwaystock

What are the best alternatives to Runwaystock?

The closest alternatives to Runwaystock are Agent-Native Analytics, Google Ads + Meta Ads + SEO/GEO MCP, dashboard-icons, ChatGPT for Amazon and Tapetide - Indian Stock Market MCP.

Where this comes from

  • Description: written from the project's README.
  • Install steps: from the official MCP Registry entry.

How AIPoweredBot checks servers