# ReplenishRadar/MCP [Health: Active]

**Category:** 🛠️ Other Tools and Integrations  
**Repository:** https://github.com/ReplenishRadar/MCP  
**GitHub Stars:** 2  
**npm Downloads (last month):** 241  
**Views:** 4  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/replenishradar-mcp

## Description
Multi-channel inventory intelligence for Shopify and Amazon sellers. 28 tools for stockout risk, demand forecasts, purchase order lifecycle, and sales analytics. Human-in-the-loop: all write operations create drafts only.

## Tools
Capabilities this server exposes over MCP:

- **rr_get_stockout_risk** — Stockout risk levels for SKUs
- **rr_get_inventory_position** — Stock-by-location for an item
- **rr_get_demand_forecast** — Demand forecast stats
- **rr_get_suggested_purchase_orders** — Suggested POs
- **rr_get_replenishment_actions** — Canonical buyer replenishment actions
- **rr_get_replenishment_action** — One replenishment action with event history
- **rr_get_alerts** — Active alerts
- **rr_list_items** — List inventory items
- **rr_get_sync_status** — Recent sync history
- **rr_list_suppliers** — List vendors/suppliers
- **rr_list_purchase_orders** — List purchase orders
- **rr_get_purchase_order** — Single PO with line items
- **rr_get_po_pdf** — PO PDF info
- **rr_get_po_documents** — Documents attached to a PO
- **rr_get_po_notes** — Notes attached to a PO
- **rr_get_sales_history** — Sales history
- **rr_get_top_sellers** — Top-selling SKUs
- **rr_get_slow_movers** — Slow-moving SKUs
- **rr_get_inventory_value** — Inventory value breakdown
- **rr_get_sku_health** — SKU-level health summary
- **rr_get_lost_sales** — Estimated lost sales from stockouts
- **rr_get_store_health** — Store connection and sync health
- **rr_get_data_freshness** — Freshness status by data dimension
- **rr_get_setup_status** — Setup milestones and next step
- **rr_get_recent_activity** — Recent alerts, POs, and sync rollup
- **search_knowledge** — ReplenishRadar product knowledge search
- **rr_evaluate_fit** — Deterministic ICP + tier fit verdict (no org data)
- **rr_acknowledge_alert** — Acknowledge an alert
- **rr_add_po_note** — Add a note to a PO
- **rr_request_approval** — Request human approval for a PO
- **rr_add_replenishment_action_note** — Add a note to a replenishment action
- **rr_dismiss_replenishment_action** — Dismiss a replenishment action
- **rr_prepare_replenishment_action** — Preview an action and get its current `updated_at` before execute
- **rr_execute_replenishment_action** — Execute an action. `mode="draft"` (default) creates a draft PO and/or transfer; `mode="autonomous"` also sends the PO / commits the transfer when the key has sensitive_write + the operation tool group + an enabled budget and all guardrails pass (else draft-fallback or hard-refuse). Idempotent, stal…
- **rr_resolve_replenishment_action** — Resolve an action as no-action with a reason; idempotent, stale-state protected
- **rr_retry_replenishment_action** — Retry a blocked or failed action via the draft execution path; idempotent, stale-state protected
- **rr_get_document_upload_url** — Get upload URL for a PO document
- **rr_record_pi_review** — Record proforma invoice details
- **rr_trigger_sync** — Trigger inventory sync
- **rr_create_purchase_order** — Create a draft PO
- **rr_update_purchase_order** — Update a draft PO
- **rr_send_purchase_order** — Send an approved PO
- **rr_cancel_purchase_order** — Cancel a PO
- **rr_create_location** — Create a manual inventory location
- **rr_set_stock_at_location** — Set stock at a manual location
- **rr_get_agent_context** — Get one entry by `namespace` + `key`; expired entries are omitted
- **rr_set_agent_context** — Store a bounded JSON object under `namespace` + `key` (optional `scope_ref`, `ttl_seconds`)
- **rr_list_agent_context** — List live entries for ONE `namespace` (required), bounded `limit` (max 200)
- **rr_delete_agent_context** — Delete one entry by `namespace` + `key`; records actor provenance

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "mcp": {
    "command": "npx",
    "args": ["-y","@replenishradar/mcp-server"],
    "env": {
      "REPLENISHRADAR_API_KEY": ""
    }
  }
}
```

**Requires environment variables:** `REPLENISHRADAR_API_KEY` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What ReplenishRadar/MCP MCP server does

ReplenishRadar/MCP connects an MCP-compatible AI client to live inventory data held by ReplenishRadar. The service is intended for sellers operating across Shopify and Amazon channels. Agents can inspect inventory conditions, identify replenishment needs, review purchasing records, and retrieve operational or product knowledge information.

The read surface covers stockout risk, inventory by location, demand forecasts, suggested purchase orders, replenishment actions, active alerts, items, suppliers, purchase orders, sales history, top sellers, slow movers, inventory value, SKU health, estimated lost sales, store health, data freshness, setup progress, recent activity, and synchronization history. It also includes a knowledge search and a deterministic fit evaluation that does not use organization data.

## How it works

The client launches the package as a local stdio MCP process with `npx`, then supplies a ReplenishRadar API key through the process environment. Tool calls use that authenticated connection to query the organization’s ReplenishRadar data. The repository documentation identifies Claude Desktop, OpenClaw, and MCP-compatible clients as supported connection options.

Write access is separated by capability and plan. Basic writes include acknowledging alerts, adding purchase-order or replenishment-action notes, requesting approval, and managing documented action workflows. Sensitive operations include syncing inventory, creating or changing purchase orders, sending or cancelling orders, creating locations, and setting stock at manual locations.

Purchase orders begin as drafts. The default replenishment execution mode creates a draft purchase order or transfer without sending it to a supplier or committing stock. Autonomous execution can send or commit only when the API key has the required capabilities, operation group, enabled budget, and passing guardrails. Preparation, timestamps, idempotency keys, and approval fallbacks support safer retries and stale-state handling.

## Setup and configuration

Obtain an API key from ReplenishRadar under Settings > API Keys. Add a server entry to the MCP client configuration using the `@replenishradar/mcp-server` package and set `REPLENISHRADAR_API_KEY` to the key value. Claude Desktop configuration is edited from Settings > Developer > Edit Config, followed by a client restart.

The repository also documents a public agent-intake endpoint that can provide a short-lived, read-only sandbox key for users without an account. API-key permissions, namespaces, sensitive capabilities, and autonomous spend budgets are managed by human administrators in ReplenishRadar settings rather than through MCP tools.

## Tools and capabilities

ReplenishRadar/MCP includes tools for:

- Reading inventory, demand, sales, alerts, suppliers, purchase orders, documents, notes, and synchronization state.
- Reviewing replenishment actions and their event history.
- Searching ReplenishRadar product knowledge and evaluating product fit without organization data.
- Acknowledging alerts, adding notes, and requesting human approval.
- Preparing, executing, resolving, retrying, or dismissing replenishment actions where the API key and plan permit.
- Managing documented agent-context entries as bounded, organization-scoped JSON objects when the relevant namespace permissions are granted.

Plan limits documented for the service are 10 calls per hour on Standard, 100 on Growth, and 1,000 on Scale. The available read, basic-write, sensitive-write, and diagnosis/status capabilities vary by plan.

_Full upstream README: https://allmcps.com/mcp/replenishradar-mcp/readme_