Generates merchant dispatch slips with embedded Google Maps GPS QR codes to help local couriers navigate to delivery locations.
Enables Meta Muse agents to search live neighborhood retail inventory and create zero-PII reservations with checkout handoff links.
Supports non-custodial direct merchant settlement via PayPal checkout links, allowing shoppers to pay merchants without exposing sensitive payment data to the LLM context.
Supports non-custodial direct merchant settlement via Stripe checkout links, allowing shoppers to pay merchants without exposing card data to the LLM context.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@muse-commerce-mcp find coffee beans in stock within 10km"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Open-source Model Context Protocol (MCP) bridge connecting Meta Muse, Llama, and Anthropic Claude agents to live neighborhood retail store shelves.
Autonomous AI shopping agents (Meta Muse, Claude, Llama) can search global e-commerce giants, but neighborhood brick-and-mortar shops remain invisible to them. Traditional local stores cannot afford expensive POS integrations or give up 30% commissions to delivery marketplaces.
muse-commerce-mcp bridges conversational AI agents directly to verified local retail inventories via the AisleNode Protocol:
Real-Time Geofenced Discovery: Query in-stock products across physically audited neighborhood stores using Haversine GPS radius filtering.
Zero-PII Agent Reservation: Agents reserve items and generate a secure checkout handoff link without handling sensitive shopper credit cards or home addresses inside the LLM context.
100% Non-Custodial Direct Settlement: Shoppers pay the merchant directly via Stripe or PayPal. The protocol never touches or pools merchant revenue.
Instant Counter & Courier Dispatch: Orders chime live on the merchant's counter web terminal and generate a thermal dispatch slip with an embedded Google Maps GPS QR code for local bike couriers.
Related MCP server: Mercora
sequenceDiagram
participant Shopper
participant Agent as Meta Muse / Claude Agent
participant MCP as muse-commerce-mcp
participant Hub as AisleNode Protocol
participant Store as Local Merchant Counter
Shopper->>Agent: "Find fresh Arabica coffee beans near me under $20"
Agent->>MCP: search_catalog(query="coffee", lat=40.756, lng=-73.988)
MCP->>Hub: POST /api/mcp (Haversine Radius Search)
Hub-->>Agent: Returns verified in-stock items + distance (0.8 km)
Shopper->>Agent: "Order 1 bag from Soltrix Store"
Agent->>MCP: create_order(store_id, items=[{product_id, qty: 1}])
Hub-->>Agent: Returns Zero-PII Checkout URL
Shopper->>Store: Completes direct Stripe/PayPal payment & road-distance fee
Store-->>Shopper: Prints GPS QR Courier Slip & dispatches local riderAdd muse-commerce-mcp to your agent configuration file (claude_desktop_config.json or Meta Muse MCP config):
{
"mcpServers": {
"muse-commerce": {
"command": "npx",
"args": ["-y", "muse-commerce-mcp"]
}
}
}If your agent framework supports direct HTTP tool execution without a local stdio wrapper, query the live production gateway directly:
Endpoint: https://aislenode.soltrixstudios.com/api/mcp
Method: POST
curl -X POST https://aislenode.soltrixstudios.com/api/mcp \
-H "Content-Type: application/json" \
-d '{
"action": "search_catalog",
"params": {
"query": "coffee",
"customer_lat": 40.756076,
"customer_lng": -73.988407,
"max_distance_km": 15
}
}'Tool Name | Description | Required Inputs |
| Search live in-stock products across verified physical stores sorted by distance. |
|
| Reserve stock and generate a direct merchant checkout URL. Zero shopper PII required from the LLM. |
|
| Poll real-time order payment, packing, and courier dispatch status. |
|
| Autonomous trust & safety reporting for phantom stock, deceptive pricing, or bad coordinates. |
|
Once connected, test your agent with natural language queries:
"Search local stores within 10km of latitude 40.756, longitude -73.988 for Arabica coffee beans."
"Create an order for 1 bag of Arabica Coffee Beans from that store and give me the checkout link."
"Check the fulfillment status of order [order_id]."
Are you a merchant or testing a full end-to-end physical counter workflow?
Sign in with Google, pin your physical storefront GPS coordinates, and upload your CSV catalog.
Once verified, your shelves are immediately discoverable by every agent running muse-commerce-mcp.
MIT License. Built and maintained by Soltrix Studios.
Check live fulfillment, payment, and courier dispatch status for an existing local retail order.
| Name | Required | Description | Default |
|---|---|---|---|
| order_id | Yes | UUID of the order to inspect |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full behavioral burden. It discloses that this is a live status check across three operational domains, implying a read-only diagnostic operation, but it does not state permissions, rate limits, caching behavior, or whether any state is changed.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
A single front-loaded sentence with zero wasted words. The action and status domains are stated immediately.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a simple one-parameter read tool, the description is nearly complete: it identifies the input context and the status domains inspected. It does not describe the response shape, but with no output schema the description could optionally say more; the remaining gap is minor.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The single order_id parameter is fully documented in the schema as a UUID. The description adds contextual scope ('existing local retail order') but no additional format, source, or example details beyond what the schema provides, so the baseline of 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description uses a specific verb, 'Check,' and a specific resource, order status, then enumerates the status domains: fulfillment, payment, and courier dispatch. It clearly distinguishes this read-only status lookup from sibling tools like create_order, search_catalog, and report_store.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The phrase 'for an existing local retail order' establishes the context and prerequisite that the order must already exist. It does not explicitly name alternatives or when-not-to-use cases, so it stops short of full routing guidance.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Reserve in-stock items at a local physical store and generate a zero-PII, non-custodial checkout URL. The shopper completes payment and delivery details directly on the verified merchant checkout page.
| Name | Required | Description | Default |
|---|---|---|---|
| items | Yes | Array of products and quantities to reserve | |
| store_id | Yes | UUID of the verified physical store | |
| agent_source | No | Identifier of the calling AI agent (e.g. "meta_muse", "claude_desktop") | meta_muse |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full behavioral disclosure burden. It helpfully reveals that the checkout URL is zero-PII and non-custodial, and that payment and delivery details are handled on the merchant page. However, it omits important operational details such as what happens if items are out of stock, reservation expiration, idempotency, or authentication requirements.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two sentences, front-loaded with the core action and result, with no wasted words. Every sentence earns its place by clarifying the tool's purpose and the checkout flow.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the absence of an output schema and annotations, the description does well to explain that a checkout URL is generated and that the shopper completes payment elsewhere. It covers the essential flow but could be slightly more complete by mentioning expected return format or error conditions.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so all parameters (store_id, items, agent_source) are already documented in the schema. The description adds no parameter-specific meaning beyond what the schema provides, so the baseline score of 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states specific verbs and resources: 'Reserve in-stock items at a local physical store' and 'generate a zero-PII, non-custodial checkout URL.' This clearly distinguishes the tool from siblings like search_catalog (search), check_order_status (status check), and report_store (reporting). An agent can immediately understand the tool's function without opening the schema.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage context ('Reserve in-stock items at a local physical store') and the flow (shopper completes payment on merchant page), but it does not explicitly state when to use this tool versus alternatives or any preconditions. However, the context is clear enough for an agent to infer appropriate use.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Submit an autonomous trust-and-safety violation report if a store exhibits deceptive pricing, phantom stock, or invalid coordinates.
| Name | Required | Description | Default |
|---|---|---|---|
| details | Yes | Detailed explanation of the discrepancy observed by the agent or user | |
| category | Yes | Violation category | |
| store_id | Yes | UUID of the store being reported | |
| agent_source | No | Identifier of the reporting AI agent | meta_muse |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full behavioral burden. It says the report is 'autonomous' but never states what happens after submission—whether it is filed immediately, whether duplicates are rejected, whether the agent needs confirmation, or what permissions are required for a write operation against a store.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
A single front-loaded sentence with the verb first and the qualifying conditions immediately after; no filler and nothing repeated from the schema.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a mutation tool with zero annotations and no output schema, the description covers selection adequately but leaves the consequence side thin—the agent cannot tell what a successful submission returns or whether the report can be amended. The rich 100%-covered schema offsets some of that gap, keeping it at minimum-viable-plus.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so store_id, category, details, and agent_source are already documented in the schema. The description's examples hint at which categories apply but add no syntax, format, or guidance for the details field or the agent_source default beyond what the schema supplies, so the baseline 3 is correct.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Names a specific verb ('Submit') and resource ('trust-and-safety violation report') and enumerates the triggering conditions (deceptive pricing, phantom stock, invalid coordinates), which map onto the enum categories. The action is unmistakably distinct from the transactional siblings search_catalog, create_order, and check_order_status.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The conditional clause 'if a store exhibits...' gives an explicit trigger for when to invoke it, which is more than most descriptions offer. It stops short of stating when not to use it (e.g., prefer other categories, or avoid duplicate filings) or naming any alternative route, so it falls short of a 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search real-time physical retail store inventories near a shopper’s GPS location. Returns verified brick-and-mortar stores, in-stock products, unit prices, and Haversine distance in kilometers.
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | Maximum number of matching products to return (default: 10) | |
| query | Yes | Product name, category, or keyword (e.g. "coffee", "milk", "usb cable") | |
| customer_lat | No | Shopper latitude coordinate for proximity ranking (optional) | |
| customer_lng | No | Shopper longitude coordinate for proximity ranking (optional) | |
| max_distance_km | No | Maximum search radius in kilometers (default: 15) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the disclosure burden. It usefully states that results are real-time, verified brick-and-mortar stores with in-stock products, unit prices, and Haversine distance, though it omits operational details such as rate limits, auth needs, or pagination behavior.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two efficient sentences with the core purpose front-loaded and no wasted wording. It immediately communicates the search scope before listing return details.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With a fully described schema but no output schema, the description adequately explains the tool's purpose and return contents. A small gap remains around how proximity ranking behaves when optional customer_lat and customer_lng are omitted.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so all five parameters are already documented in the input schema. The description reinforces the GPS-proximity nature of the search but adds no syntax or format detail beyond what the schema provides, making the baseline 3 appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states a specific verb and resource: searching real-time physical retail store inventories near a shopper's GPS location. It clearly distinguishes this product/store search from sibling tools such as create_order and check_order_status.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies the use case through 'near a shopper's GPS location' and 'physical retail store inventories,' but it does not explicitly say when to use this tool versus alternatives or when not to use it. No sibling tool is named as a contrast.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v1.0.0check_order_statuscreate_orderreport_storesearch_catalogScored across 4 tools
Each tool has a clearly distinct purpose: catalog search, order creation, order status lookup, and store reporting. There is no meaningful overlap between discovery, checkout, fulfillment tracking, and trust-and-safety reporting.
All tool names use consistent snake_case and follow a verb-led pattern. The extra noun in check_order_status is still readable and consistent with the domain resource naming.
Four tools are well-scoped for a focused local-commerce workflow. Each tool earns its place and covers a distinct stage without bloat.
The surface covers search, order creation, status checking, and safety reporting, which are the core workflows. Minor gaps exist around order cancellation or listing existing orders, but agents can likely work around them via status checks and merchant checkout.
AI-agent product catalog: search, lookup & purchase routing over verified merchant data.
AI shopping gateway for product search, inventory, carts, and merchant-hosted checkout.
Search live store shelves, verify delivery windows, and run instant AI agent checkouts.
Verified local businesses, bookable by AI agents: services, prices, availability and appointments.
