Firestarter MCP Server
Enables AI assistants to search real products, purchase with user approval, track deliveries, and run a seller business end to end via the Firestarter commerce platform.
README
Firestarter MCP Server
The Model Context Protocol server for Firestarter — AI-native commerce execution. Lets an AI assistant search real products, buy them with your approval, track deliveries, and run a seller business end to end.
Distributed two ways:
| Surface | How |
|---|---|
| Claude Desktop | Install the .mcpb desktop extension (npm run build:mcpb) |
| Any MCP client | Point it at https://api.firestarter.network/mcp with a Bearer API key |
Quick start
npm ci
npm run build # builds the inline shopping UI, then compiles
npm test
npm run build:mcpb # produces mcpb/dist/firestarter.mcpb
Set FIRESTARTER_API_KEY (fs_live_... for live orders, fs_test_... for
sandbox) to run the stdio server. Get a key at
firestarter.network/dashboard.
Safety model
Purchases never complete without an explicit approval step: firestarter_execute
finds and prices options, and only firestarter_approve charges. Tools that
move money or delete records are annotated destructiveHint: true so MCP hosts
prompt before running them; read-only tools are annotated so they don't.
Privacy Policy
Full policy: https://firestarter.network/privacy · Questions: privacy@firestarter.network
This section covers the Firestarter Desktop Extension (apps/api/mcpb), which runs the
Firestarter MCP server locally and calls the Firestarter API on your behalf.
What is collected. The extension holds one credential you supply at install time: your Firestarter API key. Claude Desktop stores it in your operating system's keychain, not in plaintext config. Beyond that, the extension sends only what a request needs — the product search or listing you ask about, and, when you approve a purchase, the order and delivery address tied to it. It reads no files and collects no telemetry of its own.
How it is used and stored. Requests go over TLS to https://api.firestarter.network
(overridable via the api_url config for sandbox use) and are handled under the policy
linked above. Order, listing, and account records are stored server-side to run the
service. Payment card numbers are never stored by Firestarter; card details are handled by
the payment processor, and stored secrets are kept as cryptographic hashes.
Third-party sharing. Data is shared only where completing the transaction requires it — principally the merchant fulfilling your order, plus payment and shipping providers. It is not sold, and there is no third-party advertising or tracking.
Retention. Data is retained while your account is active; execution logs are retained for 90 days. You may export your data or request deletion of your account from the dashboard, or by writing to the address above.
Your control. Uninstalling the extension removes the stored API key from your keychain and ends its access. Revoking the key in the Firestarter dashboard has the same effect immediately, from any device.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.