For MCP authors
Sell MCP servers on RuleSell
Distribute your Model Context Protocol servers — stdio or SSE — to every MCP-compatible client. Real GitHub attribution, automated security scans, quality scoring, and Stripe Connect payouts after beta.
What we support
stdio transport
Local subprocess servers — the most common deployment. Node, Python, Go, Rust, anything that speaks MCP over stdin/stdout.
SSE & streamable HTTP
Hosted/remote servers that stream over Server-Sent Events or the newer streamable-HTTP transport. Great for shared team tools and SaaS integrations.
Env-var declaration
List required and optional environment variables in your manifest. Users see what they need to configure before installing — no surprises after paste.
Security scan
Every listing runs automated checks for hard-coded secrets, suspicious network endpoints, and known malicious patterns before it goes live.
Clients your MCP server reaches
Compatibility varies by client and protocol version — listings declare which clients they've been tested in.
Install snippet on listing
Every MCP listing renders a tailored install snippet for each supported client. For Claude Code, that looks like:
# ~/.claude/mcp.json
{
"mcpServers": {
"your-server": {
"command": "npx",
"args": ["-y", "@your-org/your-mcp-server"],
"env": {
"YOUR_API_KEY": "${YOUR_API_KEY}"
}
}
}
}FAQ
What is the Model Context Protocol (MCP)?
MCP is an open protocol Anthropic introduced for connecting AI models to external tools and data sources. An MCP server exposes tools, prompts, or resources that any MCP-compatible client (Claude Code, Cursor, Windsurf, Cline, Continue, Gemini CLI) can invoke at runtime.
Does RuleSell support stdio and SSE transports?
Yes. Listings declare their transport (stdio for local processes, SSE / streamable-HTTP for remote servers) so users see compatibility at a glance.
How do environment variables work for MCP listings?
Servers that need API keys, database URLs, or tokens declare required env vars in their listing manifest. RuleSell shows the env-var requirements upfront and never stores or transmits user secrets — users configure secrets locally in their client.
Is my MCP server safe to publish here?
All listings go through automated security scanning before they're published — checks for hard-coded secrets, suspicious network calls, and known malicious patterns. Authors who want manual review can request a security audit before publishing.
How does the install flow look for end users?
Each MCP listing shows a one-paste config block tailored to the user's client — for example, a JSON snippet to drop into ~/.claude/mcp.json for Claude Code, or the equivalent for Cursor/Windsurf. Copy, paste, restart the client.
Get your MCP server in front of every client
Free to publish during beta. Stripe Connect payouts open after beta.