Staxly MCP Server — AI-Powered Developer Platform Data

What is the Model Context Protocol (MCP)?

The Model Context Protocol is an open standard created by Anthropic that allows AI assistants to interact with external tools and data sources in a structured way. Instead of relying on training data that may be outdated, an MCP-enabled agent can call tools exposed by an MCP server to fetch live, verified information.

The Staxly MCP server exposes the entire Staxly developer-platform catalog through this protocol. Any MCP-compatible client can query platform pricing, compare quotas, check feature availability, and estimate monthly costs — all backed by data sourced from official vendor documentation.

Quick Install

The fastest way to try the Staxly MCP server is with npx, which downloads and runs it without a global install:

npx staxly-mcp

To install globally so it's always available on your machine:

npm install -g staxly-mcp

Compatible Clients

The Staxly MCP server works with any client that supports the Model Context Protocol. Here are setup instructions for the most popular ones.

Claude Code

Claude Code is Anthropic's agentic CLI for Claude. To add the Staxly MCP server as a persistent tool source, run:

claude mcp add staxly --scope user -- npx -y staxly-mcp

Once added, Claude Code can query platform data in any conversation without additional setup.

Cursor

Cursor is an AI-powered code editor. Add the following to your Cursor MCP configuration file (.cursor/mcp.json in your project or global config):

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

Zed

Zed is a high-performance code editor with built-in MCP support. Add the server to your Zed settings under the MCP servers section using the same configuration format shown above.

Windsurf

Windsurf (formerly Codeium) supports MCP servers. Use the same JSON configuration to connect to the Staxly catalog from within the Windsurf editor.

Other MCP Clients

Any client implementing the Model Context Protocol can connect to the Staxly server. The command is npx -y staxly-mcp and it communicates over stdio using the standard MCP JSON-RPC transport.

What Data Is Exposed

The Staxly MCP server provides access to the full Staxly developer-platform catalog. All data is sourced from official vendor documentation and structured into a queryable database. The server exposes the following categories of information:

• Platform pricing — monthly and annual pricing tiers, including free tiers, with exact USD amounts and what's included at each level.
• Quotas and limits — storage caps, bandwidth limits, API rate limits, MAU limits, function invocation caps, and overage rates.
• Feature availability — which platforms offer specific features like vector search, edge functions, real-time subscriptions, SSO, or custom domains.
• Side-by-side comparisons — structured comparison data for any combination of 2 to 6 platforms in the catalog.
• Cost estimates — projected monthly cost based on your usage parameters (MAU, storage, bandwidth, function calls).
• CLI/SDK/REST operations — available commands and API calls for each platform.

MCP Tools Reference

The server exposes eight tools that your AI agent can call. Each tool returns structured JSON data that the agent can reason about and present in natural language.

• list_platforms — Browse the full catalog or filter by category (e.g., "baas", "auth", "ai-api"). Returns platform names, slugs, taglines, and category.
• get_platform — Retrieve full details for a single platform including all pricing tiers, features, quotas, and metadata.
• compare_platforms — Compare 2 to 6 platforms side-by-side with aligned pricing tiers, feature matrices, and quota tables.
• get_pricing — Get detailed pricing information: tier names, monthly and annual USD amounts, and what quotas are included at each tier.
• get_limits — Retrieve quotas, rate limits, storage caps, bandwidth allowances, and overage rates for a platform.
• search_features — Find all platforms that offer a specific feature, such as "vector search", "edge functions", or "SOC 2 compliance".
• list_operations — List available CLI commands, SDK methods, or REST API endpoints for a platform.
• estimate_cost — Project the monthly USD cost for a platform based on your expected usage: MAU, storage (GB), bandwidth (GB), and function invocations.

Example Queries

Once the Staxly MCP server is connected, you can ask your AI agent questions like these in natural language. The agent will automatically call the appropriate MCP tool and return a structured answer.

• "Compare the monthly cost of Supabase, Firebase, and Convex for 20,000 MAU with 10 GB storage."
• "Which BaaS platforms have native vector search on their free tier?"
• "What are the rate limits for the Anthropic API on the free tier?"
• "Show me every CLI command available in Supabase."
• "I need auth with social login, MFA, and under $50/month for 10k MAU. What are my options?"
• "Compare Stripe vs Paddle vs Lemon Squeezy for a SaaS selling in Europe — who handles VAT?"
• "List all platforms in the AI APIs category sorted by free tier limits."
• "Estimate my monthly cost on Vercel if I expect 50k page views, 5 GB bandwidth, and 100k serverless function invocations."

How It Works

The Staxly MCP server runs locally on your machine as a stdio-based process. When your AI agent needs developer platform data, it sends a JSON-RPC request to the server, which queries the Staxly catalog database and returns structured JSON. The agent then interprets this data and presents it in natural language.

Because the data comes from a curated database rather than LLM generation, numbers are accurate and verifiable. Every pricing figure, quota, and feature flag can be traced back to the vendor's official documentation.

Servidor MCP de Staxly — Datos de plataformas para agentes de IA

El servidor MCP de Staxly permite que tu agente de programación con IA consulte datos estructurados y actualizados de plataformas para desarrolladores: precios, cuotas, límites, características y comparaciones lado a lado, directamente desde la documentación oficial de cada vendor.

MCP (Model Context Protocol) es un estándar abierto creado por Anthropic que permite a los asistentes de IA conectarse a fuentes de datos externas. En lugar de inventar números, el agente consulta el catálogo de Staxly y obtiene datos reales.

Instalación rápida

Ejecuta el siguiente comando para iniciar el servidor MCP de Staxly sin instalación global:

npx staxly-mcp

Clientes compatibles

El servidor funciona con cualquier cliente compatible con MCP, incluyendo Claude Code, Cursor, Zed y Windsurf. Para Claude Code:

claude mcp add staxly --scope user -- npx -y staxly-mcp

Datos disponibles

El servidor expone el catálogo completo de Staxly con más de 137 plataformas en categorías como BaaS, autenticación, bases de datos, hosting, APIs de IA, pagos, CI/CD y comercio electrónico. Incluye precios por tier, cuotas del tier gratuito, límites de API, comparativas y estimaciones de costo mensual.

Ejemplos de consultas

• "Compara el costo mensual de Supabase, Firebase y Convex para 20.000 MAU y 10 GB de almacenamiento."
• "¿Qué plataformas BaaS tienen búsqueda vectorial en su tier gratuito?"
• "Necesito autenticación con login social, MFA y menos de $50/mes para 10k MAU. ¿Cuáles son mis opciones?"
• "Estima mi costo mensual en Vercel si espero 50k visitas, 5 GB de ancho de banda y 100k invocaciones serverless."