Skip to content
How the Stack Planner works

From idea to installed stack in 3 steps

No more digging through marketplaces — describe what you want and ship it.

1

Describe your project

One line is enough. Examples: 'a SaaS with Postgres and Stripe', 'a chatbot on Azure with MongoDB', 'an ML pipeline tracking experiments'.

2

AI designs your stack

Watch a complete plan stream in: architecture diagram with brand icons, MCP servers from 3,200+, matching skills, and step-by-step setup.

3

Install with one command

Run `npx mcpflix install <id>` — it writes claude_desktop_config.json, prompts for any API keys, and drops your skills into ~/.claude/skills. Restart Claude and you're done.

What makes the Stack Planner different

Other planners output a tool list. MCPFlix outputs architectural advice.

Every plan ships with three advisory outputs nobody else surfaces — grounded in published research on multi-agent systems and Claude Code workflows.

Sub-agent topology

>90% lift on complex tasks Anthropic research

When your workflow spans 5+ MCPs or several services, the planner recommends a Lead (Opus) + specialised workers (Sonnet for code, Haiku for retrieval) + Critic — with model routing baked in so token cost stays sane.

Why orchestrator-worker beats single-agent

Five-tool carpentry

Prompts • Skills • Projects • Subagents • MCP

Every Claude affordance has a job. The planner classifies which surface fits each part of your workflow, so you stop confusing Skills with Projects and Prompts with Subagents.

Read the framework

Deterministic last mile

Money • HIPAA • production writes

Stripe charges through an LLM are not testable with "call X, expect Y". The planner flags financial, compliance-bound, and system-of-record writes — and steers the irreversible step back into deterministic code.

When MCP isn't enough
Try the Stack Planner

Free. One plan per day as a guest, five if you sign in.

New to MCP?

What is the Model Context Protocol?

MCP is an open standard created by Anthropic that lets AI assistants (Claude, GPT, Cursor) connect to your real tools — databases, APIs, cloud services, and more. Think of it as USB-C for AI: one protocol, any tool.

Instead of copy-pasting data into your AI, MCP servers give it direct, secure access to query your database, run your dbt models, send Slack messages, or manage your cloud infrastructure — all from a conversation.

Your AI Assistant

Claude, GPT, Cursor, Windsurf

MCP Protocol
Snowflake
dbt
Slack

Your AI talks directly to your tools — no copy-paste needed

Community stacks

What others are building with MCPFlix

Real stacks people generated and saved. Click any to install in one command.

I'm building an ML pipeline that fetches from Hugging Face and trac…

I'm building an ML pipeline that fetches from Hugging Face and tracks experiments in Weights & Biases

751/s/kecJZWqb

Un chatbot de atención al cliente con MongoDB, montado en AKS de Azure y…

Un chatbot de atención al cliente con MongoDB, montado en AKS de Azure y modelos de Azure AI Foundry

900/s/60X3PEM8

Build a Stripe billing automation that processes HIPAA-bound patient records…

Build a Stripe billing automation that processes HIPAA-bound patient records, syncs to Salesforce, and posts daily reports to Slack — needs GitHub Actions deployment

810/s/5wI4q41j

Build me a SaaS MVP stack with Postgres, Stripe payments, and Resend emails

Build me a SaaS MVP stack with Postgres, Stripe payments, and Resend emails

770/s/3XZ4HI0y

Construye un MCP server en Python con las siguientes especificaciones…

Construye un MCP server en Python con las siguientes especificaciones: Identidad y contexto Nombre del MCP: video-generator-mcp Empresa: EDIFICACIÓN (Salfacorp) Propósito: Transformar procedimientos técnicos (PDF o texto) en guiones JSON estructurados para generación de video corporativo de capacitación Stack técnico Lenguaje: Python Transport: stdio (para Claude Desktop) Imágenes: Gemini API (gemini-2.0-flash-preview-image-generation) — solicitar API key al usuario si no está configurada Renderer de video: Creatomate (API REST, JSON → video) Generación de guión: Anthropic API (claude-sonnet-4-20250514) Herramientas (tools) que debe exponer el MCP generate_video_script Input: procedure_text (string) — texto extraído del PDF o pegado directamente pdf_path (string, opcional) — ruta al PDF si se sube archivo Proceso interno: Extrae PROJECT_NAME desde el título o encabezado del procedimiento Fija COMPANY = "EDIFICACIÓN" Determina TEMPLATE desde el nombre del procedimiento (sin prefijo numérico) Analiza el contenido y determina cantidad de escenas dinámicas (libre, sin mínimo ni máximo fijo) Genera escenas: escena 1 (intro corporativa fija) + escenas dinámicas + escena final (cierre corporativo fijo) Para cada escena dinámica genera: voz_en_off, image_prompt, video_prompt Output: JSON estructurado según el formato definido abajo generate_scene_images Input: JSON del guión generado por generate_video_script Proceso: Para cada escena dinámica, llama a Gemini API con el image_prompt y retorna las rutas o base64 de las imágenes generadas Output: JSON del guión con campo imagen_generada poblado por escena render_video Input: JSON del guión con imágenes generadas Proceso: Mapea el JSON al template de Creatomate y llama a su API REST Output: URL del video renderizado

580/s/syRnouoe

Connect Claude to my Snowflake database and Slack for daily reports

I need to connect Claude to my Snowflake database and Slack for daily reports

540/s/ui0S9UvC

Stay in the loop

Updates on new servers, skills, and features.