MCP Overview

FloImg provides an MCP (Model Context Protocol) server that lets AI agents generate and manipulate images as part of their workflows.

What is MCP?

MCP is a protocol that enables AI assistants like Claude to use external tools. With FloImg’s MCP server, AI agents can:

• Generate images with AI (DALL-E, Stability, Ollama)
• Create charts, diagrams, and QR codes
• Apply precise transforms (resize, caption, adjust colors)
• Save to local storage or cloud (S3, R2)

Why FloImg + MCP?

AI agents benefit from composable image pipelines. FloImg lets agents:

Capability	What It Enables
AI Generation	Generate creative images via DALL-E, Stability, or local models
Deterministic Transforms	Resize to exact dimensions, adjust colors precisely, add captions
Mixed Pipelines	Generate with AI → transform deterministically → save to cloud
Session State	Reference previous images by ID for iterative refinement

The power is in composition: agents can generate creatively, then apply precise transforms—all in one pipeline.

Quick Start

1. Install the MCP Server

npm install -g @teamflojo/floimg-mcp

2. Configure Your AI Client

Add FloImg to your MCP configuration:

{
  "mcpServers": {
    "floimg": {
      "command": "npx",
      "args": ["-y", "@teamflojo/floimg-mcp"]
    }
  }
}

3. Use with AI

The AI can now use FloImg tools:

“Create a bar chart showing Q1-Q4 revenue: $12k, $19k, $15k, $25k”

The agent will call the FloImg MCP tool to generate the chart.

Available Tools

The MCP server exposes these tools to AI agents:

Tool	Description
floimg_generate	Generate images from parameters
floimg_transform	Modify existing images
floimg_save	Save images to storage

See Also

• Claude Integration - Setup with Claude
• Tool Reference - All MCP tools
• Examples - Common use cases