@ekipnico/image-intel

AI-powered image content analysis using vision LLMs.

Installation

npm install @ekipnico/image-intel

Quick Start

import { createImageIntelMesh, ImageContentGenerator } from '@ekipnico/image-intel';

const mesh = createImageIntelMesh();
const generator = mesh.resolve(ImageContentGenerator);

const result = await generator.analyze(imageBuffer);

Model Configuration

Models can be specified in two ways:

1. String ID (Built-in Providers)

Use string IDs for OpenAI, Anthropic, and Google models:

// Via environment variable (recommended for defaults)
process.env.AI_MESH_DEFAULT_MODEL = 'gemini-2.0-flash';

// Or per-request
const result = await generator.analyze(imageBuffer, { model: 'gpt-4o' });

Built-in models: gpt-4o, gpt-4o-mini, gemini-flash, gemini-2.0-flash, claude-sonnet, etc.

2. LanguageModel (Any Provider)

Pass any Vercel AI SDK model directly for providers like Groq, Mistral, DeepSeek, etc:

import { createGroq } from '@ai-sdk/groq';

const groq = createGroq({ apiKey: process.env.GROQ_API_KEY });
const result = await generator.analyze(imageBuffer, {
    model: groq('llama-3.3-70b-versatile'),
});

import { createMistral } from '@ai-sdk/mistral';

const mistral = createMistral({ apiKey: process.env.MISTRAL_API_KEY });
const result = await generator.analyze(imageBuffer, {
    model: mistral('pixtral-large-latest'),
});

Environment Variables

| Variable | Description | |----------|-------------| | AI_MESH_DEFAULT_MODEL | Default model when not specified (default: gpt-4o) | | OPENAI_API_KEY | Required for OpenAI models | | ANTHROPIC_API_KEY | Required for Anthropic models | | GOOGLE_API_KEY | Required for Google/Gemini models |

API

generator.analyze(input, config?)

Analyzes an image and returns structured content.

Parameters:

• input - ImageInput (Buffer, URL, file path, or ImageInput object)
• config.model - ModelSpec (string ID or LanguageModel)
• config.language - Output language (default: 'en')
• config.context - Additional context for better analysis

Returns: ImageContent with identification, visual analysis, SEO metadata, and more.

Integrating into Existing Mesh

import { Mesh } from 'mesh-ioc';
import { registerImageIntelServices, ImageContentGenerator } from '@ekipnico/image-intel';

const mesh = new Mesh('MyApp');
// ... register your services ...
registerImageIntelServices(mesh);

const generator = mesh.resolve(ImageContentGenerator);