# Eden AI MCP for AI Agents MCP

> Eden AI connects your agent to an entire ecosystem of large language model providers. It gives you one central place to manage all your automation workflows, track API usage across services like OCR and image generation, and monitor costs instantly from multiple sources (OpenAI, Google, AWS, etc.). Stop checking dozens of dashboards; get unified control over your AI stack right where you work.

## Overview
- **Category:** ai-frontier
- **Price:** Free
- **Tags:** unified-api, llm-orchestration, ai-providers, model-management, api-usage-tracking, workflow-automation

## Description

Building complex applications with LLMs means dealing with a messy web of APIs, billing cycles, and provider limits. Eden AI solves that headache by unifying the entire AI infrastructure into one conversational interface. You can manage multi-step workflows—from generating text to extracting data or creating images—all without leaving your chat window. Instead of building separate pipelines for every service, you tell your agent what job needs doing, and it routes the request to the best provider available. It also gives you a real-time ledger: check credits, monitor costs over time, and see exactly which model is driving up your spending. If you use Vinkius, connecting Eden AI through the catalog makes sure that all these powerful capabilities are accessible instantly via any MCP-compatible client.

## Tools

### check_credits
Quickly checks your current credit balance within the Eden AI platform.

### universal_ai
Runs specialized AI tasks like OCR, translation, or image generation across providers.

### list_embedding_models
Lists all available embedding models provided by Eden AI.

### chat_completion
Generates a chat response using the best multi-provider routing, letting you select or specify the model used.

### monitor_consumption
Provides detailed reports on your API usage and associated costs over specified time periods.

### create_embedding
Generates text embeddings using the Eden AI platform for vector databases.

## Prompt Examples

**Prompt:** 
```
What are my remaining credits, and how much did I spend on embeddings today?
```

**Response:** 
```
**Eden AI Usage Report**

*   **Current Credit Balance:** $78.55
*   **Today's Consumption:** $3.12
*   **Breakdown:**
    *   `create_embedding`: 1,200 units ($1.80)
    *   `chat_completion` (Google): 4 calls ($1.32)
```

**Prompt:** 
```
Compare the pricing for OCR and translation between the three main providers.
```

**Response:** 
```
Here is the service level data I pulled:

| Feature | OpenAI Pricing | Google Pricing | AWS Pricing |
| :--- | :---: | :---: | :---: |
| **OCR** | $0.01 / 1k chars | $0.009 / 1k chars | N/A |
| **Translation** | $0.02 / 1M chars | $0.015 / 1M chars | $0.025 / 1M chars |

Google looks like the leader for cost-effective translation right now.
```

**Prompt:** 
```
List all available embedding models and tell me which ones are best for financial data.
```

**Response:** 
```
I checked the registry. We support six main types:

*   `text-general`: Best for standard articles.
*   `finance-v2`: **Recommended.** Optimized for transactional data structures.
*   `legal-compliance`: Great for policy documents.

The `finance-v2` model is currently available across all three major providers and shows high accuracy rates.
```

## Capabilities

### Monitor API Costs
Get an immediate snapshot of your credit balance and monitor consumption trends across different services.

### Execute Specialized AI Tasks
Run complex, multi-modal tasks like translating text, reading images via OCR, or generating artwork using one command.

### Generate Embeddings
Create vector embeddings from any piece of text to power search and retrieval systems.

### Route Chat Completions
Send prompts and let the agent choose the optimal provider or model for the best chat response.

### List Model Capabilities
Check which specific large language models are supported by the platform, along with their feature set.

## Use Cases

### Auditing Quarterly AI Spending
An MLOps engineer needs a full picture of spending. They ask their agent to use `monitor_consumption` for the last quarter, getting a report that breaks down costs by feature and provider.

### Building a Multi-Modal Chatbot
A developer needs a chatbot that can not only answer questions but also read receipts from uploaded images. They use `universal_ai` to integrate OCR alongside standard chat completion calls.

### Selecting the Best Model for a New Feature
A Product Manager wants to know if Google or OpenAI is cheaper for sentiment analysis. The agent uses provider intelligence (via `list_embedding_models` and pricing data) to compare costs before implementation.

### Ensuring Sufficient Funds for a Launch
An operations team lead needs to know if the current budget covers the deployment. They run a quick check using `check_credits` before initiating any large-scale testing.

## Benefits

- Cost Visibility: Use `monitor_consumption` to track exactly how much every feature costs, eliminating surprise bills.
- Unified Access: Don't switch consoles. Your agent handles specialized tasks like OCR or image generation using the `universal_ai` tool.
- Model Selection: Need a specific chat style? The `chat_completion` tool lets you route requests to different providers and models instantly.
- Budget Control: Keep track of your operational spending anytime by calling `check_credits` for an immediate balance report.
- Data Readiness: Use the `create_embedding` function to generate high-quality vector data without leaving the chat interface.

## How It Works

The bottom line is that you talk to your AI client like a coworker, and it handles all the underlying complexity of routing, billing, and execution for every major AI provider.

1. Connect your AI client to Eden AI on Vinkius and authorize it using your API key.
2. Use natural conversation to define complex operations, such as 'list all active workflows' or 'show me the cost of image generation'.
3. The agent executes the command, providing real-time data, pricing comparisons, and operational status updates.

## Frequently Asked Questions

**How does Eden AI help me compare LLM pricing in real time?**
Eden AI gives you immediate, comparative pricing reports for features like sentiment analysis or image generation. You don't need to check three different vendor websites; the agent pulls and organizes all the data instantly.

**Can Eden AI handle workflows that use multiple kinds of AI services?**
Yes, it manages everything from simple text chat completions to complex media tasks like OCR and image generation. You tell your agent the goal, and it executes all necessary steps in sequence.

**What if I need to switch providers for a single task?**
The MCP handles provider routing automatically. If you ask for chat completion, it can route that request to OpenAI or Google based on your criteria, letting you test which model performs best.

**How do I track my spending across different departments?**
You use the monitoring tool to get a comprehensive audit of API consumption. You can filter usage reports by feature and see exactly which department or project is driving up costs.

**Is Eden AI only for text-based AI tasks?**
No, it supports multi-modal inputs. You can use the platform to analyze images (OCR) and generate multimedia content, not just process pure text prompts.