# Theta MCP for AI Agents AI Agent Connect

> Theta MCP lets you manage decentralized infrastructure and blockchain data directly through your AI client. Use it to deploy AI models on EdgeCloud, query the Theta blockchain, or handle video transcoding and livestreams. It connects your agent to the Theta Network ecosystem for web3 and decentralized media workflows.

## Overview
- **Category:** image-video
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_ayHy8fShiiwAaeLJrArF1vRpm3WcKWsYnohUDBGr/ai-agent-connect
- **Tags:** theta-network, edge-computing, web3, decentralized-video, gpu-rendering

## Description

Imagine you're trying to manage a fleet of decentralized GPU nodes for your AI models. Usually, that means jumping between different dashboards, checking resource availability, and manually starting or stopping instances. This Connector changes that by letting your AI agent handle the heavy lifting. You can ask it to find the best VM resources for a specific task, launch a new deployment, or check the status of your existing nodes without ever leaving your code editor. It also handles the nitty-gritty of the blockchain. Instead of manually parsing calls to find account balances or block details, you just ask the agent to do it. If you're running a video platform, you can manage upload URLs and livestreams through simple commands. It takes all those fragmented pieces of the network and puts them into a single interface for your AI. When you use this through the Vinkius catalog, you get a direct line to decentralized infrastructure that actually scales with your projects.

## Tools

### rpc_call_smart_contract
Simulate a smart contract execution locally. This lets you test your code before you actually commit it to the blockchain.

### video_list_ingestors
See all available Edge Ingestors. Use this to find the best points for receiving your video data.

### ai_create_deployment
Launch a new AI model or a Jupyter Notebook. This is your primary tool for starting new compute tasks.

### ai_delete_deployment
Remove an AI deployment from your project. Use this to clean up your environment and stop billing.

### ai_list_deployments
See all current AI project deployments in one list. This helps you keep track of what is running on the network.

### ai_list_standard_templates
View the built-in AI templates available for use. Use this to see what pre-configured environments you can launch.

### ai_list_vm_resources
Check available machine types and GPU resources. This helps you find the right hardware for your specific AI workload.

### ai_start_deployment
Restart a stopped GPU node to get it running again. Use this to quickly resume a paused AI model.

### ai_stop_deployment
Shut down a running GPU node to save resources. This is useful for pausing tasks when you don't need them active.

### cli_list_keys
See your local account addresses in one list. It gives you a quick overview of your local Theta identities.

### cli_lock_key
Secure your local accounts by locking them. This adds a layer of security to your local key management.

### rpc_broadcast_raw_transaction_async
Submit a signed transaction in the background so you don't have to wait for a response. This keeps your agent moving.

### rpc_broadcast_raw_transaction
Submit a signed transaction immediately and wait for the confirmation. Use this when you need to know the result right away.

### rpc_get_pending_transactions
View transactions that are still in the mempool. Use this to see what is about to be finalized on the network.

### rpc_get_transaction
Get the details for a specific transaction hash. This helps you debug or verify the status of a specific transfer.

### rpc_get_version
Check the current version of the Theta blockchain. Use this to ensure your client is up to date with the network.

### video_create_livestream
Set up a new livestream object. This is the first step in starting a decentralized broadcast.

### video_create_upload_url
Generate a unique URL for video uploads. Use this to allow users to send content directly to the Theta network.

### video_get
Fetch the details for a specific video. This helps you retrieve metadata or status for stored content.

### video_list
List all videos associated with a service account. This gives you a full view of your video library.

### video_search
Find videos using specific metadata. Use this to quickly locate content in a large decentralized library.

### video_select_ingestor
Assign a specific ingestor to a live stream. This helps you optimize the path for your incoming video data.

### video_transcode
Convert a video into a different format. This is essential for ensuring your content plays across all devices.

### video_unselect_ingestor
Release an ingestor when it's no longer needed. Use this to free up resources on the network.

### cli_new_key
Create a new encrypted account on your local machine. Use this to set up a fresh, secure identity for the network.

### cli_send
Transfer Theta or TFuel between accounts. This makes it easy to move funds between your blockchain identities.

### cli_unlock_key
Unlock a local account so you can send funds. This is the first step before performing any outgoing transactions.

### rpc_get_account
Pull the specific details for a Theta account. This lets you verify balances and account status instantly.

### rpc_get_block_by_height
Fetch the finalized block at a specific height. Use this to check the history of the blockchain at a specific point.

### rpc_get_block
Retrieve block details using a specific hash. This is perfect for inspecting the contents of a particular block.

## Prompt Examples

**Prompt:** 
```
What's the current version of the Theta blockchain?
```

**Response:** 
```
I've checked the network status. The current Theta blockchain version is **v4.2.0**.
```

**Prompt:** 
```
List the available GPU resources for my AI project.
```

**Response:** 
```
I found several options for your AI deployment on EdgeCloud:

| Machine Type | Resources | Notes |
| :--- | :--- | :--- |
| **Small** | 2 vCPU, 4GB RAM | Standard use |
| **GPU-Medium** | NVIDIA T4, 16GB RAM | Recommended for AI |
| **GPU-Large** | NVIDIA A100, 80GB RAM | High-performance |

Would you like me to launch one of these for you?
```

**Prompt:** 
```
Can you start a new AI deployment for me?
```

**Response:** 
```
I'm launching your new AI model now. I'll use the standard template to set up the environment on a **GPU-Medium** node. I'll notify you as soon as the deployment is live.
```

## Capabilities

### Launch AI models on decentralized GPU nodes
Start new AI deployments or Jupyter Notebooks on the EdgeCloud network.

### Query real-time blockchain data
Fetch account details, block heights, and transaction statuses from the Theta network.

### Manage video transcoding and livestreams
Handle video upload URLs, transcoding tasks, and livestream objects programmatically.

### Create and manage encrypted local keys
Generate and secure local account keys for blockchain interactions.

### Monitor EdgeCloud VM resource availability
See which machine types and GPU resources are currently available for deployment.

## Use Cases

### Scaling AI Inference
An engineer needs 10 GPUs. They use ai_list_vm_resources to find the right tier and then ai_create_deployment to launch them on the network.

### Web3 Data Fetching
A developer needs to check a transaction. They ask the agent to use rpc_get_transaction to pull details from the Theta blockchain.

### Video Streaming
A content creator needs a new stream. They use video_create_livestream and video_select_ingestor to set up a decentralized broadcast.

### Node Management
An operator needs to clear some resources. They use ai_stop_deployment to pause nodes and ai_list_deployments to see what's still running.

## Benefits

- Deploy models faster by using ai_create_deployment to launch Jupyter Notebooks or AI models on decentralized hardware.
- Save time on data fetching with rpc_get_block_by_height and rpc_get_account for instant blockchain status.
- Automate video workflows by using video_transcode and video_create_upload_url to handle content at scale.
- Manage local security with cli_new_key and cli_lock_key to keep your Theta accounts safe while interacting with the network.
- Optimize costs by using ai_list_vm_resources to find the cheapest GPU nodes for your specific AI workloads.

## How It Works

The bottom line is you get a direct command line for the Theta Network inside your AI client.

1. Provide your Theta RPC URL and EdgeCloud or Video API credentials to the Connector.
2. Ask your AI agent to perform tasks like listing deployments or querying blockchain blocks.
3. Receive real-time status updates and confirmation of successful actions from the network.

## Frequently Asked Questions

**What can I do with the Theta MCP?**
You can use it to manage decentralized AI models, query blockchain data, and handle video streaming all through your AI client.

**How do I manage AI models on Theta?**
The Connector allows your agent to find available GPU resources and launch deployments for AI models or Jupyter Notebooks automatically.

**Can I use this for video streaming?**
Yes, you can manage livestreams, create upload URLs, and handle video transcoding using the Theta video infrastructure.

**Does this Connector help with blockchain data?**
It provides direct access to query account balances, block details, and transaction statuses on the Theta blockchain.

**How do I secure my Theta accounts?**
You can use the Connector to create new encrypted local keys and lock them to keep your account information secure.

**Can I use this to find specific GPU nodes?**
The Connector can list all available machine types and GPU resources currently available on the EdgeCloud network.

**How can I check the balance and sequence of a specific Theta wallet?**
Use the `rpc_get_account` tool by providing the wallet address (0x...). The agent will return the coin balances and the current sequence number for that account.

**Can I manage AI model deployments on Theta EdgeCloud?**
Yes. You can use `ai_list_deployments` to see active ones, and `ai_create_deployment` or `ai_stop_deployment` to control your AI infrastructure resources.

**How do I search for existing video assets in my Theta library?**
Simply use the `video_search` tool. You can provide a query string to filter through your uploaded and transcoded video content.