---
title: "TronSave MCP Server: Let an AI Agent Buy TRON Energy (2026)"
description: "TronSave MCP server v1.3.0 (Nov 2, 2026): 53 tools, 18 with no login, OAuth sign-in, quotes, the price table and on-chain orders signed in your wallet."
url: https://blog.tronsave.io/tronsave-mcp-server-ai-agent-tron-energy/
canonical: https://blog.tronsave.io/tronsave-mcp-server-ai-agent-tron-energy/
lang: en-US
datePublished: 2026-08-20T17:47:23+00:00
dateModified: 2026-11-07T00:00:00+00:00
author: TronSave
publisher: TronSave (https://tronsave.io/)
categories:
  - Tron Guidelines
keywords: TronSave MCP server
translations:
  ru: https://blog.tronsave.io/ru/mcp-server-tronsave-ii-agent-energiya-tron/
  es: https://blog.tronsave.io/es/servidor-mcp-tronsave-agente-ia-energia-tron/
  pt: https://blog.tronsave.io/pt/servidor-mcp-tronsave-agente-ia-energia-tron/
  zh: https://blog.tronsave.io/zh/tronsave-mcp-fuwuqi-ai-zhineng-ti-nengliang/
  de: https://blog.tronsave.io/de/tronsave-mcp-server-ki-agent-tron-energie/
---

# TronSave MCP Server: Let an AI Agent Buy TRON Energy (2026)

*Updated: November 2, 2026 — checked against the live server, version 1.3.0 (adds OAuth sign-in).*

**Short answer:** TronSave runs a live **MCP server** at `https://mcp.tronsave.io/mcp`. On November 2, 2026 (version 1.3.0) it gave an AI assistant **53 tools** for the TronSave market. **18 of them work without logging in**, including quotes, the price table and an order paid from your own wallet. Your private key never goes near it: every signature happens in your wallet.

With it an assistant can price a rental, read the order book, place, cancel and extend orders, configure auto-buy and auto-sell, and manage API keys and alerts.

## What MCP is, in one paragraph

The Model Context Protocol is a standard way to hand an AI assistant a set of callable tools. Instead of the model guessing at an API from documentation, the server publishes typed tool definitions and the client calls them. Practically, it is what turns “explain how to buy TRON energy” into “buy me 65,000 energy for an hour, but check the price first.”

TronSave ships two separate integrations. The **MCP server** is the full one, aimed at desktop assistants and agent frameworks. The **WebMCP** integration is a small, deliberately read-only set of tools the web app offers to in-browser agents. They are covered separately below because their safety models are not the same.

## Endpoints

| Network | Endpoint |
| --- | --- |
| Mainnet | `POST https://mcp.tronsave.io/mcp` |
| Testnet | `POST https://mcp.tronsave.io/testnet/mcp` |
| Mainnet, OAuth sign-in | `POST https://mcp.tronsave.io/oauth/mcp` |
| Testnet, OAuth sign-in | `POST https://mcp.tronsave.io/oauth/testnet/mcp` |

The network is chosen by which endpoint you connect to — there is no network parameter to get wrong. All four respond to a standard MCP `initialize` handshake and advertise the same tool set, so you can develop an agent against the testnet and move it to mainnet by changing one URL.

In a client that reads a JSON config, that is all the setup there is:

```
{
  "mcpServers": {
    "tronsave": {
      "url": "https://mcp.tronsave.io/mcp"
    }
  }
}
```

**New in 1.3.0: OAuth sign-in.** Apps such as Claude, Claude Code and Cursor can connect to `/oauth/mcp` instead. The app opens a TronSave sign-in page where you enter an API key or sign with TronLink, so you never paste credentials into a config file. Step-by-step setup for each app: [How to Connect TronSave to Claude Code, Cursor and Claude](https://blog.tronsave.io/connect-tronsave-claude-code-cursor/).

## Do you need to log in?

Not for market data. The 1.0.0 server this guide first described required a session for every call; public market tools arrived in version 1.1.0, and since 1.2.0 every tool declares one of three access levels (in its `_meta["io.tronsave/auth"]` field), and you can check where you stand at any time with `tronsave_get_session_status`:

| Access level | Tools | What it unlocks |
| --- | --- | --- |
| Public, no login | 18 | Quotes, minimum price, order book, price table, market overview, system info, account resources, order suggestions, flash-unstake quotes, public order list, and `create_order` paid on-chain from your wallet. |
| API key | 9 | The internal-account tools: balance, deposit address, internal orders and extends, paid from your deposited balance. |
| Wallet signature | 26 | Your profile, order update and cancel, auto-buy and auto-sell settings, API keys, notifications, referrals, seller stats and bulk internal orders. |

The API key is sent with `tronsave_login {apiKey}` or as a client header `Authorization: ApiKey <key>`. The wallet-signature route works like this:

```
1. tronsave_get_sign_message        -> { message, timestamp }
2. (sign `message` in your own wallet, client-side)
3. tronsave_login { signature: "<signature>_<timestamp>" }
                                    -> { sessionId, walletAddress, ... }
4. the login is bound to this MCP connection; never put a session id in tool arguments
```

Note step 2. The server issues a message; *you* sign it in your wallet and send back only the signature. **Never send a private key to the server** — the protocol never asks for one, and any tool or client that does is not TronSave’s.

The API-key route is simpler and suits headless jobs, but it scopes you to the internal-account tools rather than the full platform set.

A public on-chain order still needs your wallet: `tronsave_create_order` with `paymentMethod=onchain` takes a TRX transfer that you sign yourself, to the deposit address in the quote. The server never signs for you.

## Two tool families

| Family | Count | Backed by | What it reaches |
| --- | --- | --- | --- |
| `tronsave_*` | 46 | GraphQL market | Market data, on-chain orders, and signature-session tools: cancel and update orders, auto-buy, auto-sell, API keys, notifications. |
| `tronsave_internal_*` | 7 | REST, api-key account | Estimate, create, history, details, order book and extend — all scoped to your deposited internal balance. |

The split matters for how an agent spends money. Internal-account tools draw on a balance you deposited in advance, which is a natural spending cap: an agent working through that family can only ever spend what you funded. Platform mutations go through your signature session instead.

## What do the 53 tools cover?

| Area | Tools |
| --- | --- |
| Session & auth | `get_sign_message`, `login`, `logout`, `get_session_status` |
| Market data (public) | `get_market_overview`, `get_price_table`, `get_system_info`, `get_account_resources`, `suggest_create_order` |
| Pricing & depth | `get_min_price`, `list_order_books`, `estimate_buy_resource`, `internal_order_book`, `internal_order_estimate` |
| Account | `get_user_info`, `get_user_permissions`, `get_internal_account`, `get_deposit_address`, `list_internal_transactions` |
| Orders (platform) | `create_order`, `get_order`, `list_orders`, `cancel_order`, `update_order`, `list_delegations` |
| Orders (internal) | `internal_order_create`, `internal_order_history`, `internal_order_details`, `bulk_create_internal_orders` |
| Extend | `list_extendable_delegates`, `list_extendable_receivers`, `internal_extend_delegates`, `internal_create_extend_request` |
| Auto-buy | `list_auto_buy_settings`, `create_auto_buy_setting`, `update_auto_buy_setting`, `delete_auto_buy_setting` |
| Selling | `sell_order_manual`, `register_auto_sell`, `update_auto_sell_setting`, `get_user_auto_setting`, `get_seller_stats` |
| Flash unstake | `list_flash_unstake_orders`, `estimate_flash_unstake`, `get_flash_unstake_request` |
| Alerts & programs | `list_notifications`, `mark_notifications_read`, `get_telegram_notify_setting`, `update_telegram_notify_setting`, `get_referral`, `get_agency_bot` |
| API keys | `generate_api_key`, `revoke_api_key` |

(All names carry the `tronsave_` prefix; it is dropped in the table for width.) The extend tools are worth knowing about before an agent starts buying repeatedly — [extending an existing delegation](https://blog.tronsave.io/extend-feature-extend-top-up-energy-without-ever-worrying-about-reclaim/) is usually cheaper than placing a fresh order.

Two of those deserve a flag before you let an agent loose. `revoke_api_key` immediately breaks every future request using the previous key, and `delete_auto_buy_setting` permanently stops all future executions of a rule. Both are documented as having exactly those side effects — they are not reversible by re-running the tool.

## How should an agent buy?

The server publishes its own playbook: a `buy_resource` prompt, an `account_overview` prompt, and a full guide as the resource `tronsave://guide/agent`. The buying flow it prescribes is four steps:

1.  `tronsave_estimate_buy_resource` returns the unit price, the total (`estimateTrx`, in SUN) and how much is available.
2.  Show the user the receiver, amount, duration, unit price and total in TRX, and get an explicit yes.
3.  `tronsave_create_order` with the price and payment amount from step 1.
4.  Confirm with `tronsave_get_order` or `tronsave_list_orders`.

Errors come back as `{"error":{"code","message","retryable","hint"}}`. An agent should follow the hint, never repeat a write when `retryable` is false, and check the order state before retrying after a timeout.

Use `estimateTrx` for the total rather than multiplying a unit price yourself: rentals of a day or longer are [priced per day](https://blog.tronsave.io/tron-energy-rental-cost-per-day-formula/), so a 30-day price from the price table is a daily rate.

## The freshness rule that will bite you

This is the single most important operational detail: **market prices and availability can change from one TRON block to the next, roughly every 3 seconds.** Re-run your estimate or order-book read *immediately* before creating or updating an order, and never reuse a quote more than a few seconds old.

It matters more with an AI agent than with ordinary code, because a model may spend several seconds reasoning between reading a price and acting on it. An agent that quotes, thinks, explains its plan, and then orders is working from a stale number. If you are writing the prompt, make “re-check the price right before you order” an explicit instruction rather than assuming it.

Two conventions to keep straight while you are at it: **all prices are in SUN** (1 TRX = 1,000,000 SUN) and **all durations are in seconds**.

## The browser side: WebMCP

The TronSave web app separately exposes a much smaller tool set to in-browser AI agents through the experimental `navigator.modelContext` API. The design here is conservative by intent. The production app offers six tools (October 2026):

-   **Overview** — what TronSave is, its key pages, and how to reach the API programmatically.
-   **Resource price** — the current estimated minimum unit price in SUN plus the estimated TRX cost for an amount. Read-only.
-   **Market overview** — energy and bandwidth available versus total supply, and the APY sellers earn. Read-only.
-   **USDT transfer cost** — the energy and bandwidth a number of USDT transfers need, the TRX burned without them, and the cost of renting the energy for one hour instead. Read-only.
-   **Open page** — client-side navigation to a known page. It never signs or sends a transaction.
-   **Prepare USDT transfer** — opens the Transfer USDT tool with the recipient and amount filled in. It does not send anything: you review the form and sign in your own wallet.

There are **no money-moving or signing tools** in the browser set — that is a deliberate line, not an oversight. Anything that spends requires explicit wallet interaction from you. The integration is also feature-detected, so on a browser without the API it does nothing at all, and it withdraws its tools when the page unmounts rather than leaving a stale set advertised.

## A sane safety model

If you are giving an agent access to something that spends money, the boundaries are worth being deliberate about:

-   **Start on the testnet endpoint.** Same tools, no real TRX.
-   **Confirm before any write.** The server’s own instructions tell agents to ask the user before calling any tool that is not read-only.
-   **Prefer the internal-account family for autonomous work.** A funded balance is a hard ceiling on what an agent can spend.
-   **Use a dedicated API key** you can revoke without disturbing anything else.
-   **Keep signing manual for anything large.** The signature stays in your wallet by design; do not engineer that away.

## FAQ

### Which AI clients can use this?

Any MCP client that can connect to a remote server over HTTP — Claude Desktop and Claude Code among them, plus agent frameworks with MCP support. The server speaks the standard protocol, so nothing TronSave-specific is needed on the client. Setup for Claude Code, Cursor and Claude, including OAuth sign-in: [How to Connect TronSave to Claude Code, Cursor and Claude](https://blog.tronsave.io/connect-tronsave-claude-code-cursor/).

### How would an agent find this on its own?

Through the discovery files TronSave publishes — `llms.txt`, an [API catalog, MCP server card and agent-skills manifest](https://blog.tronsave.io/llms-txt-well-known-agent-discovery-tronsave/) under `/.well-known/`. They point an agent at this endpoint without anyone pasting a URL.

### Do I need an API key, or is a wallet enough?

Neither, for quotes and market data: 18 tools are public. An API key unlocks the internal-account tools, and a wallet signature unlocks account management such as order updates, auto-buy, auto-sell and API keys. You can generate and revoke keys through the server itself.

### Can the server sign transactions for me?

No, and that is the point. Signing is always client-side. The server issues a message to sign and accepts the resulting signature — it never handles your key.

### Can I try it without spending real TRX?

Yes. Connect to the testnet endpoint instead of the mainnet one. Everything else about the integration is identical.

### Is market data available without logging in?

Yes. On the live server (version 1.3.0, checked November 2, 2026) quotes, the minimum price, the order book, the price table, the market overview and on-chain orders paid from your own wallet all work without a login. The 1.0.0 server this guide first described required a session for every call.

### Can an agent pay with x402 instead?

Not on TRON yet. As of October 28, 2026 the x402 TRON binding is still an open proposal; see [x402 on TRON: can AI agents pay with USDT yet?](https://blog.tronsave.io/x402-on-tron-ai-agent-usdt-payments/) Today an agent pays with an ordinary TRX transfer signed in the user’s wallet, or from a prepaid internal balance.

### How does this compare to the SDKs?

The [official SDKs](https://blog.tronsave.io/buy-tron-energy-with-code-tronsave-sdk/) are for code you write and control, and the [REST API](https://blog.tronsave.io/automate-buy-tron-energy-with-api-guide/) sits under both. MCP is for handing capability to an assistant that decides what to call. For a deterministic backend pipeline, use an SDK; for exploration, ops assistance, or an agent that reacts to conditions, MCP fits better. They all hit the same market.

## Bottom line

An AI assistant that can read the TronSave order book and place an order is a genuinely useful thing to have when you are running a payout pipeline or watching energy prices. The integration is built with the right instincts — signing stays client-side, the browser tools cannot spend, and the internal-balance family gives you a natural cap. Start on the testnet endpoint, teach your agent to re-check prices before it orders, and keep the irreversible tools out of reach until you trust it.

**Connect an agent:** point your MCP client at `https://mcp.tronsave.io/testnet/mcp` to experiment, then switch to the mainnet endpoint and watch it price against [the live energy market](https://tronsave.io/market).

*TronSave publishes this blog and operates the MCP server described here. Tool names, counts, and behaviour reflect server version 1.3.0 (checked November 2, 2026) and can change — call `tools/list` for the current set. Giving an AI agent access to tools that spend funds carries risk; start on the testnet and scope credentials tightly. This is not financial advice.*
