Can ChatGPT Pro place trades through Fensory?

Not through the current full MCP path. OpenAI states that Pro custom MCP apps support read and fetch permissions, while full write actions are currently limited to Business, Enterprise and Edu beta access.

Can ChatGPT agent mode use the Fensory app?

No. OpenAI states that agent mode does not use custom apps. Use a regular conversation with Fensory selected for the message that needs a tool call.

Can deep research place a trade?

No. Deep research can use a custom app for read and fetch actions only. It can inspect account or market information but cannot use the write action required for an order.

Does ChatGPT see my private key?

No. ChatGPT connects over OAuth. On the current MCP path, Hyperliquid trades use a scoped signing key that can sign trades and cannot withdraw, transfer or bridge funds.

Why start with a read-only prompt?

It verifies the account, plan, mode, pair and limits without creating an order. It also catches the common case where OAuth completed with a different Fensory account than the one you intended.

Agent Skill · version 0.1.0

Connect ChatGPT to a trading account with MCP

Verify a Fensory MCP connection, inspect Hyperliquid spot account state and prepare a bounded USDC-quoted spot order when write tools are available. Use for Fensory connection checks and supervised spot-order setup.

Download SKILL.md Read the guide

Download and review

Download the public file, review every instruction, then place it in the skills directory documented by the compatible client you use. Client-specific paths are omitted until each workflow is verified.

Download: curl -fsSL https://fensory.com/skills/chatgpt-trading-setup/SKILL.md -o SKILL.md

What the skill instructs

Read the account and instrument before preparing an action.
Keep the proposal inside the user's ceiling and the account's per-order cap.
Show the exact symbol, side, order type and size before a write call. Include the limit price for a limit order, or the current quote and expected execution conditions for a market order, then ask for confirmation.
Report a prepared order, submitted order and fill as three different states.

SKILL.md

---
name: fensory-spot-setup
description: Verify a Fensory MCP connection, inspect Hyperliquid spot account state and prepare a bounded USDC-quoted spot order when write tools are available. Use for Fensory connection checks and supervised spot-order setup.
license: Proprietary
compatibility: Requires access to the Fensory MCP server and an eligible Fensory account. Fensory is not available to US persons.
metadata:
  author: Fensory AG
  version: "0.1.0"
  source: "https://fensory.com/skills/chatgpt-trading-setup"
---

# Connect ChatGPT to a trading account with MCP

## When to use

This guide is for an eligible ChatGPT user who wants to verify a Fensory connection and understand whether the selected workspace, mode and account support read-only access or real trading actions.

It is not a workaround for ChatGPT plan restrictions. A successful read call on Pro does not mean the same connection can submit an order. The example stays read-only until every required write permission is present.

## Operating rules

- Start with read tools and identify the signed-in account, market, venue and instrument before preparing an action.
- Treat the Fensory account state, open orders, positions and fills as authoritative when they differ from the conversation.
- Fensory enforces three controls on the account: a kill-switch, a market allow-list and a per-order cap. Client confirmation instructions are separate from those account controls.
- Never describe a perpetual future as a share or tokenized equity. State the instrument and its risks before discussing an order.
- Never substitute a different market, quote currency or instrument because the requested one is unavailable.
- Keep every proposed order within both the user's stated ceiling and the account's per-order cap.
- Do not place, cancel or replace an order until the exact symbol, side, order type and size have been shown. For a limit order, show the limit price. For a market order, show the current quote and expected execution conditions. The user must confirm that exact action.
- Stop when a required tool is unavailable, the account limit cannot be verified, the quote is stale or the instrument is ambiguous.

## Risk and scope

This file is operational guidance, not financial advice. Trading can lose the full amount committed. Perpetual futures use margin and can be liquidated. Spot assets can lose value. Stop if the instrument, current quote or account controls cannot be verified.

## Procedure

1. **Confirm what the ChatGPT plan can do.** Pro custom MCP access is read and fetch only. Business, Enterprise and Edu currently receive full MCP write actions in beta. Check this before treating a successful connection as a trading setup.
2. **Read the spot account.** ChatGPT reads the available balance and any resting orders before it prepares another. This works as a useful connection test even when the workspace is read-only. Tools: `get_spot_account_state`, `get_open_orders`.
3. **Resolve the spot pair.** The tool identifies the actual Hyperliquid spot pair and whether it is USDC quoted. Fensory rejects agent orders where the per-order USD limit cannot be applied, so the quote currency is part of the safety check. Tools: `get_spot_metadata`.
4. **Read the fee schedule.** The account-specific venue fee and Fensory builder fee should be visible before confirmation. The public Fensory builder-fee range is 0.01% to 0.05% on Hyperliquid spot and perpetual futures. Tools: `get_user_fees`.
5. **Prepare or submit according to the mode.** A read-only workspace stops after showing the proposed arguments. In a workspace with full MCP, instruct ChatGPT to submit only after it has shown the exact action and you have confirmed it. The Fensory account controls still apply. Tools: `place_hl_spot_order`.

## Request template

Use the Fensory app to read my account only. Call get_spot_account_state, get_spot_metadata, get_user_fees and get_open_orders. Find the BTC/USDC spot market, state whether it is quoted in USDC, show my available balance, current open orders, applicable fees and per-order cap.

Do not place, cancel or replace an order. If this ChatGPT workspace and mode expose write actions, prepare an illustrative $10 BTC/USDC market order and show every argument that place_hl_spot_order would receive. Ask for my explicit confirmation for that exact action. If write actions are unavailable, say that clearly and stop. Never substitute a non-USDC pair or a perpetual future.

## Failure handling

- **The plan is read-only:** ChatGPT Pro can connect and read the account without exposing the write action. Treat that as a supported research workflow, not as failed OAuth and not as proof that trading is enabled.
- **The conversation is in the wrong mode:** Agent mode does not use custom apps. Deep research uses custom apps for read and fetch actions only. Use a regular conversation and select Fensory for the message.
- **OAuth used the wrong Fensory account:** The browser may already be signed in elsewhere. Compare the account balance and controls returned by the read tools before enabling a write action.
- **The pair is not USDC quoted:** The per-order USD limit cannot be applied to a non-USDC spot pair, so the backend rejects the agent order. The prompt requires a USDC pair and stops instead of substituting another market.
- **A successful submission is not a fill:** Market conditions, liquidity and venue behavior still matter. Read the fill and open-order records after any confirmed action before assuming the position changed.

## Response contract

Report the account used, the exact instrument, the tools called, the controls that applied and the current order or position state. Separate a prepared order from a submitted order, and a submitted order from a fill. If no action was taken, say so directly.

## Sources

- [OpenAI: developer mode and full MCP apps in ChatGPT](https://help.openai.com/en/articles/12584461-developer-mode-and-full-mcp-connectors-in-chatgpt)
- [Connect ChatGPT to Fensory](https://fensory.com/connect/chatgpt)
- [Fensory pricing](https://fensory.com/pricing)

Product behavior and client permissions verified 2026-10-08.

Questions people ask

Can ChatGPT Pro place trades through Fensory?: Not through the current full MCP path. OpenAI states that Pro custom MCP apps support read and fetch permissions, while full write actions are currently limited to Business, Enterprise and Edu beta access.
Can ChatGPT agent mode use the Fensory app?: No. OpenAI states that agent mode does not use custom apps. Use a regular conversation with Fensory selected for the message that needs a tool call.
Can deep research place a trade?: No. Deep research can use a custom app for read and fetch actions only. It can inspect account or market information but cannot use the write action required for an order.
Does ChatGPT see my private key?: No. ChatGPT connects over OAuth. On the current MCP path, Hyperliquid trades use a scoped signing key that can sign trades and cannot withdraw, transfer or bridge funds.
Why start with a read-only prompt?: It verifies the account, plan, mode, pair and limits without creating an order. It also catches the common case where OAuth completed with a different Fensory account than the one you intended.

Important information

Fensory does not provide investment advice or recommendations, does not generate buy or sell signals, and does not solicit any transaction. It supplies market data and software tooling only; all trading decisions are made by you or by an agent you configure and operate. Fensory’s fees, where charged, are published and paid by the user: a builder fee between 0.01% and 0.05% on perpetual futures and Hyperliquid spot, and no Fensory fee on tokenized equities, yields or bridging. The fee is set in advance and does not depend on the outcome of a trade.

Strategies run inside your own agent and execute only while that agent is running, only within the per-trade limits and allowed markets you set. Pre-IPO exposure is in development, with timelines subject to change. Third-party costs such as network fees and venue or issuer spreads apply. Tickers, prompts, strategies and execution examples shown on this page are illustrative only and are not recommendations; past performance of any strategy or trader does not guarantee future results.

All trading carries risk, including total loss of capital. Perpetual futures and leveraged products carry substantially higher risk and can lose more than the initial margin committed. Digital and tokenized assets may be volatile, illiquid, or subject to regulatory change.