Archived — below our codifiability bar

ChatGPT AI Trading Bot Prompt Guide

Learn to prompt ChatGPT for AI trading bots and indicators. This guide covers specifying indicators like MACD, RSI, and Fibonacci for effective strategy generat

Published · Archived · Methodology: Mixed

  • Algo score: 45%
  • Discretionary score: 55%

This strategy was decoded from a public trading video but did not clear Strategy Decoder's codifiability bar: the extraction could not pin the rules down precisely enough to be turned into a reviewable specification. It is kept here as a reference post-mortem rather than as a strategy you can trade or backtest.

Part of: Moving Average Strategies

  • Methodology: Mixed
  • Content type: educational
  • Timeframes: Not specified, but the video advises to 'specify what time frame' to ChatGPT.
  • Markets: Not specified, but the video advises to 'specify which Equity' to ChatGPT.

Indicators mentioned

  • MACD
  • RSI
  • Moving Average
  • Fibonacci Sequence

Why this strategy was archived

Using a large language model to draft trading-bot code is now a common on-ramp for traders who want automation without writing everything from scratch. This entry decodes Trade Tactics' video "How To Make an AI Trading Bot With ChatGPT", a prompt-oriented walkthrough that shows how to ask ChatGPT for a bot built around familiar technical components — MACD, RSI, a moving average, and Fibonacci levels — while instructing the user to specify their own timeframe and equity in the prompt.

**Why this entry is archived.** Our extraction scored this video below the codifiability bar. The named indicators give the entry real structure, but the strategy layer stays open by design: the video's method is to hand the specifics — which timeframe, which market, which parameter values, and how the four components combine into a single entry or exit decision — to the person writing the prompt. What our extractor recovered is a prompt template with unbound variables rather than a rule set, and the conditions that are referenced could not be pinned down precisely enough to automate without guesswork. An entry that cannot be coded faithfully does not belong in the active catalog.

**What it still offers.** The video is useful as workflow material. It is explicit that a vague prompt produces a vague bot, and its core instruction — name your instrument, name your timeframe, name your indicators before asking for code — is the right discipline, and the same discipline our own extraction demands of a strategy. If you are experimenting with LLM-assisted automation, treat this as a guide to framing the request; then bring a fully specified strategy to it. The MACD, RSI and moving-average concept hubs, and the active catalog, list decoded entries where those rules were fully extracted and are ready to be specified.

Source video

Decoded from: How To Make an AI Trading Bot With ChatGPT by Trade Tactics — watch the original

Key timestamps:

  • 0:00 - Introduction to AI trading bots with ChatGPT
  • 0:12 - Importance of being specific in prompts
  • 0:19 - Using keywords for indicators (MACD, RSI, Moving Average)
  • 0:35 - Specifying timeframe and equity
  • 0:44 - Mention of Pine Script coding
  • 0:49 - Providing examples to ChatGPT
  • 0:53 - Example of a short entry condition (Fibonacci, MACD)
  • 1:00 - Example of a long entry condition (Fibonacci, MACD)

Frequently asked questions

Why is this ChatGPT trading bot guide archived?

The video names its indicators — MACD, RSI, a moving average and Fibonacci levels — but leaves the operational specifics to the user's own prompt: no fixed timeframe, market, parameter values, or defined logic for combining the signals. Our extraction scored it below the codifiability bar we require for the active catalog.

Is the video still worth watching?

Yes, as workflow material rather than as a strategy. It shows how to structure a prompt so an AI-generated bot reflects what you actually intend, including the instruction to state your timeframe and instrument explicitly.

Can ChatGPT build a trading bot from a video like this?

It can generate code, but only as specific as the rules you give it. The gap this entry illustrates is the same one automation always runs into: a strategy has to be defined before it can be coded, by an AI or by anyone else.

Where can I find codifiable MACD or RSI strategies on Strategy Decoder?

The MACD, RSI and moving-average concept hubs, along with the active catalog, list decoded video strategies with complete entry and exit rules extracted.

Other archived strategies

Browse all archived strategies · Explore the strategies that did make the cut