# 10 Top Business CRAFT™ Prompts for ChatGPT-5

<figure><img src="/files/YgbcPCdCdCneewvHMbQs" alt=""><figcaption></figcaption></figure>

CRAFT™ is a five-part prompt that tells GPT-5 (and any other LLM for that matter) exactly what you want, why you want it and how to present it.

**Context** = The situation, background, and constraints

**Role** = The specific expertise needed

**Attributes & Actions** = The characteristics of the desired response

**Facts, Figures & Functions** = The relevant data and information

**Target Outcome** = The specific outcome needed

GPT-5 is very steerable when instructions are explicit, contradiction-free and paired with clear outcomes. Use low verbosity (i.e. be more concise) with minimal reasoning for routine tasks and raise them only for complex work.

When you need structured fields, supply a schema so the model returns valid JSON (getting techie now\... these practices are recommended in OpenAI’s GPT-5 docs and cookbooks).

**The CRAFT™ base prompt you can reuse:**

Choose your GPT-5 mode (Instant / Thinking Mini / Thinking / Pro) and paste this at the top of any prompt, then fill in the blanks.

**Context** =\
Organisation: \[who we are]\
Audience: \[who this is for and what they care about]\
Constraints: \[brand, tone, legal, length, UK English]

**Role** =\
Act as: \[job title] for \[team or function] with authority over \[scope]

**Attributes + Actions** =\
Style: clear, plain UK English, concise\
Workflow: plan → do → quality check → finalise\
Verbosity: \[low | medium | high]\
Reasoning effort: \[minimal | standard | elevated]\
If data is missing: list assumptions and proceed

**Facts & Figures + Functions** =\
Inputs: \[paste notes, links, data]\
Structured output: \[schema or “not required”]\
Do not invent facts beyond Inputs

**Target Outcome** =\
Deliverable: \[what you need and why]\
Format: \[Markdown sections | email | table | JSON schema]\
Success criteria: \[3 to 5 checks]\
Quality check: confirm criteria met, note gaps<br>

**How to Use This Resource:**

* Search for the category you’re looking for
* Copy/paste the prompt into your LLM
* Attach all required documents - or as many as you can
* Complete anything in square brackets as it relates to you or your business
* Refine the prompts to make them your own as needed


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://bob-e-ai-docs.gitbook.io/craft-prompting-for-ai/10-top-business-craft-tm-prompts-for-chatgpt-5.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.