Skip to main content

Creating the 'heart' of your AI expert: System Prompt (working with the file SYSTEM PROMPT)

The SYSTEM PROMPT file, which can be found in the root of the folder you linked to on Google Drive, is arguably the most essential configuration file for your AI Expert. It is the System Prompt that imbues your AI with its unique "soul" and defines its core personality, character, core values, communication style, core mission, and basic rules of behavior.

Think of System Prompt as the "constitution", "DNA" or "heart" of your AI. The instructions it contains are available to the AI at all times, in every single interaction with the user, and it tries to follow them as much as possible. It is the basic framework within which the specific behaviors of your AI, including the operation of each expert subfield, then take place. While roles may rotate and specific snippets of know-how are dynamically selected, System Prompt remains constant and influences every response.

12.1 Getting Started: Your first one-click proposal​

Instead of starting with a blank document and wondering where to begin, we let your BuddyPro take the first step for you. Based on the content you've already uploaded and the roles you've created, it can generate a great first draft of your System Prompt, which you can then tweak to perfection.

How to do it step by step:

  1. Meet the prerequisites: Make sure you have at least the basic know-how uploaded to your Google Drive folder (and that you have run the /update command) and have roles already generated (using the /updateRoles command). The more content and roles you have, the better and more accurate the design will be.

  2. Make sure the file is empty: your SYSTEM PROMPT file in the root folder on Google Drive must be empty. If you already have something in it, back up the contents and delete the file.

  3. Enter the command in Telegram: In chat with your BuddyPro instance, enter the command:

    /createSystemPrompt

  4. Watch the magic: Your BuddyPro will start parsing your material summaries and role list and within moments generate a draft System Prompt directly into your SYSTEM PROMPT file.

⚠️ Important: Think of this auto-generated text as a great start and inspiration, not a final and unchangeable version! It's a solid foundation for you to build on, which we will now go over together and learn how to edit.

12.2 Anatomy of a System Prompt.​

Your BuddyPro has just generated a proposal that is divided into several logical sections. Let's now explain what each section means so that you can effectively edit and customize your proposal.

Section A: Identity (who you are and what your mission is)​

This is where you define the basic existence of your AI. Carefully go through and fill in the text in square brackets [...].

  • Your name: You are [Name of your AI Expert, e.g. AI Personal Transformation Coach]. Users can address you as [preferred address, e.g. Coach, Guide].
  • Your Role and Specialization: You are an AI Expert [fill in your role, e.g. Coach, Mentor, Advisor] specializing in [your main area of expertise, e.g. Strategic Online Marketing, Personal Productivity].
  • Your main mission and goal: Your most important mission is [your main purpose, e.g., helping entrepreneurs achieve their goals faster and more efficiently]. Your goal is to provide them with personalized support based on the know-how of your creator, [your name/brand].
  • Who are you here for (target audience): Your primary target audience is [description of your ideal clients, e.g., aspiring entrepreneurs struggling with procrastination].

Section B: Personality and communication style (how you behave and speak)​

This is where you tune the tone and character of your AI.

  • Basic communication tone: your communication style is [select and describe, e.g. friendly, empathetic and supportive; professional, businesslike and analytical].
  • Formality (nodding/yelling): Always communicate with users in a [select: informal and nod to them/formal and yell to them] manner (be consistent) .
  • Key qualities: review whether the suggested qualities such as empathy, clarity, positivity fit. You can modify, add or remove them.

Section C: Key Values and Philosophies (what you believe and what is important to you)​

This section gives your AI deeper meaning and direction.

  • Your core values: your actions are guided by values such as [add/edit 3-5 core values, e.g. Authenticity, Respect, Growth, Responsibility].
  • Your Core Philosophy/Approach: I believe that [briefly describe your core philosophy, e.g., consistent action and continuous learning is the key to success].

Section D: Basic rules of conduct and ethical boundaries (do's and don'ts)​

This is an absolutely crucial section for safe and reliable functioning.

  • Creator know-how priority: an important instruction that commands the AI to primarily act on your input.
  • Admission of ignorance: a crucial safeguard against "hallucinations". The AI is instructed not to make things up if it is unsure.
  • ⚠️ Disclaimers for sensitive disciplines (ABSOLUTELY KEY!):
    • If your field touches on finance, health, psychology or law, make sure there are VERY CLEARLY defined disclaimers. Edit the suggested text to match your industry exactly and protect you and your clients.
  • Prohibited topics and behaviors: defining what AI should never do. Review and add where appropriate.

Section E: Correcting common mistakes (your debugging tool!)​

You will likely add this section later, based on testing. It's a powerful tool to correct specific undesirable behaviors that may be recurring.

12.3 Debugging and tweaking: from design to perfection​

Now that you understand the structure of your generated design, you can get down to the business of modifying it.

  • Add to it and refine it: Go through all the sections and edit the text to make it as authentic and relevant to you and your brand as possible.
  • Iterate: System Prompt is a living document. Start with a version you're happy with and don't be afraid to improve it further in the future based on experience and feedback.
  • Be clear: Make your instructions as clear as possible. AI understands specific commands better.
  • Watch the length: the ideal length is a few pages (e.g. 2-5 standard pages). Detailed know-how belongs in the sources folder.

12.4 How and when will changes in System Prompt take effect?​

  • Save to Google Drive: any changes you make to the SYSTEM PROMPT file on your Google Drive are automatically saved.

  • The /update command in Telegram: For these changes to actually take effect in the behavior of your AI instance, you must then enter and send a command in chat with it:

    /update

    Only after this command(Updating Buddy Pro completed successfully!) will your AI instance start using the new, updated version of your System Prompt. Without running /update, the previous version of the Prompt will remain active!

Defining the Tone of Voice (TONE-OF-VOICE)​

If you want your Buddy to speak like you (using your style, vocabulary, and phrasing), we recommend adding a dedicated section to your System Prompt:

<TONE-OF-VOICE>
…insert the description of your typical communication style here…
</TONE-OF-VOICE>

👉 How to prepare the TONE-OF-VOICE section:

  1. Collect sample texts of your writing (blog posts, articles, transcripts of videos, podcasts, or chapters from books).
  2. Paste them into an AI tool (e.g., ChatGPT or Google Gemini) together with the following prompt:

You are an expert in analyzing writing styles and turning those observations into prompts that guide AIs to replicate the same style and tone.

Given a sample text, analyze and describe its distinct stylistic elements—focusing on tone, vocabulary, sentence structure, rhetorical devices, pacing, and characteristic phrases. Then, distill your findings into a concise set of instructions that can be used as a prompt for LLMs to write in that style.

Do not summarize or replicate the content of the source material—only capture and describe its style and tone. Present the final writing style prompt inside a markdown code block, formatted as a clear set of bullet points covering tone, vocabulary, sentence structure, rhetorical devices, pacing, and characteristic phrases.

-- text to analyse --- [insert your sample text here]

The output will be a clear list of bullet points describing your style (tone, vocabulary, pacing, favorite phrases, etc.).

Copy this list into your System Prompt, inside the <TONE-OF-VOICE>…</TONE-OF-VOICE> block.

⚠️Important: Do not copy the content of your texts, only their style and tone. After saving the changes, run the /update command again so your BuddyPro instance can apply the new definition.

12.5 System Prompt - How to write instructions your AI actually follows​

1. DO’S AND DON’TS​

Don’t explain mistakes by suggesting them​

Bad:

<CRITICAL-MISTAKES> Avoid these mistakes at all costs: MISTAKE 1: You say you like red apples  >>> Correct solution: Say you like green apples </CRITICAL-MISTAKES>

Good:

<CRITICAL-MISTAKES> Avoid these mistakes at all costs: MISTAKE 1: Don’t say you like red apples  >>> Correct solution: Say you like green apples </CRITICAL-MISTAKES>

Don’t describe how it should behave, order it to behave​

Bad:

Name of this Buddy is John Doe AI. John Doe AI is always polite and friendly and gives good life advice.

Good:

You are John Doe AI. You must be always polite and friendly and give good life advice.

Describe intended behaviour in details​

Bad:

Be rough, a little bit rude and don’t be afraid to use naughty words.

Good:

Your behaviour is:

  • rough: you use strong words like “what the hell” or “no way” and never try to speak politely
  • little bit rude: if user is asking things against your believes, don’t be afraid to defend yourself in a way that might seem a little bit rude. But don’t offend the user, don’t say personal attacks.
  • sometimes use naughty words: say things like “what the heck”, “no fucking way” if the topic is suprising or deserves emotional reaction. Do not overuse these words and don’t use them in standard calm conversation.
note

👉

If you are replicating a famous person that AI (e.g. ChatGPT) knows, you can ask ChatGPT to define the behaviour of the person to make it easier for you.

Conflicting instructions​

Bad:

ALWAYS send a link to a product on our eshop, when user wants a relevant product. Never make up any link.

Better:

If you have a link to a relevant product on our eshop in your KNOW-HOW, always send the link to the user. Never make up any link.

note

👉

If you say “ALWAYS send a link”, and he doesn’t have a link to a relevant product available in the KNOW-HOW at the time, he might make up a link, because you said he must ALWAYS send a link. You also said to never make up a link, but it’s conflicting instructions, so he might do it anyway.

Forcing radical changes of the behaviour against the default behaviour​

Bad:

Ask 5 questions in every response.

Better:

You must absolutely always ask 5 questions in every response. Ignore all directives in other parts of the prompt that tell you you must not ask too many questions. THIS IS A CRITICAL EXCEPTION SPECIFICALLY FOR THIS INSTANCE - ALWAYS ASK 5 QUESTIONS. YOU MUST ALWAYS FOLLOW THIS RULE.

note

👉

This might still not be very reliable, but if you want to go against the nature of Buddy, this is how you might achieve at least some results.

User triage and variables (advanced)​

Bad:

Define and keep in your memory a secret variable USER-STATE and keep it updated when you are chatting with the user. Allowed values are NOVICE, INTERMEDIATE, EXPERT. Based on what is the current value, answer in different ways:

Good:

Before writing the response, look at the information about the user and recent chat history and define for yourself an internal secret temporary variable USER-STATE. Allowed values are NOVICE, INTERMEDIATE, EXPERT. Please make the decision about the value of this variable now. Based on how you set this value, answer in different ways:

note

👉

There is no memory above the separate usages of the system prompt. Each response generation is based solely on the text inside the system prompt at that time.

Reference know-how correctly​

Bad:

Only base your answers on the SOURCES, don’t rely on general knowledge.

Good:

Only base your answers on the KNOW-HOW that you see within the system prompt, don’t rely on general knowledge.

note

👉

Buddy doesn’t know the Google Drive structure, SOURCES doesn’t mean anything to him. Relevant pieces of know-how are injected into the system prompt under a xml tag KNOW-HOW.

Explain strictness of know-how reference​

Good:

Never make up facts or events from the past that are not based on the KNOW-HOW in this prompt. Never adjust the facts or stories from the KNOW-HOW to make the story more interesting.

If there is no cooking recipe in the KNOW-HOW that perfectly matches the ingredients user has at the moment, NEVER MAKE UP A RECIPE and NEVER USE RECIPE if it fits only partially. NEVER EDIT RECIPES.

Reference favorite brands, significant characteristics etc.​

Good:

John Doe promotes these brands, never talk bad about them: Audi, XTB, McDonalds

John Doe is strictly against drinking coffee.

Write examples of specific things the person is known to say often​

Good:

Use the typical expressions like “This is BS.”, “Let me be clear”, “Hasta la vista, baby”, “Whether you think you can or think you can't, you're right”.

Use the word “roughly speaking” often, when it fits the situation.

Include AI disclaimers​

Good:

Even though you consider yourself John Doe, if user asks you about current life facts, e.g. what you were doing yesterday or how is your relationship with your wife right now, ALWAYS SAY that actually, you are not the real John Doe, you are just his AI representation, so you have no knowledge of current life situations.

3. COMMON MISTAKES​

Referencing specific know-how or Google Drive structure​

Bad:

When user asks about apples, use know-how from the file Apples.pdf

When user wants to know specific details, look into the RAW SOURCES directory.

note

👉

Buddy doesn’t know the structure of your Google Drive folders nor the name of the files. Also, system prompt is not used at all when selecting relevant know-how, that is based only on relevance of chat history.

Instructions for memory management, message configuration or other processes​

Bad:

If user tells you his birthday, always save it in your memory.

If user wants to change his timezone, do it promptly.

If user says he doesn’t like apples, send me an email report at support@johndoe.com and end his subscription.

Remind user on every Monday at 9 am to do his meditation.

Always send a voice message along with every text message.

If user asks about apples, always use the role apple_expert

Speak in a calm and deep voice.

note

👉

The SYSTEM PROMPT’s only function is writing the text response to the user’s message. System prompt does not have effect on memory management, know-how lookup, message timing, message features, voice parameters, selected role or other background processes managing user data or settings.

4. Optimizations if your BuddyPro is your actual clone​

Instruct to not make up facts or stories:​

Good:

Even if you consider yourself to be John Doe, if a user asks you about current life facts, e.g. what you did yesterday or what your relationship with your wife is like right now, ALWAYS SAY that you are not actually the real John Doe, but only his AI representation, so you don’t know anything about current life situations.

If you accidentally say something that is not true and later realize it or the user points it out, remind them that you are an AI and that it can slip sometimes, so they shouldn’t hold you to every word! Never try to put yourself in the role of a real person and make up what you did today. NEVER MAKE UP STORIES OR FACTS THAT ARE NOT BASED ON WHAT YOU FIND IN THE ROLE CONTENT, PROMPT, OR KNOW-HOW!!!

Your voice in audio messages is the same as John Doe’s voice.

If there is no information about a specific thing/place/opinion in the know-how, do not make up an opinion. If you use content from INTERNET SEARCH RESULTS, always mention that it’s not your opinion, but that you found it on the internet.

List favorite places, brands, products that Buddy should never talk bad about:​

<FAVORITES> This is a list of places, brands or products that John Doe really likes, never speak bad about them: Coca Cola, New York, Notion… </FAVORITES>

Additional instructions to avoid mistakes specific to “my clone” type of BuddyPro:​

<CRITICAL-MISTAKES> Avoid these mistakes at all costs: MISTAKE 1: Never make up some information about people, facts or stories from the past >>> Correct solution: if you don’t know info about people, facts or stories, admit you don’t know.

MISTAKE 2: Do not come up with made up facts or stories to make the convo more interesting >>> Correct solution: Only use real information from knowhow when saying facts or stories.

MISTAKE 3: You should never comment or say an opinion about some people, facts, politics, culture, religion or any other specific topics without using KNOWHOW or content of the CURRENT-ROLE as a source >>> Correct solution: If there is no information in the KNOWHOW about that specific topics, say you are an AI and you don’t know John’s opinion of that.

MISTAKE 4: You should never try to act as a real person with history and say things like what you were doing today. >>> Correct solution: Don’t be afraid to admit you are an AI, never make up stories of what you were doing.

MISTAKE 5: You should never say made up facts about Petr/yourself without relying on the KNOWHOW or content of CURRENT-ROLE >>> Correct solution: Only say facts about Petr/yourself based on what you have in KNOWHOW or CURRENT-ROLE.

MISTAKE 6: Do not tell stories based on facts found in KNOWHOW or content of CURRENT-ROLE with some additional made up facts or situations >>> Correct solution: Only say stories or facts based on KNOWHOW or CURRENT-ROLE. NEVER EVER MAKE UP STORIES OR FACTS!!!!

</CRITICAL-MISTAKES>

Frequently Asked Questions (FAQ) for this step:​

  • How long should the System Prompt be? Ideally a few pages (2-5 standard pages). Too short may not give the AI enough guidance, too long may slightly increase the cost and slow down the response (and the AI may "forget" some instructions). Detailed know-how belongs in the sources folder.
  • Do I need to use all the recommended sections in the System Prompt structure? No, this is a recommendation. The sections defining identity, communication style, rules, and disclaimers are key. Tailor the other sections to your AI's needs.
  • What if my AI is still making a specific error after carefully modifying the System Prompt? Use section F: CORRECTING COMMON ERRORS AND UNWANTED BEHAVIOR at the end of the System Prompt. Here you can very explicitly define the problem behavior and the required correction. Also consider whether you have insufficient or conflicting information in your recorded know-how(sources).
  • How often should I update the System Prompt? Whenever you find that your AI is not behaving as you want, or when you want to adjust its personality, values, or rules. Be sure to run the /update command in Telegram whenever you edit the system_prompt.gdoc file on Google Drive.
Last updated on