What We Actually Tell the AI

This is the exact prompt we send to GPT-4o. It takes the raw form inputs and turns them into clear, natural messages your team would be comfortable sending.

Role Instructioning → Sets the assistant’s identity as a professional blog writer to define purpose and writing standard.

‍Content Constraints → Instructs the assistant to write a short-form blog post between 400 to 800 words.

‍Structural Prompting → Specifies the blog layout: introduction, 2 to 3 core sections, and a short conclusion.

‍Tone Conditioning → Requires a natural, clear, and slightly conversational tone. Bullet points and listicles are not allowed.

‍Output Formatting → Mandates output of plain blog text only. No titles, explanations, formatting, or markdown.

‍Input Referencing → Provides the blog topic using the variable input: {{ $json.Topic }}.

Role Instructioning → Sets the assistant’s identity as a social media copywriter for Facebook to establish voice and context.

‍Task Specification → Instructs the assistant to write a single Facebook caption based on the given blog post.

‍Tone Conditioning → Requires the caption to hook the reader, sound friendly, and close with a casual call to action.

‍Content Enhancement → Encourages the use of 2 to 3 emojis and relevant hashtags to improve engagement.

‍Output Formatting → Mandates that only the caption be returned with no other text or explanation.

‍Input Referencing → Provides the blog post content using the input variable: {{ $json.message.content }}.

Role Instructioning → Sets the assistant’s identity as a professional LinkedIn content strategist to guide the tone and intent.

Task Specification → Directs the assistant to write a single LinkedIn caption based on the provided blog post.

‍Tone Conditioning → Instructs the assistant to open with a thought-provoking hook and maintain a confident, professional tone.

‍Content Enhancement → Specifies that no links should be included and encourages the use of relevant hashtags for visibility.

‍Output Formatting → Requires returning only the caption with no formatting, markdown, or commentary.

Input Referencing → Provides the blog post using the input variable: {{ $('Blog Creation').item.json.message.content }}.

Role Instructioning → Sets the assistant’s identity as a digital strategist writing for X (formerly Twitter) to shape voice and brevity.

‍Task Specification → Instructs the assistant to summarize the blog post into a single, punchy tweet.

‍Tone Conditioning → Directs the assistant to focus on one key insight or hook and stay within 280 characters.

‍Content Enhancement → Specifies that no links should be included and encourages the use of relevant hashtags and light emojis to increase engagement.

‍Output Formatting → Requires returning only the tweet with no formatting, markdown, or extra commentary.

‍Input Referencing → Provides the blog post input using the variable: {{ $('Blog Creation').item.json.message.content }}.

Role Instructioning → Sets the assistant’s identity as a friendly, conversational email copywriter to define the tone and purpose.

‍Task Specification → Instructs the assistant to write a 3 to 6 sentence email based on the provided blog post, beginning with a curiosity-driven hook and highlighting the main takeaway.

‍Tone Conditioning → Directs the assistant to maintain a warm, approachable tone that reflects a helpful, human brand voice.

‍Output Formatting → Requires only the email body to be returned, with no subject line or preview text. Includes a clear, inviting call to action.

‍Input Referencing → Provides the blog content using the variable: {{ $('Blog Creation').item.json.message.content }}.

‍Content Injection → Instructs the assistant to include a placeholder blog link at the end of the email: .