Best n8n AI Automation Prompts: 25 Examples

Fill in the blanks below, or click a highlighted word in the prompt.

You are an automated customer support triage system.
Analyze the following incoming support ticket:

Ticket Content: "{{ $json.ticket_body }}"

Tasks:
1. Determine sentiment: POSITIVE, NEUTRAL, or NEGATIVE.
2. Determine urgency level: LOW, MEDIUM, HIGH, or URGENT.
3. Classify category: BILLING, TECHNICAL, FEATURE_REQUEST, ACCOUNT_ACCESS, or OTHER.

Output strictly valid JSON with no markdown tags or introductory text:
{
  "sentiment": "STRING",
  "urgency": "STRING",
  "category": "STRING",
  "summary": "1 sentence brief summary of the issue"
}

Fill in the blanks below, or click a highlighted word in the prompt.

You are a helpful customer support agent for a SaaS platform.
Draft a polite, professional reply to the customer based ONLY on the context provided.

Customer Email: "{{ $json.email_body }}"
Knowledge Base Context: "{{ $json.kb_context }}"

Rules:
- If the knowledge base does not contain the answer, set "can_answer" to false and provide a escalation draft.
- Keep the tone helpful, concise, and professional.

Output Format (JSON only):
{
  "can_answer": true,
  "draft_reply": "Your draft email body here...",
  "suggested_action": "RESOLVE or ESCALATE_TO_HUMAN"
}

Fill in the blanks below, or click a highlighted word in the prompt.

Analyze the following web form submission for spam, promotional fluff, phishing attempts, or automated bot messages.

Form Submission:
Name: {{ $json.name }}
Email: {{ $json.email }}
Message: {{ $json.message }}

Analyze the input and return pure JSON:
{
  "is_spam": true/false,
  "confidence_score": 0.00 to 1.00,
  "reason": "Brief explanation for the score"
}

Fill in the blanks below, or click a highlighted word in the prompt.

You are a multilingual communication processor.
Analyze the incoming text message: "{{ $json.incoming_text }}"

Perform the following:
1. Detect source language (ISO 639-1 code).
2. Translate text into English if it is not already English.

Output pure JSON:
{
  "language_code": "en",
  "is_english": true/false,
  "english_translation": "Translated text here",
  "original_text": "{{ $json.incoming_text }}"
}

Fill in the blanks below, or click a highlighted word in the prompt.

You are an SLA monitor for an IT service desk.
Evaluate the following ticket metrics:

Ticket ID: {{ $json.ticket_id }}
Created At: {{ $json.created_at }}
Last Update: {{ $json.updated_at }}
Customer Tier: {{ $json.customer_tier }}
Latest Message: "{{ $json.last_message }}"

Analyze risk of SLA breach. Return JSON:
{
  "sla_risk": "LOW" | "MEDIUM" | "HIGH" | "CRITICAL",
  "recommended_action": "Short string describing next step",
  "escalate": true/false
}

Fill in the blanks below, or click a highlighted word in the prompt.

Extract key invoice information from the following unformatted text input:

Input Text:
"{{ $json.text_content }}"

Extract and return JSON according to this exact structure:
{
  "vendor_name": "String or null",
  "invoice_number": "String or null",
  "invoice_date": "YYYY-MM-DD or null",
  "due_date": "YYYY-MM-DD or null",
  "total_amount": 0.00,
  "currency": "USD",
  "line_items": [
    {
      "description": "String",
      "amount": 0.00
    }
  ]
}
Return only JSON. Do not add markdown backticks.

Fill in the blanks below, or click a highlighted word in the prompt.

Read the following meeting notes and extract key decision points and explicit action items.

Transcript:
"{{ $json.transcript }}"

Output pure JSON format:
{
  "key_decisions": ["Decision 1", "Decision 2"],
  "action_items": [
    {
      "task": "Task description",
      "owner": "Person name or Unassigned",
      "deadline": "Mentioned deadline or Unknown"
    }
  ]
}

Fill in the blanks below, or click a highlighted word in the prompt.

You are a data transformation engine. Convert the raw scraped product text into standard e-commerce fields.

Raw Data:
"{{ $json.scraped_text }}"

Return standard JSON format:
{
  "product_title": "Clean Title",
  "brand": "Brand Name or Unknown",
  "price": 0.00,
  "in_stock": true/false,
  "features": ["Feature 1", "Feature 2"],
  "specs_dict": {}
}

Fill in the blanks below, or click a highlighted word in the prompt.

Parse the following candidate resume text into a structured profile:

Resume Text:
"{{ $json.resume_text }}"

JSON Output Schema:
{
  "full_name": "String",
  "email": "String",
  "phone": "String",
  "years_experience": Number,
  "skills": ["Skill1", "Skill2"],
  "education_level": "Bachelor's / Master's / PhD / Other",
  "current_company": "String or null"
}

Fill in the blanks below, or click a highlighted word in the prompt.

Extract structured financial receipt data from this raw text string:

Raw Text:
"{{ $json.ocr_text }}"

Return JSON strictly:
{
  "merchant_name": "String",
  "transaction_date": "YYYY-MM-DD",
  "tax_amount": 0.00,
  "tip_amount": 0.00,
  "grand_total": 0.00,
  "category": "MEALS" | "TRAVEL" | "SUPPLIES" | "SOFTWARE" | "OTHER"
}

Fill in the blanks below, or click a highlighted word in the prompt.

You are an expert social media strategist.
Take the following article excerpt and write tailored posts for multiple channels.

Article Excerpt:
"{{ $json.article_text }}"

Requirements:
- Twitter/X: Under 280 characters, engaging hook, 2 hashtags.
- LinkedIn: Professional tone, bullet points, clear call to action.
- Facebook: Friendly tone, conversational tone.

Return JSON only:
{
  "twitter_post": "String",
  "linkedin_post": "String",
  "facebook_post": "String"
}

Fill in the blanks below, or click a highlighted word in the prompt.

You are an editor for a tech newsletter.
Summarize the following list of news articles into an engaging 3-bullet newsletter section.

Articles Input:
"{{ $json.articles_combined }}"

Rules:
- Include a catch title.
- Provide 3 distinct highlights with single-sentence summaries.
- Keep output markdown-formatted for direct publishing.

JSON Output:
{
  "subject_line": "Catchy Subject",
  "markdown_body": "Markdown text here"
}

Fill in the blanks below, or click a highlighted word in the prompt.

Generate an SEO-optimized title tag and meta description based on the provided page content.

Page Title: "{{ $json.title }}"
Content Body: "{{ $json.content }}"
Target Keyword: "{{ $json.keyword }}"

Guidelines:
- Meta Title: Max 60 characters, include target keyword near the front.
- Meta Description: Max 155 characters, includes target keyword, strong CTA.

Output pure JSON:
{
  "meta_title": "String",
  "meta_description": "String",
  "character_counts": {
    "title_len": 0,
    "desc_len": 0
  }
}

Fill in the blanks below, or click a highlighted word in the prompt.

Organize this raw video transcript into logical chapters with timestamps.

Transcript Input:
"{{ $json.raw_transcript }}"

JSON Format:
{
  "chapters": [
    {"timestamp": "00:00", "title": "Introduction"},
    {"timestamp": "02:15", "title": "Topic 1"}
  ],
  "overall_summary": "Short paragraph text"
}

Fill in the blanks below, or click a highlighted word in the prompt.

Rewrite the following developer Git commits into polished, customer-facing product update release notes.

Commit Messages:
"{{ $json.commit_logs }}"

Formatting Requirements:
- Group into "New Features", "Improvements", and "Bug Fixes".
- Remove technical jargon and internal PR numbers.

Output JSON:
{
  "changelog_markdown": "Formatted markdown text"
}

Fill in the blanks below, or click a highlighted word in the prompt.

You are a Sales Operations AI. Evaluate incoming lead details to calculate an ICP fit score (0-100).

Lead Data:
Email: {{ $json.email }}
Company Size: {{ $json.company_size }}
Job Title: {{ $json.job_title }}
Use Case Description: "{{ $json.use_case }}"

Scoring Criteria:
- Enterprise domain (@company.com vs @gmail.com): +20
- Senior decision maker (Director, VP, C-Level): +30
- High match use case: +30
- Company size > 50 employees: +20

Return pure JSON:
{
  "lead_score": 85,
  "qualification_status": "HOT" | "WARM" | "COLD",
  "scoring_breakdown": "Explanation of score"
}

Fill in the blanks below, or click a highlighted word in the prompt.

Analyze the web page text of a target company and return firmographic intelligence.

Scraped Web Page:
"{{ $json.website_text }}"

JSON Output:
{
  "company_name": "String",
  "industry": "String",
  "business_model": "B2B" | "B2C" | "Marketplace" | "Unknown",
  "core_value_proposition": "1 sentence description",
  "target_audience": "Description of target customer"
}

Fill in the blanks below, or click a highlighted word in the prompt.

Craft a personalized 3-sentence sales email to a prospect.

Prospect Name: {{ $json.prospect_name }}
Prospect Title: {{ $json.prospect_title }}
Company: {{ $json.company }}
Recent News / Bio Snippet: "{{ $json.bio_snippet }}"
Our Product Value Prop: "{{ $json.our_value_prop }}"

Guidelines:
- No generic openers ("I hope this email finds you well").
- Refer to their specific context in line 1.
- Offer a soft call to action.

Return JSON:
{
  "subject_line": "String",
  "email_body": "String"
}

Fill in the blanks below, or click a highlighted word in the prompt.

Analyze this sales call snippet for mentions of market competitors.

Transcript Segment:
"{{ $json.call_segment }}"

Output pure JSON:
{
  "competitor_mentioned": true/false,
  "competitors_found": ["Competitor A", "Competitor B"],
  "customer_sentiment_towards_competitor": "POSITIVE" | "NEGATIVE" | "NEUTRAL",
  "key_concerns_raised": ["Concern 1"]
}

Fill in the blanks below, or click a highlighted word in the prompt.

Summarize the following client meeting transcript for sales record logging.

Transcript Text:
"{{ $json.transcript }}"

Return JSON:
{
  "executive_summary": "Brief 2-3 sentence overview.",
  "client_pain_points": ["Point 1", "Point 2"],
  "agreed_next_steps": ["Step 1", "Step 2"],
  "deal_health": "STRONG" | "NEUTRAL" | "AT_RISK"
}

Fill in the blanks below, or click a highlighted word in the prompt.

You are a DevOps assistant monitor. Diagnose this system execution failure log.

Node Name: {{ $json.node_name }}
Error Message: "{{ $json.error_message }}"
Execution Stack: "{{ $json.stack_trace }}"

Tasks:
1. Explain the root cause in simple language.
2. Recommend immediate fix steps for the developer.

Output JSON:
{
  "summary": "1 sentence plain language error explanation",
  "probable_cause": "Detailed technical cause",
  "action_required": "Step-by-step resolution guidance",
  "severity": "CRITICAL" | "WARNING" | "INFO"
}

Fill in the blanks below, or click a highlighted word in the prompt.

Transform the dynamic payload into the required destination structure.

Input JSON Payload:
{{ JSON.stringify($json.raw_payload) }}

Target Fields:
- external_id (map from id, uuid, or user_id)
- user_email (map from email, contact_email)
- status (default to "PENDING" if not explicitly provided)

Return target structure as pure JSON only.

Fill in the blanks below, or click a highlighted word in the prompt.

An n8n SQL node executed with an error. Analyze the statement and error message.

Query: "{{ $json.query }}"
SQL Error: "{{ $json.error }}"

Return JSON:
{
  "error_type": "SyntaxError" | "ConstraintError" | "ConnectionError" | "Unknown",
  "corrected_sql": "Suggested corrected SQL statement",
  "explanation": "Why the original query failed"
}

Fill in the blanks below, or click a highlighted word in the prompt.

Screen the following text for safety compliance (hate speech, severe profanity, personal identifiable information leaks).

Input Text:
"{{ $json.user_input }}"

Return pure JSON:
{
  "passed_filter": true/false,
  "flagged_categories": ["CategoryName"],
  "sanitized_text": "Text with PII/redacted items replaced by [REDACTED] if needed"
}

Fill in the blanks below, or click a highlighted word in the prompt.

You are a routing decision engine for a business process workflow.
Evaluate the incoming payload and decide which workflow branch should handle it.

Payload:
"{{ $json.payload_description }}"

Available Branches:
- BRANCH_A: Financial transactions, refund requests, billing inquiries.
- BRANCH_B: Technical bug reports, API questions, platform outages.
- BRANCH_C: General inquiries, marketing messages, sales leads.

Output strictly valid JSON:
{
  "selected_branch": "BRANCH_A" | "BRANCH_B" | "BRANCH_C",
  "reasoning": "Brief explanation for this choice"
}