Global Prompt

Set consistent instructions and tone for all AI workflows and agent responses.

Global Prompt provides consistent context and instructions that apply to all your AI interactions across workflows and agents.

What is Global Prompt?

A global prompt is a set of instructions that gets prepended to every AI request. Use it to:

  • Define your preferred communication tone (formal, casual, technical)
  • Provide personal context (role, expertise, preferences)
  • Set response guidelines (length, format, language)
  • Establish domain knowledge (industry terms, company info)

Enable Global Prompt

  1. Go to Settings
  2. Find the Global Prompt section
  3. Toggle Enable Global Prompt on
  4. Enter your prompt in the text area
  5. Click Save Changes

Global Prompt section Fig : Global Prompt settings with enable toggle and prompt text area

Example Prompts

Professional Tone

You are assisting a software engineer. Keep responses technical and concise.
Use code examples when relevant. Prefer bullet points over paragraphs.

Customer Support Agent

You are a friendly customer support assistant for [Company Name].
Always be polite and empathetic. Offer solutions proactively.
If unsure, suggest escalating to a human agent.

Domain Expert

You are assisting with financial analysis. Use industry-standard terminology.
Always cite data sources. Format numbers with proper currency symbols.
Consider regulatory compliance in recommendations.

How It Works

User Query → Global Prompt + Workflow System Message + Query → AI Response

The global prompt combines with workflow-specific instructions, giving you both consistent baseline behavior and workflow-specific customization.

Best Practices

  • Keep it focused — Include only universally applicable instructions
  • Don't repeat — Avoid duplicating what's in workflow system messages
  • Test thoroughly — Verify it works well across different workflows
  • Update as needed — Refine based on AI response quality

Use Cases

Use CaseGlobal Prompt Focus
Personal assistantPreferences, schedule patterns, communication style
Business workflowsCompany context, brand voice, compliance requirements
DevelopmentCode style, tech stack, documentation preferences
ResearchCitation format, academic tone, source requirements