Generate a System Prompt for an Agent
Create an agent system prompt from a target objective, examples, and operational requirements.
Prompt template
- Format
- Text Parameterized
- Inputs
- target_objective (text) · Required correct_response_examples (text) · Required context (text) · Required output_format (text) · Required
- Outputs
- system_prompt (text)
Write a system prompt for an agent.
Target objective: ${target-objective}
Examples of correct responses: ${correct-response-examples}
Context: ${context}
Exact output format: ${output-format}
The system prompt must include five prohibitions and explain how the agent should behave when information is missing. Instruct the agent to ask follow-up questions instead of making things up.
Return only the text of the system prompt.
Replace the ${…} placeholders with your values
before using this prompt. See parameters
and defaults below
Parameters
Replace each placeholder with your own value, or use its default when one is provided
${target-objective}
target-objective
Required
${correct-response-examples}
correct-response-examples
Required
${context}
context
Required
${output-format}
output-format
Required
What you'll create
The resulting instruction specifies five prohibitions and a clear missing-information policy while preserving the exact requested response format.
Frequently asked questions
Should the agent guess missing facts?
No. It must ask follow-up questions instead of inventing information.
Comments
Tried this prompt? Share what you made, compare results, or ask a question