Write a System Prompt for an AI Agent
Turn a description of what your agent should do into a structured, production-ready system prompt with explicit boundaries and failure behavior.
0 likes
0 dislikes
Sign in to rate this prompt
Prompt
Write a system prompt for an AI agent based on the description below. Treat it as an operating manual, not a personality sketch.
What the agent is for:
{{purpose}}
Who uses it, and in what context:
{{context}}
What it must NOT do / its boundaries:
{{constraints}}
Structure the system prompt with clear sections covering:
- **Role & objective** — what the agent is responsible for, in one or two lines.
- **How to behave** — tone, level of detail, and how it should handle the main tasks.
- **Boundaries** — what it must refuse or escalate, and what's out of scope.
- **Failure behavior** — what to do when it's unsure, lacks information, or a request is ambiguous (for example, ask a clarifying question rather than guess).
- **Output expectations** — the format and structure of its responses, if that matters.
Keep instructions plain and direct; avoid vague superlatives, and don't stack aggressive "you must always" language. If a key detail (like the boundaries) is missing, ask me for it before writing.
Why this prompt works
-
Claude vs Gemini vs ChatGPT
Google's Flash rate doubles on 1 January 2027 and its flagship is still preview; OpenAI spans 50x from Luna to Astra with no retirement d...
Read the guide → -
ChatGPT vs Claude
Both flagships now cost exactly $10/$50 — but Anthropic's tokenizer fits 555k words into a million tokens where the old one fit 750k, so ...
Read the guide → -
How to Deploy an AI Agent
Every deployment problem follows from one property: you don't know how long a run takes, because you didn't write the path. Why the run m...
Read the guide →