RISEN Prompt Framework: Complete Guide with Examples (2026)

Last updated June 28, 2026 · 2 examples · Works with ChatGPT, Claude, Gemini
Quick Answer

RISEN is a five-part prompt engineering framework: Role sets the AI persona, Instructions describe the overall task, Steps define the exact sequence, End Goal specifies the desired outcome, and Narrowing constrains scope and format. It is the best framework for complex, multi-phase tasks where you need to control both process and output.

What Is the RISEN Prompt Framework?

RISEN extends simpler frameworks like RACE by adding two critical components: explicit Steps that tell the AI exactly how to proceed, and Narrowing that constrains what the output must and must not include. The result is a framework that gives you fine-grained control over complex tasks while still being readable and maintainable.

What Does RISEN Stand For?

Best for complex multi-step tasks requiring precise output control.

R
Role
The expert persona or professional identity you assign to the AI. Same as in RACE — sets the knowledge base and perspective.
I
Instructions
A high-level description of the overall task. What needs to happen, without specifying how yet.
S
Steps
The exact sequence of actions the AI must follow. Number them. This is what distinguishes RISEN from simpler frameworks — you control the process, not just the outcome.
E
End Goal
What a successful final output looks like. The target state after all steps are complete.
N
Narrowing
Scope constraints, exclusions, and format rules. What the AI must NOT do, include, or assume. Prevents hallucination and scope creep.
+R
Rules (optional)
Cross-cutting constraints appended at the end. Added by Promptary as an optional sixth field.

When to Use RISEN

Use RISEN when a task has multiple phases that must follow a specific sequence, when output format needs to be precisely controlled, or when previous attempts with simpler prompts produced results that were too broad or went off-track. For quick self-contained tasks, RACE is faster.

RISEN Examples

Technical blog post
Role: You are a senior developer advocate who writes technical content for a developer audience.
Instructions: Write a technical blog post about using REST APIs to manage AI prompts across multiple applications.
Steps: 1. Open with the problem: prompts scattered across tools. 2. Introduce the concept of a prompt API endpoint. 3. Show a curl example fetching a prompt. 4. Show the same in JavaScript. 5. Close with next steps.
End Goal: A complete 800-word blog post that a developer could publish immediately.
Narrowing: Do not explain what REST APIs are. Do not use marketing language. No headers like Introduction or Conclusion. Code blocks must be syntax-highlighted markdown.
Competitive analysis
Role: You are a product strategist with experience analyzing developer tools markets.
Instructions: Analyze the prompt management tools market and identify positioning opportunities for a new entrant.
Steps: 1. List the top 5 existing tools with one-line descriptions. 2. Identify the 3 most common feature gaps. 3. Map which user segments are underserved. 4. Propose 3 differentiation angles with rationale.
End Goal: A structured analysis I can use to brief a product team.
Narrowing: Focus only on tools with active users in 2026. Do not include enterprise-contract-only tools. Each differentiation angle must be achievable by a solo developer.

RISEN vs Other Frameworks

RISEN vs RACE: RACE lets the AI choose its own approach; RISEN specifies the exact steps. RISEN vs CO-STAR: CO-STAR optimizes for audience and style fit; RISEN optimizes for process and output precision. RISEN takes longer to write but produces dramatically more consistent results on complex tasks.

Use this RISEN template in Promptary — free

Save your first RISEN prompt in the structured editor and get a permanent REST API endpoint. Personal plan is free forever, no credit card required.

Start free

Frequently Asked Questions

What does RISEN stand for?

RISEN stands for Role, Instructions, Steps, End Goal, and Narrowing. The two additions over simpler frameworks — Steps and Narrowing — are what make it effective for complex, multi-phase tasks.

When should I use RISEN instead of RACE?

Use RISEN when your task has multiple phases that must follow a specific sequence, when the AI keeps going off-track with simpler prompts, or when output format must be precisely controlled. RACE is better for quick, self-contained tasks.

How detailed should the Steps field be?

Each step should be a single, clear action. Number them. Aim for 3–7 steps. If a step needs sub-steps, break the task into two separate prompts.

Can I use RISEN for image generation prompts?

RISEN is designed for language tasks. For image generation, use Promptary's dedicated Image Gen framework.