PromptMake

RISEN Prompt Framework Generator

Describe your task — get a RISEN prompt with Role, Instructions, numbered Steps, End goal, and Narrowing constraints ready to paste into ChatGPT or Claude.

Related guide: Read on the blog →

What RISEN stands for

RRole

Perspective the model adopts — reviewer, tutor, analyst, editor. Tie to audience or domain, not empty praise.

IInstructions

Primary task in imperative language. One clear job statement before steps expand it.

SSteps

Numbered workflow: 3–7 actionable steps the model should follow in order.

EEnd goal

Success criteria — deliverable shape, acceptance tests, what the user can do with the output.

NNarrowing

Constraints: scope limits, banned behaviors, factual guardrails, escalation paths.

Empty template

Role: [perspective]
Instructions: [primary task]
Steps:
1. [step]
2. [step]
End goal: [success criteria]
Narrowing: [constraints]

Filled example

Role: Senior security engineer reviewing a teammate's pull request.
Instructions: Audit the payment retry handler for error-handling gaps.
Steps:
1. Read the diff for unhandled exceptions and missing timeouts.
2. List must-fix issues with file and function names.
3. Suggest should-fix improvements separately.
4. Skip style nits unless they hide bugs.
End goal: Markdown review comment ready to post on GitHub.
Narrowing: No invented CVEs. Flag uncertainty as NEEDS VERIFY. Max 2 sentences per bullet.

Generate your RISEN prompt

Describe the task or paste a rough draft — get labeled sections plus a combined paste-ready prompt.

0/800

Upgrade your limits

1 credit = 1 generation. Go unlimited with Pro or buy credit packs.

Free

$0/month

  • 5 text / day · 3 image / day
  • Prompt history
Get Started

Pro

$9/month

  • Unlimited generations
  • Batch image processing
  • Pro options

Pro Annual

$79/year

  • Everything in Pro
  • Best value

Or pay as you go

Starter

$1.99

50 credits

Pro Pack

$4.99

200 credits

Ultimate

$12.99

1000 credits

Batch image processing

Upload up to 10 images at once on Image to Prompt. Pro includes unlimited batches. Otherwise: $1.99 one-time per batch, or 10 credits.

Log in to purchase. Full pricing page →

RISEN for multi-step prompt design

Single-sentence prompts fail when a task has phases: read, analyze, compare, recommend. RISEN's Steps block turns implicit expectations into an ordered checklist the model follows before declaring done.

End goal and Narrowing close the loop. End goal states what deliverable success looks like; Narrowing prevents scope creep, invented facts, and unsafe outputs.

Strong Role and Narrowing patterns

Effective Role lines specify audience or review style: "Write for a CFO with two minutes" beats "You are an expert." Narrowing should include at least one factual guardrail — no invented statistics, cite only pasted sources, or flag unknowns explicitly.

PromptMake generates both labeled sections and a combined prose prompt so you can paste into chat UIs that prefer paragraphs or structured blocks.

RISEN in production workflows

Teams store RISEN templates in Notion or git for recurring jobs: weekly competitive brief, PR review, customer ticket triage. Swap only Instructions and Context per run; keep Steps and Narrowing stable for consistent evals.

When a RISEN prompt still drifts, add one example output in Narrowing or split Steps finer — usually faster than rewriting the whole prompt from scratch.

FAQ

What is the RISEN prompt framework?

RISEN stands for Role, Instructions, Steps, End goal, and Narrowing. It structures prompts for multi-step work where the model needs a workflow and explicit constraints, not just a one-line ask.

When should I use RISEN instead of CO-STAR?

Use RISEN for procedural tasks: code review, research synthesis, checklist-driven analysis, onboarding flows. Use CO-STAR when style, tone, and audience drive the output — marketing and comms.

How is RISEN different from RTF?

RTF covers Role, Task, and Format in three blocks. RISEN adds numbered Steps, explicit End goal acceptance criteria, and Narrowing guardrails — better for tasks that can drift or hallucinate facts.

What makes a good Narrowing section?

Include scope limits (files, topics), factual rules (do not invent pricing), output boundaries (max bullets), and what to do when information is missing (write NEEDS INPUT).

Can I use RISEN for agents and tool use?

Yes. Steps map well to agent loops: plan, call tool, verify, summarize. Pair RISEN prompts with explicit stop conditions in Narrowing for safer automation.

Is it free?

Yes — same free tier as PromptMake /text. Guests get 3 generations per day without signup.