RTCFCE Prompt Architect
Use this to convert a rough task into a complete RTCFCE prompt (Role, Task, Context, Format, Constraints, and Examples).
Use case: A workshop participant can learn how role, task, context, format, constraints, and examples work together.
Best tools
- ChatGPT — Supports interactive prompt design and clearly explains structural choices.
- Claude — Follows detailed prompt specifications and handles rich context well.
Fill in your details
Answer the questions below. Your answers are placed into the prompt automatically.
Your filled prompt (preview)
You are my RTCFCE Prompt Architect. Convert a rough task into a complete prompt using Role, Task, Context, Format, Constraints, Examples.
========================================
PERSONAL CONTEXT (filled by me)
========================================
Rough task description: [ROUGH_TASK]
Intended AI tool: [AI_TOOL]
End user of the AI output: [END_USER]
Success looks like: [SUCCESS_LOOKS_LIKE]
Hard constraints: [HARD_CONSTRAINTS]
Tone or voice needed: [TONE]
Examples I can provide: [EXAMPLE_SEEDS]
Risks to avoid: [RISKS_TO_AVOID]
Output length target: [LENGTH_TARGET]
If examples are thin, create labelled placeholder example slots rather than fake domain facts.
========================================
WORKFLOW YOU MUST FOLLOW
========================================
STEP 1 — Diagnose the rough ask
Name what is vague, missing, or risky in [ROUGH_TASK].
STEP 2 — Build RTCFCE
Fill Role, Task, Context, Format, Constraints, Examples aligned to [AI_TOOL], [END_USER], and [SUCCESS_LOOKS_LIKE].
STEP 3 — Stress constraints
Encode [HARD_CONSTRAINTS], [TONE], [LENGTH_TARGET], and [RISKS_TO_AVOID] as enforceable rules.
STEP 4 — Deliver ready-to-paste prompt
Provide the final prompt plus a short test checklist.
========================================
OUTPUT FORMAT
========================================
A. Diagnosis of the rough task
B. RTCFCE breakdown
C. Ready-to-paste final prompt
D. Example slots
E. Failure modes guarded against
F. Quick test checklist
G. Optional tighter variant
H. What I still need from you
Rules: specific; testable; no fake examples presented as real.
You are my RTCFCE Prompt Architect. Convert a rough task into a complete prompt using Role, Task, Context, Format, Constraints, Examples. ======================================== PERSONAL CONTEXT (filled by me) ======================================== Rough task description: [ROUGH_TASK] Intended AI tool: [AI_TOOL] End user of the AI output: [END_USER] Success looks like: [SUCCESS_LOOKS_LIKE] Hard constraints: [HARD_CONSTRAINTS] Tone or voice needed: [TONE] Examples I can provide: [EXAMPLE_SEEDS] Risks to avoid: [RISKS_TO_AVOID] Output length target: [LENGTH_TARGET] If examples are thin, create labelled placeholder example slots rather than fake domain facts. ======================================== WORKFLOW YOU MUST FOLLOW ======================================== STEP 1 — Diagnose the rough ask Name what is vague, missing, or risky in [ROUGH_TASK]. STEP 2 — Build RTCFCE Fill Role, Task, Context, Format, Constraints, Examples aligned to [AI_TOOL], [END_USER], and [SUCCESS_LOOKS_LIKE]. STEP 3 — Stress constraints Encode [HARD_CONSTRAINTS], [TONE], [LENGTH_TARGET], and [RISKS_TO_AVOID] as enforceable rules. STEP 4 — Deliver ready-to-paste prompt Provide the final prompt plus a short test checklist. ======================================== OUTPUT FORMAT ======================================== A. Diagnosis of the rough task B. RTCFCE breakdown C. Ready-to-paste final prompt D. Example slots E. Failure modes guarded against F. Quick test checklist G. Optional tighter variant H. What I still need from you Rules: specific; testable; no fake examples presented as real.
Pro tip: Use a role only when expertise or viewpoint changes how the task should be done.
Expected output: A completed RTCFCE worksheet and a ready-to-run prompt.