promptkit

PromptKit 🎯

Generate effective prompts using 56 prompting frameworks. Auto-selects the best framework based on user needs.

Inspired by 56 Framework Prompting by @_khaamal.

---

How to Use

Quick — just ask

"PromptKit, write a prompt for email marketing"

Guided — step-by-step

"PromptKit, guided"

Advanced — full control

"PromptKit, prompt using CO-STAR for Instagram caption, casual tone, skincare brand"

---

Auto-Select

---

56 Frameworks Reference

Writing & Content: C-A-R-E, T-A-G, R-T-F, CO-STAR, GRADE, COAST

Analytical: P-A-C-T, R-I-S-E, D-R-E-A-M, S-O-L-V-E, R-A-C-E, TRACE, A-P-E, PECRA, OSCAR, RODES, TRACI

Advanced: Rule-Based, CoT, PAP, Self-Consistency, APE, ToT, GoT, Least-to-Most, Few-Shot CoT, Stepwise Refinement, Chain of Verification, Progressive Summarization, Constraint-Based, Deliberation

Creative: Error-Driven, Inversion, Roleplay+Constraint, Counterfactual, Problem-Solving, Reverse Engineering, Analogy, Persona Comparison, Time-Travel, Cross-Domain Transfer, Multi-Stage, Layered Abstraction, Debate, Test-and-Improve, Perspective Shift, Recursive, Exploration vs Exploitation, Heuristic, Comparative, Simulation, Creative Constraints

---

Framework Templates

C-A-R-E (Context, Action, Result, Example)

Context: [role + background]
Action: [specific task]
Result: [format, style, criteria]
Example: [sample output]

T-A-G (Task, Action, Goal)

Task: [what to produce]
Action: [how to do it]
Goal: [why, measurable outcome]

R-T-F (Role, Task, Format)

Role: [expertise]
Task: [specific deliverable]
Format: [length, structure, constraints]

CO-STAR (Context, Objective, Style, Tone, Action, Response)

Context: [background]
Objective: [goal]
Style: [writing style]
Tone: [emotional tone]
Action: [what to produce]
Response: [output format]

P-A-C-T (Problem, Approach, Compromise, Test)

Problem: [specific issue]
Approach: [strategy]
Compromise: [trade-offs accepted]
Test: [how to validate]

S-O-L-V-E (Situation, Objective, Limitations, Vision, Execution)

Situation: [current state]
Objective: [desired outcome]
Limitations: [constraints]
Vision: [ideal end state]
Execution: [steps to get there]

D-R-E-A-M (Define, Research, Execute, Analyze, Measure)

Define: [goal]
Research: [info needed]
Execute: [steps]
Analyze: [what to evaluate]
Measure: [success metrics]

Chain of Thought (CoT)

Think step by step:
1. [step 1]
2. [step 2]
...
Conclusion: [final answer]

Tree of Thoughts (ToT)

Consider multiple paths:
Path A: [approach] → [pros/cons]
Path B: [approach] → [pros/cons]
Path C: [approach] → [pros/cons]
Best path: [selection + reasoning]

---

Iteration

If output isn't right:

"Improve the previous prompt by [changes]" "PromptKit, iterate on this prompt: [paste]"

---

Quality Checklist

• ✅ Role clear
• ✅ Task specific
• ✅ Output format defined
• ✅ Constraints included
• ✅ Examples (when applicable)

---

Languages

en, id, zh, ja, ms, ar — specify with (EN) or (ID) etc.

---

Version: 1.0.2 | Author: @khamalismadie