
A Claude skill that writes the accurate prompts for any AI tool. Zero tokens or credits wasted. Full context and memory retention. No re-prompting your way to an answer you should have gotten on attempt one.
Works with: Claude, ChatGPT, Codex, Grok, Gemini, o1/o3, MiniMax, Cursor, Claude Code, GitHub Copilot, Windsurf, Bolt, v0, Lovable, Devin, Perplexity, Midjourney, DALL-E, Stable Diffusion, ComfyUI, Sora, Runway, ElevenLabs, Zapier, Make, and any AI tool you throw at it.
๐ Installation
RECOMMENDED - Claude.ai (browser)
- Download this repo as a ZIP
- Go to claude.ai โ Sidebar โ Customize โ Skills โ Upload a Skill
OR Clone directly into Claude Code skills directory (Not Suggested)
mkdir -p ~/.claude/skills
git clone https://github.com/nidhinjs/prompt-master.git ~/.claude/skills/prompt-master
๐ฅ The Problem This Solves
Every AI user wastes credits the same way:
Write vague prompt โ get wrong output โ re-prompt โ get closer โ re-prompt again โ finally get what you wanted on attempt 4
That's 3 wasted API calls. Multiply by 50 prompts a day. That's real money and real time gone.
The key insight
"The best prompt is not the longest. It's the one where every word is load-bearing."
Most "prompt generators" make prompts longer. This skill makes them sharper.
๐ฏ Usage
In Claude, you can invoke the skill naturally:
Write me a prompt for Cursor to refactor my auth module
I need a prompt for Claude Code to build a REST API โ ask me what you need to know
Here's a bad prompt I wrote for GPT-4o, fix it: [paste prompt]
Generate a Midjourney prompt for a cyberpunk city at night
I have a reference image โ help me write a prompt to edit just the head angle
Break this prompt down and adapt it for Stable Diffusion
Or explicitly invoke it:
/prompt-master
I want to ask Claude Code to build a todo app with React and Supabase
How It Works
Prompt Master runs a structured pipeline on every request:
- Detects the target tool โ figures out which AI system the prompt is for and routes silently to the right approach
- Extracts 9 dimensions of intent โ task, input, output, constraints, context, audience, memory, success criteria, examples
- Asks targeted clarifying questions โ max 3 questions if critical info is missing, never more
- Routes to the right framework โ picks and applies the correct prompt architecture automatically, never shown to the user
- Applies safe techniques only โ role assignment, few-shot examples, XML structure, grounding anchors, memory block as needed
- Checks model recency โ verifies exact models and controls against official provider docs when the request depends on "latest"
- Runs a token efficiency audit โ strips every word that doesn't change the output
- Delivers the prompt โ one clean copyable block with a one-line strategy note
Full Example #1: Generating Prompts for Images
User Input
Write me a midjourney prompt for a realistic samurai standing in the rain at night
Generated Prompt
lone samurai standing in heavy rain at night, traditional armor,
neon reflections on wet cobblestone street, cinematic lighting,
dramatic shadows, fog, ultra detailed, photorealistic,
shallow depth of field --ar 16:9 --v 6 --style raw
negative: blurry, low quality, watermark, cartoon, anime, extra limbs
๐ฏ Target: Midjourney ยท โก Framework: Visual Descriptor ยท ๐ฐ Tokens: Light (~60) ยท ๐ก Strategy: Comma-separated descriptors over prose, lighting and mood anchored early, aspect ratio and version locked, negative prompt prevents style drift.
Full Example #2: Generating Prompts for Coding
User Input
Build a claude code prompt for a landing page for a business
dashboard that looks and feels exactly like notion - smooth animations,
clean ui
Generated Prompt
Objective:
Build a fully detailed, production-quality static landing page for a
business dashboard product. Design must match Notion's exact aesthetic โ
clean, minimal, typographic, with smooth animations throughout.
Stack:
Single HTML file with embedded CSS and vanilla JS. No frameworks.
No external dependencies except Google Fonts (Inter).
Design Spec โ match these exactly:
- Background: #ffffff
- Primary text: #1a1a1a
- Secondary text: #6b7280
- Border color: #e5e7eb
- Accent: #000000
- Font: Inter, weights 400 / 500 / 600 / 700
- Base spacing unit: 8px
- Border radius: 6px on cards, 4px on buttons
- Box shadows: 0 1px 3px rgba(0,0,0,0.08) only โ no dramatic shadows
- Buttons: black fill, white text, no border radius above 6px
Sections to build in order:
1. Navbar โ logo left, nav links center, "Get started free" CTA right.
Sticky on scroll, adds subtle border-bottom on scroll with JS.
2. Hero โ large left-aligned heading (max 3 lines), subtext, two CTAs
(primary black + secondary outlined), hero screenshot mockup
placeholder (gray rounded card, 16:9 ratio).
3. Logos bar โ "Trusted by teams at..." with 5 placeholder company names
in muted gray, centered.
4. Features โ 3-column grid. Each card: icon (use emoji), bold title,
2-line description. Hover state: border darkens to #d1d5db,
translateY(-2px), transition 200ms ease.
5. How it works โ alternating left/right layout, 3 steps. Number badge,
heading, description, placeholder screenshot card.
6. Pricing โ 2 tiers side by side. Free tier: outlined card. Pro tier:
black background, white text, "Most popular" badge. List 4 features
each with checkmarks.
7. CTA banner โ full-width dark section, centered heading, subtext,
single CTA button.
8. Footer โ 4-column links, copyright line, minimal.
Animations โ implement all of these:
- Fade-in + translateY(20px) to translateY(0) on scroll for every section,
using IntersectionObserver. Threshold 0.15. Duration 500ms ease-out.
- Staggered animation on feature cards: 100ms delay between each card.
- Navbar CTA button: background transitions from black to #333 on hover,
200ms ease.
- Hero CTA primary: same hover transition. Secondary: background goes from
transparent to #f3f4f6 on hover.
- Smooth scroll on all anchor links.
- Navbar border-bottom fades in on scroll past 60px using JS scroll event.
Constraints:
- Single file only โ all CSS and JS embedded, no external files
- No dependencies except Google Fonts via CDN link tag
- All placeholder images replaced with styled gray div elements
- Must be pixel-perfect at 375px mobile and 1440px desktop
- No CSS frameworks, no utility classes from external sources
Done When:
- All 8 sections render correctly at both breakpoints
- All animations trigger on scroll without layout shift
- Navbar sticks and border appears on scroll
- Hover states work on all interactive elements
- Opens in browser with zero console errors
๐ฏ Target: Claude Code ยท ๐ฐ Tokens: Medium (~380) ยท ๐ก Strategy: Every vague Notion aesthetic cue translated into exact hex values and pixel specs โ Claude Code cannot guess wrong. Animations defined with exact timing, method, and trigger so there is no interpretation needed.
๐ค Works With Any AI Tool
Prompt Master includes specific profiles for 20+ tools. For anything not on the list, it uses a Universal Fingerprint: 4 questions that let it write a quality prompt for any AI system it has never seen before.
๐ 14 Prompt Templates (Auto-Selected)
Prompt Master picks the right architecture for every task automatically and routes silently โ you never see the framework name, just the prompt.
๐ก๏ธ 5 Safe Techniques, Applied When Needed
Prompt Master only uses techniques with reliable, bounded effects. Methods known to produce hallucinations or unpredictable output (Tree of Thought, Graph of Thought, Universal Self-Consistency, prompt chaining) are explicitly excluded.
๐ซ 37 Credit-Killing Patterns Detected (with Before/After Examples)
Context Patterns
๐ง Memory Block System
When your conversation has history, Prompt Master pulls out prior decisions and prepends a Memory Block so the AI never contradicts earlier work:
## Memory (Carry Forward from Previous Context)
- Stack: React 18 + TypeScript + Supabase
- Auth uses JWT stored in httpOnly cookies, not localStorage
- Component naming convention: PascalCase, no default exports
- Design system: Tailwind only, no custom CSS files
- Architecture: no Redux, context API only
This is the single biggest fix for long sessions. Most wasted re-prompts come from the AI forgetting what you already decided.
โน๏ธ Version History
- 1.8.0 โ Current-model refresh. Added Claude Fable 5, Opus 5, Sonnet 5, GPT-5.6 Sol/Terra/Luna, Codex, and Grok 4.6 routing. Replaced hidden chain-of-thought requests with auditable reasoning and generalized the Claude task brief for current adaptive-thinking models.
- 1.7.0 โ Opus 4.8 compatibility. Made Claude 4.x routing version-aware: durable advice generalized across 4.6/4.7/4.8, added Opus 4.8 (current default) profile, kept Opus 4.7 labeled. De-hardcoded the effort-level note (now harness-managed). Template M and pattern 36 cover 4.7 and 4.8. Fixed a stray fragment in patterns.md.
- 1.6.0 โ Opus 4.7 update. Added Template M (Opus 4.7 Task Brief). Updated Claude and Claude Code routing for literalism, adaptive thinking, xhigh effort, and session hygiene. Added patterns 36โ37.
- 1.5.0 โ Added more tool routing. New Agentic AI and 3D Model AI routing added. Fixed description to 189 chars. Removed token estimate from output. Added instruction layer and copywriting placeholders
- 1.4.0 โ Added reference image editing detection, ComfyUI support, Prompt Decompiler mode. Fixed trigger description to invoke correctly in Claude Code. 3 new templates added to references folder
- 1.3.0 โ Rebuilt around PAC2026 positional structure (30/55/15). Silent routing replaces user-facing framework selection. References folder introduced
- 1.2.0 โ Restructured for attention architecture. Removed fabrication-prone techniques (ToT, GoT, USC, prompt chaining). Templates and patterns moved to references folder
- 1.1.0 โ Expanded tool coverage, added memory block system, 35 credit killing patterns
- 1.0.0 โ Initial release
๐ License
MIT: See LICENSE for details.