nidhinjs/prompt-master

What it solves

Prompt Master is designed to eliminate the "trial-and-error" cycle of prompting, where users waste API credits and time by repeatedly refining vague prompts to get the desired output. It transforms imprecise user requests into high-precision, "load-bearing" prompts tailored to specific AI tools.

How it works

It operates as a Claude skill that processes requests through a structured pipeline:

1. Tool Detection: Identifies the target AI tool (e.g., Midjourney, Cursor, GPT-4o) and routes to the appropriate approach.
2. Intent Extraction: Analyzes nine dimensions of intent, including task, constraints, and success criteria.
3. Clarification: Asks a maximum of three targeted questions if critical information is missing.
4. Framework Selection: Automatically applies one of 13 specialized prompt architectures (such as RTF, CO-STAR, or RISEN) without showing the framework name to the user.
5. Optimization: Applies safe techniques like role assignment and XML tags, performs a token efficiency audit to strip unnecessary words, and verifies model recency against provider docs.
6. Memory Management: Uses a "Memory Block" system to carry forward prior decisions and project context to prevent contradictions in long conversations.

Who it’s for

AI users who want to maximize the quality of their first-attempt outputs across a wide variety of AI tools, including LLMs, image generators, coding agents, and autonomous agents.

Highlights

• Broad Tool Support: Includes specific profiles for over 30 tools, including Claude, ChatGPT, Midjourney, Sora, and Devin.
• ** uma Universal Fingerprint**: A fallback mechanism to generate quality prompts for unknown AI systems.
• Credit-Saving Patterns: Detects and fixes 37 common "credit-killing" patterns across task, context, format, scope, and reasoning.
• Precision-Focused: Prioritizes sharp, efficient prompts over long ones to reduce token waste.