This one activates only when you explicitly ask it to write or improve a prompt for a specific AI tool. It extracts intent across nine dimensions (task, target tool, output format, constraints, etc.), then outputs a single production-ready prompt optimized for that system. The routing logic is detailed: it knows not to add chain of thought to o3 or o4-mini, understands that Opus 4.7 takes instructions literally and needs full context upfront, and adjusts for Gemini's citation hallucination tendencies. It won't pad output with theory unless you ask, won't activate for general coding or writing tasks, and caps clarifying questions at three. Useful if you work across multiple LLMs and want prompts tuned to each one's actual behavior, not generic templates.
npx -y skills add nidhinjs/prompt-master --skill prompt-master --agent claude-codeInstalls to .claude/skills
Select a file.
sickn33/antigravity-awesome-skills
github/awesome-copilot
github/awesome-copilot