Skip to content
mcp/skillhub

Agent Skill

Prompt Master skill

prompt-masternidhinjs/prompt-masterSKILL.md

Generates optimized prompts for AI tools. Activates only when the user explicitly asks to write, fix, improve, or adapt a prompt for a specific AI tool (LLM, Cursor, Midjourney, image AI, video AI, coding agents, etc.). Does not activate for general conversation, coding tasks, document writing, or other non-prompt-engineering work.

15kMIT

always on

77 tokname + description

once used

6,804 tokSKILL.md body

bundled files

222 KB

repo stars

15kGitHub

Install the Prompt Master skill

Skills CLI (Claude Code, Codex, Cursor and others)

npx skills add https://github.com/nidhinjs/prompt-master --skill prompt-master

Claude Code, by hand (all projects)

git clone --depth 1 https://github.com/nidhinjs/prompt-master.git /tmp/prompt-master
mkdir -p ~/.claude/skills
cp -R /tmp/prompt-master ~/.claude/skills/prompt-master

For one project only, copy the folder into .claude/skills/ in that project instead. The agent picks it up when a task matches its description.

What it costs your context

Agent Skills load in three steps. Tokens counted with the o200k tokenizer; your model's count will differ a little.

WhenWhat loadsCost
Every sessionName and description77 tok
When the agent uses itThe SKILL.md instructions6,804 tok
Only if opened2 bundled files (22 KB)varies

Until it is used, prompt-master costs about 18× less than the median MCP server we measured (1,362 tokens on every turn). Skills vs MCP study

Checks

  • The instructions are over 5,000 tokens, the Agent Skills specification's guideline for SKILL.md.

Checked 2026-10-22 · method

What's inside

The headings of its SKILL.md. Read the full file on GitHub.

  • PRIMACY ZONE — Identity, Hard Rules, Output Lock
  • MIDDLE ZONE — Execution Logic, Tool Routing, Diagnostics
  • Intent Extraction
  • Tool Routing
  • Model Recency Gate
  • Credential Safety
  • Input Sanitization -- Pasted Prompts
  • Diagnostic Checklist
  • Memory Block
  • Safe Techniques — Apply Only When Genuinely Needed
  • Agentic Output Warning
  • RECENCY ZONE — Verification and Success Lock
  • Reference Files