From prompt-the-prompter
You're about to hand a task to a coding agent and want the prompt to actually land. Trigger: Write a prompt for a coding agent to <task>: include file:line anchors, acceptance criteria, and a verify-first step.
How this skill is triggered — by the user, by Claude, or both
Slash command
/prompt-the-prompter:prompt-the-prompterThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
You're about to hand a task to a coding agent and want the prompt to actually land.
You're about to hand a task to a coding agent and want the prompt to actually land.
Write a prompt for a coding agent to <task>: include file:line anchors, acceptance criteria, and a verify-first step.
A ready-to-paste agent prompt that names the files and lines to touch, states explicit acceptance criteria, and builds in a verify-first step so the agent confirms the problem exists before it changes anything.
The generalised discipline:
Confirm the skill actually fires — in your own session, on your own data. No data leaves your machine.
Does it fire? Paste the prompt in demo-input.md. You should get the shape shown in expected-output.md.
On your data. Run it on your real input.
Confirm the shape. Save the output to a file and run:
node ${CLAUDE_SKILL_DIR}/scripts/check.mjs <your-output-file>
It reports ENGAGED when the output carries this skill's characteristic structure — the deterministic sign it fired. It checks shape, not quality; judge the content yourself.
The check runs entirely locally — nothing leaves your machine. If you want to add to the public "confirmed working" count, opt in explicitly:
node ${CLAUDE_SKILL_DIR}/scripts/check.mjs <your-output-file> --report https://skills-field-guide-proxy.nesper6311.workers.dev --skill prompt-the-prompter
That sends only an anonymous +1 — never your input, output, or identity.
Installed from the public field guide. Customize freely — this is a starting point built from the published projection of prompt-the-prompter.
Provides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub marmalade7929/skills-registry-site --plugin prompt-the-prompter