Write a SKILL.md From a Workflow You Keep Repeating
Interviews you about a task you do over and over, then writes a complete, save-ready SKILL.md with a trigger description, steps, output format and stop conditions.
37 prompts
Prompts for Claude Code, Anthropic's coding agent that runs in your terminal and works across a repository. They describe the task, the files and checks involved, and what done looks like, so the agent can plan, edit and verify.
9 results
Interviews you about a task you do over and over, then writes a complete, save-ready SKILL.md with a trigger description, steps, output format and stop conditions.
Turns a pasted CLAUDE.md or AGENTS.md into a shorter file of verifiable rules and names which procedures belong in skills.
A SKILL.md that makes Claude write and edit in your brand's voice: it reads a VOICE.md in the project, or builds one from samples, then checks every draft against it.
Has Codex inspect your repository and write a short, accurate AGENTS.md with the real build, test and lint commands, conventions and boundaries, checked by actually running them.
A ready-to-save SKILL.md that makes your coding agent write Conventional Commits messages and pull request descriptions from the staged diff, never from guesses.
A SKILL.md that turns any meeting transcript or rough notes into a decision log, owner-and-date action items, open questions and a ready-to-send recap email.
A SKILL.md that writes release notes from the commits between two tags, split into a plain-language section for users and a technical changelog for developers.
A SKILL.md that has your agent profile a messy CSV or spreadsheet, fix it in a copy with a script, and report every change so the cleaned data can be trusted.
A SKILL.md that forces your coding agent to reproduce a bug with a failing test before changing code, apply the smallest fix, and prove it with the test suite.
Start claude in the repository's root folder and paste the prompt as your first message. Fill in the blanks with real paths, commands and limits first; the agent can't guess them.
Only rules that apply to every task, such as build commands and code style, belong in CLAUDE.md. Keep one-off task prompts in the chat so they don't steer later work.
Most do. Codex, Cursor and Copilot's agent mode follow the same kind of task brief; swap any mention of CLAUDE.md for that tool's instruction file, such as AGENTS.md.