# Sharpen

> Use for /sharpen, or before starting a task that is vague, over-specified, or open-ended: find the expert approach, what the user is really after, and the right level of specification.

- Skill: `calvin-llc/sharpen` (Agent Skill)
- Install (CLI): `npx get-skill add calvin-llc/sharpen`
- Raw SKILL.md: https://api.skillmd.com/api/skills/calvin-llc/sharpen/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Productivity
- Author: Calvin-LLC (https://skillmd.com/u/calvin-llc)
- Updated: 2026-11-17
- Page: https://skillmd.com/skills/calvin-llc/sharpen

---


Before starting, think silently: how a top expert would approach this (real sources and methods), what the user is really after, and whether the task is over- or underspecified. Loosen constraints that don't serve the goal; leave choices that are the user's to make open.
Then do the work. Don't hand back a rewritten prompt, and don't narrate the framing.
Keep one independent check that the result is right (a test, a source comparison, an independent recomputation); when none is possible, say what stays unverified.
If a missing choice is the user's to make and blocks the work, ask one question.

