Skip to main content
skill CX Codex

ask-questions-if-underspecified — Claude Code skill: clarify before

The skill makes the agent ask 1–5 clarifying questions before it starts work, so you don't get the wrong thing built. Invoked manually.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add MoizIbnYousaf/ai-agent-skills --skill ask-questions-if-underspecified -a claude-code
Open on GitHub

What ask-questions-if-underspecified does

This skill kicks in when a request is vague. The agent checks six things: the objective, what counts as done, scope, constraints, environment, and how reversible the change is. If several readings of the task are plausible, it treats the request as underspecified.

Then comes a short round of questions. They're numbered, answers are lettered, so you can reply with something compact like 1b 2a 3c. The recommended option is marked as the default, and there's a fast path: reply defaults to accept them all. Until answers arrive, the agent won't edit files or run commands. Only a low-risk look around is allowed, such as reading repo structure or config files.

Ask it to proceed anyway and the skill wants your assumptions written out as a numbered list first, with confirmation before any work. It also rules out two habits: asking things a quick config or docs read would answer, and asking open-ended questions where a multiple-choice would settle it faster. Nothing here fires on its own, so in day-to-day use the skill stays quiet until you call it.

What it helps with

  • pin down acceptance criteria before the agent writes code
  • split one vague request into distinct interpretations
  • set scope: which files are in and which stay untouched
  • surface runtime and build environment targets
  • talk through data migration risk and rollback

Example prompts for the agent

  • Use the ask-questions-if-underspecified skill on my request to add caching to the API, then wait for my answers
  • ask-questions-if-underspecified: what do you need to know before refactoring the payments module?
  • Run ask-questions-if-underspecified on my schema migration task and hold off until I reply

How to install ask-questions-if-underspecified

1Skills CLI (Vercel)

npx skills add MoizIbnYousaf/ai-agent-skills --skill ask-questions-if-underspecified -a claude-code

2Manually: put the files in place

~/.claude/skills/ask-questions-if-underspecified/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/ask-questions-if-underspecified/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about ask-questions-if-underspecified

How do I install ask-questions-if-underspecified in Claude?

Run npx skills add MoizIbnYousaf/ai-agent-skills --skill ask-questions-if-underspecified -a claude-code. The skill lands at ~/.claude/skills/ask-questions-if-underspecified/SKILL.md.

What does ask-questions-if-underspecified do?

It checks whether a request has enough detail to act on, then asks 1–5 questions with answer options. No files get touched until you reply.

Does it run automatically?

No. Its description says it is not used automatically and only runs when invoked explicitly.

Source text

SKILL.md · MIT
---
name: ask-questions-if-underspecified
description: Clarify requirements before implementing. Do not use automatically, only when invoked explicitly.
version: 4.1.0
---

# Ask Questions If Underspecified

## Goal

Ask the minimum set of clarifying questions needed to avoid wrong work; do not start implementing until the must-have questions are answered (or the user explicitly approves proceeding with stated assumptions).

## Workflow

### 1) Decide whether the request is underspecified

Treat a request as underspecified if after exploring how to perform the work, some or all of the following are not clear:
- Define the objective (what should change vs stay the same)
- Define "done" (acceptance criteria, examples, edge cases)
- Define scope (which files/components/users are in/out)
- Define constraints (compatibility, performance, style, deps, time)
- Identify environment (language/runtime versions, OS, build/test runner)
- Clarify safety/reversibility (data migration, rollout/rollback, risk)

If multiple plausible interpretations exist, assume it is underspecified.

### 2) Ask must-have questions first (keep it small)

Ask 1-5 questions in the first pass. Prefer questions that eliminate whole branches of work.

Make questions easy to answer:
- Optimize for scannability (short, numbered questions; avoid paragraphs)
- Offer multiple-choice options when possible
- Suggest reasonable defaults when appropriate (mark them clearly as the default/recommended choice; bold the recommended choice in the list, or if you present options in a code block, put a bold "Recommended" line immediately above the block and also tag defaults inside the block)
- Include a fast-path response (e.g., reply `defaults` to accept all recommended/default choices)
- Include a low-friction "not sure" option when helpful (e.g., "Not sure - use default")
- Separate "Need to know" from "Nice to know" if that reduces friction
- Structure options so the user can respond with compact decisions (e.g., `1b 2a 3c`); restate the chosen options in plain language to confirm

### 3) Pause before acting

Until must-have answers arrive:
- Do not run commands, edit files, or produce a detailed plan that depends on unknowns
- Do perform a clearly labeled, low-risk discovery step only if it does not commit you to a direction (e.g., inspect repo structure, read relevant config files)

If the user explicitly asks you to proceed without answers:
- State your assumptions as a short numbered list
- Ask for confirmation; proceed only after they confirm or correct them

### 4) Confirm interpretation, then proceed

Once you have answers, restate the requirements in 1-3 sentences (including key constraints and what success looks like), then start work.

## Question templates

- "Before I start, I need: (1) ..., (2) ..., (3) .... If you don't care about (2), I will assume ...."
- "Which of these should it be? A) ... B) ... C) ... (pick one)"
- "What would you consider 'done'? For example: ..."
- "Any constraints I must follow (versions, performance, style, deps)? If none, I will target the existing project defaults."
- Use numbered questions with lettered options and a clear reply format

```text
1) Scope?
a) Minimal change (default)
b) Refactor while touching the area
c) Not sure - use default
2) Compatibility target?
a) Current project defaults (default)
b) Also support older versions: <specify>
c) Not sure - use default

Reply with: defaults (or 1a 2a)
```

## Anti-patterns

- Don't ask questions you can answer with a quick, low-risk discovery read (e.g., configs, existing patterns, docs).
- Don't ask open-ended questions if a tight multiple-choice or yes/no would eliminate ambiguity faster.

---

*Originally created by [@thsottiaux](https://x.com/thsottiaux)*

© MoizIbnYousaf · MIT · MoizIbnYousaf/ai-agent-skills

Similar skills

All skills for Agent setup: memory, context, orchestration →

From the same set6

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor