Agent skill

ask-questions-if-underspecified

Ask clarifying questions when requirements are underspecified

Stars 115
Forks 18

Install this agent skill to your Project

npx add-skill https://github.com/alfredolopez80/multi-agent-ralph-loop/tree/main/.claude/skills/ask-questions-if-underspecified

SKILL.md

Skill: Ask Questions If Underspecified

v2.88 Key Changes (MODEL-AGNOSTIC)

  • Model-agnostic: Uses model configured in ~/.claude/settings.json or CLI/env vars
  • No flags required: Works with the configured default model
  • Flexible: Works with GLM-5, Claude, Minimax, or any configured model
  • Settings-driven: Model selection via ANTHROPIC_DEFAULT_*_MODEL env vars

ultrathink - Take a deep breath. We're not here to write code. We're here to make a dent in the universe.

The Vision

Clarity is the foundation of inevitable solutions. Every question should narrow the path to truth.

Your Work, Step by Step

  1. Analyze requirements: Identify missing inputs and ambiguities.
  2. Separate MUST vs NICE: Block on essentials, assume the rest.
  3. Ask precisely: Short, concrete questions with defaults.
  4. Confirm understanding: Summarize before proceeding.

Ultrathink Principles in Practice

  • Think Different: Question hidden assumptions.
  • Obsess Over Details: Align questions with real constraints.
  • Plan Like Da Vinci: Build the question set before asking.
  • Craft, Don't Code: Precision over volume.
  • Iterate Relentlessly: Refine questions as context evolves.
  • Simplify Ruthlessly: Ask only what matters.

Purpose

Ensure task clarity BEFORE implementation by identifying ambiguities.

When to Use

  • ANY new task or feature request
  • Complex modifications
  • Unclear requirements

Process

1. Analyze Requirements

Identify:

  • Missing technical details
  • Unclear scope boundaries
  • Ambiguous terminology
  • Unstated assumptions

2. Categorize Questions

MUST_HAVE (Blocking)

Questions that BLOCK implementation until answered:

  • Critical architecture decisions
  • Security requirements
  • Data model choices
  • Integration points

NICE_TO_HAVE (Assumptions)

Questions where you can make reasonable assumptions:

  • UI/UX preferences
  • Performance targets
  • Edge case handling

3. Output Format

markdown
## 🔍 Clarification Needed

### MUST_HAVE (Please answer before I proceed):
1. [Critical question 1]
2. [Critical question 2]

### NICE_TO_HAVE (I'll assume these if not specified):
- [Optional question] → I'll assume: [default value]
- [Optional question] → I'll assume: [default value]

### My Understanding:
[Summarize what you understand so far]

4. Wait for Answers

DO NOT proceed with implementation until MUST_HAVE questions are answered.

Examples

Good Clarification

MUST_HAVE:
1. Should auth support both email/password AND OAuth providers?
2. What's the session timeout requirement?

NICE_TO_HAVE:
- Rate limiting? → I'll assume: 100 req/min
- Password complexity? → I'll assume: min 8 chars, 1 number, 1 special

Bad (Too Vague)

What do you want?
Can you give more details?

Expand your agent's capabilities with these related and highly-rated skills.

alfredolopez80/multi-agent-ralph-loop

browser-test

Browser testing using Chrome DevTools MCP and Playwright for visual verification. Start dev server, navigate, screenshot, Lighthouse audit, console errors, network check. Use when: (1) verifying frontend changes, (2) accessibility auditing, (3) performance testing, (4) visual regression. Triggers: /browser-test, 'test in browser', 'visual test', 'lighthouse audit'.

115 18
Explore
alfredolopez80/multi-agent-ralph-loop

crafting-effective-readmes

Use when writing or improving README files. Not all READMEs are the same — provides templates and guidance matched to your audience and project type.

115 18
Explore
alfredolopez80/multi-agent-ralph-loop

stop-slop

A skill for removing AI-generated writing patterns ('slop') from prose. Eliminates telltale signs of AI writing like filler phrases, excessive hedging, overly formal language, and mechanical sentence structures. Use when: writing content that should sound human and natural, editing AI-generated drafts, cleaning up prose for publication, or any content that needs to sound authentic rather than AI-generated. Triggers: 'stop-slop', 'remove AI tells', 'clean up prose', 'make it sound human', 'edit AI writing'.

115 18
Explore
alfredolopez80/multi-agent-ralph-loop

bugs

Bug hunting with Codex CLI Use when: (1) /bugs is invoked, (2) task relates to bugs functionality.

115 18
Explore
alfredolopez80/multi-agent-ralph-loop

adversarial

Multi-Agent Adversarial Analysis System for code security

115 18
Explore
alfredolopez80/multi-agent-ralph-loop

senior-software-engineer

Global skill enforcing senior software engineering best practices

115 18
Explore

Didn't find tool you were looking for?

Be as detailed as possible for better results