ask-user
Gather user input effectively
AI agents often lack clarity on user requirements. This skill provides patterns for asking targeted questions that gather exactly what you need to proceed effectively.
์คํฌ ZIP ๋ค์ด๋ก๋
Claude์์ ์ ๋ก๋
์ค์ โ ๊ธฐ๋ฅ โ ์คํฌ โ ์คํฌ ์ ๋ก๋๋ก ์ด๋
ํ ๊ธ์ ์ผ๊ณ ์ฌ์ฉ ์์
ํ ์คํธํด ๋ณด๊ธฐ
"ask-user" ์ฌ์ฉ ์ค์ ๋๋ค. Make the search faster
์์ ๊ฒฐ๊ณผ:
- I understand you want faster search. To help you best, I have a few questions:
- โข What does 'faster' mean to you? (response time under 100ms, faster than current, instant)
- โข What is your current search implementation? (database full-text, external service, custom)
- โข How many records are you searching through? (hundreds, thousands, millions)
๋ณด์ ๊ฐ์ฌ
์์ This is a prompt-based documentation skill with no executable code. It provides guidance for AI agents on how to interact with users using the AskUserQuestion tool. All 19 static findings are false positives caused by the analyzer misidentifying documentation code blocks and keywords as security patterns. No scripts, network calls, filesystem access, or external commands are present.
์ํ ์์ธ
๐ ๋คํธ์ํฌ ์ ๊ทผ (1)
๐ ํ์ผ ์์คํ ์ก์ธ์ค (1)
โ๏ธ ์ธ๋ถ ๋ช ๋ น์ด (3)
ํ์ง ์ ์
๋ง๋ค ์ ์๋ ๊ฒ
Clarify project scope
Use structured questions to understand exact requirements before implementation begins
Resolve technical decisions
Present technology choices with pros and cons for user selection
Reduce clarification loops
Batch related questions and provide context to minimize user interaction cycles
์ด ํ๋กฌํํธ๋ฅผ ์ฌ์ฉํด ๋ณด์ธ์
The user said they want the app to be 'fast'. Ask them to clarify whether they mean response time, load time, or development speed.
Present three database options with their trade-offs: PostgreSQL for relational data, MongoDB for flexible schemas, Redis for caching. Ask which fits their use case.
Summarize the requirements you understood and ask the user to confirm before proceeding with the build.
Ask whether a feature should also handle edge cases or just the main use case. Provide examples of what is included versus excluded.
๋ชจ๋ฒ ์ฌ๋ก
- Always provide context about why you are asking and what depends on the answer
- Offer concrete options with descriptions rather than asking open-ended questions
- Batch related questions together to reduce interaction cycles
ํผํ๊ธฐ
- Asking 'What do you want?' without any specific options or context
- Making assumptions instead of asking for clarification
- Asking multiple unrelated questions in separate interactions instead of batching