ask-questions
Skill0 starsChanged 6 months agoMITView on GitHub
Gathers structured input from the user through targeted questions with discrete options. Use when invoked via /ask-questions, or when the user says phrases like "ask me questions", "clarify first", "gather requirements", "make sure you have context", or "what do you need to know".
4 other projects use a version of this.
Adopted by
Most recently taken up by
- BrickerP/Personal-Skillstheir own version, first seen Nov 1, 2026
- LeeorNahum/ask-questions-skilltheir own version, first seen Oct 25, 2026
- alastairlundy/skillstheir own version, first seen Jul 17, 2026
- Codagent-AI/agent-skillstheir own version, first seen Apr 25, 2026
Other versions4 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 4
- Codagent-AI/agent-skills★ 31same name
- LeeorNahum/ask-questions-skill★ 2same name
- alastairlundy/skills★ 0same name
- BrickerP/Personal-Skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository7 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- commit-messageSkill
- make-todosSkill
- review-planSkill
- smoke-testSkill
- verify-changesSkill
The file
We split this skill into the rules it gives an agent
6 rules. 0 are common, 5 are distinctive, and 1 is very specific to this project.
0
Common
Most of the 4 edited versions we read keep these.
5
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Read the project's agent instructions (e.g.
- Distinctive: line 16
Do not use this skill for:
- Distinctive: line 55
If answers contradict established project conventions, conventions win.
- Distinctive: line 56
If the user says "you decide" or "just do it," proceed with your best judgment and state the assumptions you are making.
- Distinctive: line 57
Do not re-ask questions the user has already answered, even indirectly.
- Very specific: line 61
After gathering answers, chain into /make-todos to decompose the work or /explore-codebase to investigate the relevant code.
“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Ask Questions
Read the project's agent instructions (e.g. AGENTS.md) before applying this skill to ensure conventions are followed.
When to apply
- User explicitly invokes
/ask-questions - User asks you to clarify, gather requirements, or check assumptions before proceeding
- You face unknowns that would significantly change your approach
Do not use this skill for:
- Single clarification questions mid-task (just ask inline)
- Things you can determine by reading project docs, code, or git history
- Implementation details with a single obvious answer given established patterns
- Whether to run tests, lint, or verify (always do it)
Procedure
- Identify the topic from the slash command argument (e.g.
/ask-questions about the deployment strategy) or surrounding context. - Categorize your unknowns using the tiers below to prioritize what to ask.
- Present structured questions with discrete options. Use a structured-question tool (AskQuestion, AskUserQuestion) if available; otherwise present numbered text options the user can reply to by number.
- Ask all questions at once unless a later question depends on an earlier answer. Batch to minimize round trips.
- Proceed after answers: take the most appropriate next action (implement, plan, explore). Only pause if you genuinely cannot determine what to do next.
Question categories
Prioritize questions by how much they change your approach:
Tier 1 -- Blocking unknowns: Choices that change the implementation approach entirely. Ask these first.
- "Should this be a new handler or an extension of the existing one?"
- "REST API or GraphQL?"
Tier 2 -- Preference unknowns: Either option works, but the user has an opinion. Ask after Tier 1.
- "Return plain text or structured JSON?"
- "Single PR or split by concern?"
Tier 3 -- Confirmable assumptions: Things you believe are true based on code/docs, but want to verify. Ask last, or skip if confidence is high.
- "I'll follow the existing error handling pattern -- correct?"
- "This should go in the utils package based on the existing structure -- right?"
Question design
- Match option count to the actual decision space. Binary questions get 2 options -- do not inflate to 3-5 artificially.
- For genuinely open-ended option spaces, include an escape hatch ("Other" or "Agent decides").
- Allow multiple selections when choices can co-exist (e.g. "Which concerns apply?").
- One sentence per question. Front-load the decision being made.
After receiving answers
- If answers contradict established project conventions, conventions win. Explain why and ask the user to confirm they want to override.
- If the user says "you decide" or "just do it," proceed with your best judgment and state the assumptions you are making.
- Do not re-ask questions the user has already answered, even indirectly.
Related skills
- After gathering answers, chain into
/make-todosto decompose the work or/explore-codebaseto investigate the relevant code.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

