Ask Questions If Underspecified
This one solves the classic problem of Claude diving into implementation before understanding what you actually want. When invoked, it forces a pause to ask 1-5 targeted questions about objectives, scope, constraints, and acceptance criteria before writing any code. The questions are deliberately efficient with multiple choice options, suggested defaults, and quick reply formats like "1a 2b 3c". It won't activate automatically, so you stay in control of when to use it. Most useful when you're handing off vague requirements or working in unfamiliar codebases where assumptions are expensive. The workflow explicitly blocks file edits and commands until must-have answers arrive, which feels rigid but prevents a lot of wasted work.
- skills.sh installs
- 380
- GitHub stars
- 1.2K
- last updated
- May 5, 2026
Install
$ npx -y skills add moizibnyousaf/ai-agent-skills --skill ask-questions-if-underspecified --agent claude-codeInstalls into .claude/skills of the current project.
Files
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.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageDetails
- Repository
- github.com/moizibnyousaf/ai-agent-skills
- Category
- AI & Agent Building
- License
- MIT
- First seen
- Jun 11, 2026
More from moizibnyousaf/ai-agent-skills
All 17- Browse And Evaluate177
- Review A Skill177
- Build Workspace Docs172
- Audit Library Health170
- Update Installed Skills168
- Install From Remote Library161
- Share A Library161
- Curate A Team Library160
- Migrate Skills Between Libraries158
- Backend Development874
- Database Design615
- Code Documentation569
- Best Practices531
- Llm Application Dev466
- Content Research Writer412
- Changelog Generator397
Recommended
More AI & Agent Building
agent-browsershanraisshan/claude-code-best-practiceBrowser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking scr379skills.sh installs
gen-ai-usePicsArt/gen-ai-skillsGenerate AI images, videos, audio via Picsart gen-ai CLI.367skills.sh installs
agent-developmentjezweb/claude-skillsDesign and build custom Claude Code agents with effective descriptions, tool access patterns, and self-documenting prompts. Covers Task tool delegation, model s356skills.sh installs
ai-writing-detectionmike-coulbourn/claude-vibesComprehensive AI writing detection patterns and methodology. Provides vocabulary lists, structural patterns, model-specific fingerprints, and false positive pre354skills.sh installs