Agent skill
spec-clarify
Ask targeted questions to clarify scope, edge cases, and acceptance before coding.
Stars
163
Forks
31
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/data/spec-clarify
SKILL.md
Spec Clarify
Objective
- Surface ambiguities and edge cases early.
- Confirm inputs, outputs, and constraints before implementation.
When to use
- Requirements are vague, conflicting, or incomplete.
- Before starting on a new feature or significant change.
Inputs
- User request and any provided specs.
- AGENTS.md and applicable rules.
- Existing behavior inferred from code/tests.
Process
- Restate the goal and current understanding.
- List assumptions explicitly.
- Ask precise questions on inputs/outputs, data shapes, edge cases, and metrics.
- Highlight risks of proceeding without answers.
- Propose a minimal path once questions are resolved.
Outputs
- Concise list of questions and assumptions.
- Risks of proceeding without answers.
- Suggested next step once clarified.
Edge cases
- If blocking ambiguity exists, do not proceed to implementation; ask first.
- Keep questions specific and answerable; avoid broad surveys.
Didn't find tool you were looking for?