clarify

Resolves ambiguities in feature specifications through targeted stakeholder questions.

22|Updated Apr 1, 2026
One-click install
npx skills add https://github.com/handarbeit/fabrik --skill clarify-handarbeit
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: clarify
Source: https://github.com/handarbeit/fabrik/tree/main/plugin/entwurf/skills/clarify
Command: npx skills add https://github.com/handarbeit/fabrik --skill clarify-handarbeit

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Feature specifications often contain vague language, missing details, and unresolved decisions that surface as costly rework during development. This Skill finds those gaps and resolves them by asking the specification owner up to five targeted questions, writing each answer directly back into the spec document. ## Core Features & Use Cases - Ambiguity Scanning: Assesses the spec against a coverage taxonomy spanning functional scope, data model, non-functional requirements, edge cases, and terminology, marking each area Clear, Partial, or Missing. - Guided Questioning: Asks one question at a time in plain business language, with recommended options and reasoning, so non-technical stakeholders can answer unaided. - Incremental Spec Updates: Writes every accepted answer back into the spec immediately, maintaining a Clarifications session log and re-validating the requirements checklist. - Use Case: After drafting a spec for a new reporting feature, run this Skill to discover that "fast response times" and "appropriate permissions" are undefined, resolve them into measurable criteria with the product owner, and hand engineering a testable document. ## Quick Start Ask the assistant to review the current feature specification for gaps and ambiguities and clarify what's missing before handing it to engineering.

Frequently Asked Questions about clarify

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I find gaps in a feature specification before development?

Run a clarification pass that scans the spec against a coverage taxonomy of functional scope, data model, non-functional requirements, and edge cases. Each gap is marked Clear, Partial, or Missing, and the highest-impact ones become targeted questions for the specification owner.

How many questions should a spec clarification session ask?

This workflow caps clarification at five questions per session, ranked by impact and uncertainty. Each question must be answerable by a non-technical stakeholder, either by picking from two to five options or in five words or fewer.

Does this work with GitHub Spec Kit specifications?

Yes, it follows the Spec Kit clarify workflow and locates specs via the .specify/feature.json pointer or the highest-numbered specs/NNN-* folder. It also reads .specify/memory/constitution.md to avoid asking questions the project has already settled.

What happens to the answers after a clarification session?

Each accepted answer is written directly into the spec file under a Clarifications section with a dated session heading, then applied to the relevant requirements, entities, or success criteria. The requirements checklist is re-validated and checkbox states updated to reflect the changes.

When should I not run a spec clarification pass?

Skip clarification when no spec exists yet, since the workflow will not create one, or when the work is an exploratory spike where ambiguity is acceptable. It is designed for the stage after a spec is written and before engineering builds against it.