speckit.clarify

Detects underspecified areas in feature specs and records targeted clarification answers.

Updated Aug 20, 2026
One-click install
npx skills add https://github.com/hduoc2003/leviathan-guardian --skill speckit-clarify-hduoc2003
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: speckit.clarify
Source: https://github.com/hduoc2003/leviathan-guardian/tree/main/.agents/skills/speckit-clarify
Command: npx skills add https://github.com/hduoc2003/leviathan-guardian --skill speckit-clarify-hduoc2003

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Feature specifications often contain ambiguous requirements, missing decisions, and vague language that cause rework during planning and implementation. This Skill systematically scans a spec for gaps and resolves them through focused clarification questions before planning begins. ## Core Features & Use Cases - Structured Ambiguity Scan: Evaluates the spec across a taxonomy covering functional scope, data model, UX flows, non-functional requirements, integrations, edge cases, constraints, and terminology. - Interactive Questioning Loop: Asks up to 5 high-impact questions one at a time, each with a recommended option or suggested answer based on best practices. - Incremental Spec Updates: Writes each accepted answer into a ## Clarifications section and updates the relevant spec sections immediately, keeping the document consistent. - Use Case: Before running a planning step on a new feature spec, run this Skill to resolve open questions about security posture, data entities, and acceptance criteria so the plan is built on firm decisions. ## Quick Start Ask the assistant to clarify the current feature spec and resolve any ambiguities before planning.

Frequently Asked Questions about speckit.clarify

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

FAQPage Schema
How do I clarify ambiguities in a feature specification?▼

Run the clarify workflow on the active spec file. It scans categories like functional scope, data model, and non-functional requirements, then asks up to five targeted questions and writes the answers directly into the spec.

When should I run spec clarification in a spec-driven workflow?▼

Run clarification after creating the feature spec and before generating the technical plan. Skipping it is allowed for exploratory spikes, but downstream rework risk increases when ambiguities reach the planning phase.

How many clarification questions does the workflow ask?▼

It asks a maximum of five questions per session, selected by impact and uncertainty. Each question offers multiple-choice options with a recommendation or a short-answer format constrained to five words.

What happens if the feature spec file is missing?▼

The workflow aborts and instructs you to run the specify step first. It does not create a new spec, since clarification only operates on an existing feature specification.

Where are clarification answers recorded in the spec?▼

Accepted answers are appended as bullets under a Clarifications section with a dated session subheading, and the relevant spec sections such as Functional Requirements or Data Model are updated in place.