Can I use this for existing projects?

Yes, it is highly effective for documenting new features or refactors within existing projects by defining the specific 'Intent' of the change before implementation begins.

What is a 'Scope Guard'?

A Scope Guard is a mechanism that checks every proposed feature or decision against the 'Anchor'—a one-sentence purpose statement—to ensure the project stays focused and avoids unnecessary bloat.

What is the difference between Phase A and Phase B?

Phase A focuses exclusively on questioning and recording decisions in a flat file (decisions.md), while Phase B is an explicit step that organizes those decisions into a structured, line-limited technical specification (INTENT.md).

Why does the skill enforce a line budget for documentation?

Maintaining a line budget (typically 300-500 lines) ensures that specifications remain high-density and readable. This helps AI agents and human developers process the context more effectively without getting lost in verbose descriptions.

Intent Interview

Name: Intent Interview
Author: ArcBlock

byArcBlock

•

学習とドキュメンテーション

Transforms vague product ideas into structured, implementation-ready technical specifications through a guided interviewing process.

Intent Interview is a specialized skill designed for Intent-Driven Development (IDD), helping teams move from abstract concepts to concrete technical documentation. It employs a rigorous two-phase workflow: first, it conducts a structured deep-dive interview to capture architectural decisions and edge cases in a decisions.md file; second, it synthesizes these decisions into a concise, budget-aware INTENT.md specification. By enforcing 'Scope Guards' and strict line-count limits, the skill prevents requirement bloat and ensures that every technical requirement is traceable to a core purpose, making it ideal for seeding complex tasks for AI coding agents.

主な機能

01Two-phase workflow separating raw decision-making from formal specification composition

027 GitHub stars

03Budget-constrained output generation to keep specifications concise and readable

04Systematic deep-dive questioning covering data, rendering, architecture, and edge cases

05Contradiction resolution to identify and fix conflicting technical choices early

06Automated 'Scope Guard' that flags features drifting from the project's core anchor

ユースケース

01Defining MVP boundaries and API contracts before writing a single line of code

02Standardizing requirements gathering within teams using the AINE methodology

03Converting high-level product ideas into actionable technical specs for developers

主な機能

01Two-phase workflow separating raw decision-making from formal specification composition

027 GitHub stars

03Budget-constrained output generation to keep specifications concise and readable

04Systematic deep-dive questioning covering data, rendering, architecture, and edge cases

05Contradiction resolution to identify and fix conflicting technical choices early

06Automated 'Scope Guard' that flags features drifting from the project's core anchor

ユースケース

01Defining MVP boundaries and API contracts before writing a single line of code

02Standardizing requirements gathering within teams using the AINE methodology

03Converting high-level product ideas into actionable technical specs for developers