Can this skill handle variables from earlier development phases?

Yes, it uses Question Piping to inject variables like project goals, constraints, or technical approaches directly into the current question text for better continuity.

How does it prevent asking the same question twice?

It utilizes Parameter Chain Logic to scan the .progress.md file for existing answers before posing a question, automatically skipping any topics that have already been resolved in previous phases.

What happens if I select the 'Other' option during an interview?

The framework triggers an Adaptive Depth loop, generating context-specific follow-up questions tailored to your unique input instead of using generic prompts.

What is the Interview Framework skill?

It is a specialized capability for Claude Code that provides a standardized, adaptive algorithm for interviewing users during the technical specification process to ensure high-quality project requirements.

Interview Framework

Name: Interview Framework
Author: tzachbon

bytzachbon

•

144

•

生产力与工作流

Standardizes AI-led requirement gathering through an adaptive, single-question loop to build comprehensive technical specifications.

The Interview Framework is a core component of the ralph-specum plugin designed to streamline the discovery phase of spec-driven development. It implements a sophisticated, single-question interview loop that dynamically adjusts its depth based on user input, ensuring Claude gathers all necessary technical context without being repetitive. By leveraging a parameter chain that remembers previous answers and piping existing context into new questions, it creates a cohesive and efficient planning experience that eliminates redundant questioning and focuses on resolving project ambiguities.

主要功能

01Intelligent parameter chaining to skip previously answered questions

02Strict UX rules limiting choices to 4 options for faster decision making

03144 GitHub stars

04Adaptive single-question loop with dynamic follow-up logic

05Dynamic variable piping for context-aware questioning using .progress.md

06Semantic key mapping for standardized state persistence across development phases

使用场景

01Automating the discovery phase of architectural and system design

02Maintaining structured progress logs and context for complex development tasks

03Streamlining technical requirement gathering for new software projects

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add tzachbon/smart-ralph interview-framework

For use in Claude.ai and ChatGPT

主要功能

01Intelligent parameter chaining to skip previously answered questions

02Strict UX rules limiting choices to 4 options for faster decision making

03144 GitHub stars

04Adaptive single-question loop with dynamic follow-up logic

05Dynamic variable piping for context-aware questioning using .progress.md

06Semantic key mapping for standardized state persistence across development phases

使用场景

01Automating the discovery phase of architectural and system design

02Maintaining structured progress logs and context for complex development tasks

03Streamlining technical requirement gathering for new software projects

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add tzachbon/smart-ralph interview-framework

For use in Claude.ai and ChatGPT