flow-next-interview

Conduct structured interviews with 40+ questions to refine specifications via flowctl.

Updated Jan 15, 2026

Search Tags:#documentation #interview #epic #task #flowctl #spec-refinement

Ready to run

npx skills add https://github.com/raydocs/inktale --skill flow-next-interview-raydocs

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: flow-next-interview
Source: https://github.com/raydocs/inktale/tree/main/plugins/flow-next/skills/flow-next-interview
Command: npx skills add https://github.com/raydocs/inktale --skill flow-next-interview-raydocs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Conduct in-depth interviews to flesh out epic, task, or spec details and produce refined, actionable implementation plans before building. This approach reduces ambiguity and misalignment by guiding a structured questioning process and recording decisions in a centralized flow-tracking system.

Core Features & Use Cases

Structured, in-depth interviews (40+ questions) to convert vague ideas into complete specifications.
Supports Epic, Task, or file-based inputs; fetches context, records decisions in the .flow directory, and writes refined plans back via flowctl.
Outputs refined specifications or task descriptions ready for implementation, including acceptance criteria and edge-case considerations.

Quick Start

Start by providing a Flow epic/task ID or a target file path to begin the interview and refine the spec with 40+ questions.

Frequently Asked Questions about flow-next-interview

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

FAQPage Schema

How do I refine vague epic or task specifications into actionable plans?▼

Spec refinement through structured interviews uses 40+ questions to extract requirements and record decisions in flow tracking. It outputs refined specifications with acceptance criteria and edge-case considerations.

What is the best way to elicit requirements before starting implementation?▼

Requirement elicitation via in-depth interviews fetches current context and records decisions in a centralized .flow directory. This structured questioning process reduces ambiguity before building begins.

Can I use flowctl to write refined task descriptions back to a local directory?▼

Yes, flowctl writes refined task descriptions and plans back to the .flow directory. The system enforces flowctl to ensure all task state persists locally and avoids global installations.

Does spec refinement work with file-based specifications or only epic and task IDs?▼

Spec refinement works with Epic IDs, Task IDs, and file-based specifications. The system fetches current context from the provided input and applies the structured interview process to produce refined documentation.

Do I need to install flowctl globally to conduct in-depth spec interviews?▼

No, global installation is not required. The system enforces the bundled flowctl tool and ensures all task state persists in the local .flow directory, avoiding global installations.

What are the limitations of automated requirement elicitation for complex epics?▼

Automated requirement elicitation relies on the .flow directory for state persistence and requires a valid epic/task ID or target file path. Without initial context, the structured questioning process cannot begin effectively.