gsd-discuss-phase

Load project documents and repository patterns to produce a phase-numbered CONTEXT.md.

Updated Jan 25, 2026
One-click install
npx skills add https://github.com/MINT-IA/MINT --skill gsd-discuss-phase-mint-ia
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gsd-discuss-phase
Source: https://github.com/MINT-IA/MINT/tree/main/.claude/skills/gsd-discuss-phase
Command: npx skills add https://github.com/MINT-IA/MINT --skill gsd-discuss-phase-mint-ia

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill collects and clarifies the unresolved implementation questions and contextual constraints for a given project phase so downstream researcher and planner agents can act without repeated clarifications.

Core Features & Use Cases

  • Prior-context loading: Automatically load PROJECT.md, REQUIREMENTS.md, STATE.md, and previous CONTEXT.md files to avoid re-asking decided questions.
  • Codebase scouting: Search the repository for reusable assets and patterns relevant to the phase.
  • Adaptive discussion: Identify gray areas, present them to the user for selection, and deep-dive into each selected area via interactive questions or automated defaults.
  • Decision capture: Produce a phase-numbered CONTEXT.md that records concrete implementation decisions for downstream research and planning.
  • Use Case: When preparing a new development sprint, run this Skill to convert scattered requirements and partial decisions into a single actionable context file for engineers and planners.

Quick Start

Run gsd-discuss-phase for the target phase number with optional flags like --auto or --chain to generate a phase-numbered CONTEXT.md that captures actionable implementation decisions.

Frequently Asked Questions about gsd-discuss-phase

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

FAQPage Schema
How do I capture phase-specific implementation decisions before starting sprint planning?

Loading prior project artifacts like PROJECT.md and REQUIREMENTS.md, then interactively resolving gray areas, captures phase-specific implementation decisions. This produces a phase-numbered CONTEXT.md with actionable constraints for downstream researchers and planners.

What is phase context gathering and when do I need it in a software workflow?

Phase context gathering collects unresolved implementation questions and contextual constraints for a specific development phase. You need it when prior artifacts like PROJECT.md and STATE.md exist but require clarification before research and planning tasks can proceed.

How do I generate a CONTEXT.md file from existing project requirements and state files?

Generate a CONTEXT.md file by running an interactive workflow that loads PROJECT.md, REQUIREMENTS.md, STATE.md, and previous CONTEXT.md files. The workflow scouts the codebase for patterns, identifies gray areas, and records your decisions in the phase-numbered output file.

Do I need existing repository artifacts like PROJECT.md and REQUIREMENTS.md to use interactive phase planning?

Yes, interactive phase planning requires prior artifacts like PROJECT.md, REQUIREMENTS.md, STATE.md, and earlier CONTEXT.md files to exist. The workflow analyzes these documents and repository patterns to avoid re-asking previously decided questions and prepare actionable research tasks.

Can I automate context gathering and decision capture for a development sprint without manual interaction?

You can automate context gathering and decision capture by running the workflow with the --auto flag. This applies automated defaults to identified gray areas instead of interactive questions, still producing a phase-numbered CONTEXT.md containing actionable implementation decisions for downstream agents.

What's the best way to resolve gray areas in software project requirements before handing off to planner agents?

The best way to resolve gray areas before handing off to planner agents is to use an adaptive discussion workflow that identifies unresolved questions and presents them for user selection. It deep-dives into each selected area via interactive questions, then outputs a CONTEXT.md with concrete implementation decisions.