kata-research-phase

Generate a RESEARCH.md documenting stacks, patterns, and pitfalls for a phase.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/gannonh/kata-orchestrator --skill kata-research-phase
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kata-research-phase
Source: https://github.com/gannonh/kata-orchestrator/tree/main/skills/kata-research-phase
Command: npx skills add https://github.com/gannonh/kata-orchestrator --skill kata-research-phase

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps planners determine how to implement a project phase by performing standalone research or re-researching after planning. It guides the creation of a dedicated RESEARCH.md that captures standard stacks, architecture patterns, and pitfalls for a phase, ensuring research outcomes are actionable for planning.

Core Features & Use Cases

  • Standalone phase research: Investigate how to implement a phase without committing to planning.
  • Context-constrained research: Respect decisions in CONTEXT.md to narrow scope.
  • Spawn and debrief: Spawn the kata-phase-researcher agent, gather evidence from Context7, official docs, and WebSearch, and produce a structured RESEARCH.md for the planner.

Quick Start

Run phase research by supplying phase details and CONTEXT.md decisions; the agent will generate a RESEARCH.md and store it at .planning/phases/XX-name/XX-RESEARCH.md.

Frequently Asked Questions about kata-research-phase

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

FAQPage Schema
How do I document standard stacks and architecture patterns for a project phase?

You can document standard stacks and architecture patterns for a project phase by generating a structured RESEARCH.md file that captures technical decisions, pitfalls, and evidence gathered from official docs to guide planning.

What is standalone phase research used for in project planning?

Standalone phase research investigates how to implement a specific project phase without committing to a full planning cycle, producing a dedicated research artifact that outlines standard stacks and potential pitfalls.

How do I perform re-research after initial project planning?

You perform re-research after initial project planning by triggering a standalone investigation phase that re-evaluates architecture patterns and pitfalls, updating the RESEARCH.md artifact with new evidence from official documentation.

Does phase research respect existing constraints in CONTEXT.md?

Yes, phase research respects existing constraints in CONTEXT.md by enforcing those decisions during the investigation to narrow scope and ensure the generated RESEARCH.md aligns with prior architectural choices.

What is the best way to gather technical evidence for a RESEARCH.md file?

The best way to gather technical evidence for a RESEARCH.md file is to orchestrate a researcher agent that queries Context7, official documentation, and WebSearch, compiling the findings into a structured artifact.

When should I not use standalone phase research?

You should not use standalone phase research if your project phase lacks existing CONTEXT.md decisions to constrain scope, or if the phase requires immediate implementation rather than preliminary investigation into architecture patterns.