trellis-brainstorm

Guide collaborative requirements discovery and converge on MVP scope.

52|7|Updated Apr 29, 2026
One-click install
npx skills add https://github.com/RebornQ/fl_api_hub --skill trellis-brainstorm-rebornq
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: trellis-brainstorm
Source: https://github.com/RebornQ/fl_api_hub/tree/main/.cursor/skills/trellis-brainstorm
Command: npx skills add https://github.com/RebornQ/fl_api_hub --skill trellis-brainstorm-rebornq

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Facilitates collaborative requirements discovery before implementation, enabling a shared understanding, seeds PRD, and establishes an iterative flow to converge on MVP scope when requirements are unclear or multiple approaches exist.

Core Features & Use Cases

  • Phase-driven planning: create a task directory, seed prd.md, and kick off a one-question-at-a-time interview flow to surface key decisions.
  • Evidence-first questioning: inspect code, tests, docs, and task history to answer questions with repository evidence rather than asking for facts that the project can provide.
  • iterative artifact evolution: update planning artifacts (prd.md, design.md, implement.md) as decisions and dependencies become clearer.

Quick Start

Create a Trellis planning task by generating a new task directory, seeding prd.md with initial questions and decisions, and starting the one-question-at-a-time interview flow.

Frequently Asked Questions about trellis-brainstorm

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

FAQPage Schema
How do I define MVP scope when product requirements are vague?

MVP scope is defined by running a Phase 1 planning interview that asks one high-value question at a time, turning vague requests into concrete decisions captured in prd.md. This iterative flow converges on a shared understanding of the minimal viable product.

What is the best way to start requirements discovery before implementation?

Requirements discovery starts by creating a Trellis task directory and seeding a prd.md file. The process then kicks off a collaborative interview flow to surface key decisions and establish shared understanding before any coding begins.

How do I generate a PRD from existing repository evidence?

You generate a PRD by inspecting existing code, tests, docs, and task history to answer questions with repository evidence. This evidence-first questioning approach seeds the prd.md with factual project context rather than assumptions.

How do planning artifacts evolve during product discovery?

Planning artifacts like prd.md, design.md, and implement.md evolve iteratively as decisions and dependencies become clearer. The process updates these files continuously to reflect the shared understanding achieved through the one-question-at-a-time interview flow.

Can I use this for task-management when multiple product approaches exist?

Yes, this task-management approach is designed specifically for situations where multiple approaches exist. It facilitates collaborative requirements discovery to converge on a single MVP scope by asking targeted questions and updating planning artifacts.

Why do I need to inspect repository evidence before asking requirements questions?

Inspecting repository evidence prevents asking for facts the project already provides. By checking code, tests, and docs first, the questioning flow focuses on high-value decisions that actually surface new requirements rather than redundant information.