trellis-brainstorm

Guides collaborative requirements discovery and converges on an MVP scope before implementation.

Updated Apr 4, 2026
One-click install
npx skills add https://github.com/elebirds/panoptes --skill trellis-brainstorm-elebirds
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: trellis-brainstorm
Source: https://github.com/elebirds/panoptes/tree/main/.opencode/skills/trellis-brainstorm
Command: npx skills add https://github.com/elebirds/panoptes --skill trellis-brainstorm-elebirds

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Starting implementation with vague or evolving requirements leads to rework, scope creep, and misaligned solutions. This Skill structures the pre-implementation phase by capturing ideas into a task directory immediately, asking only high-value questions one at a time, and converging on a clearly scoped MVP with explicit acceptance criteria. ## Core Features & Use Cases - Task-first capture: Creates a task directory and seeds a PRD immediately so no idea is lost, then refines it through structured Q&A. - Research-first technical decisions: Delegates technology and pattern research to sub-agents that persist findings to research files, then presents 2-3 concrete approaches with trade-offs. - Diverge-then-converge workflow: Expands thinking across future evolution, related scenarios, and edge cases before locking an MVP scope with explicit out-of-scope items and an ADR-lite decision record. - Use Case: When a user describes a new feature like "add multiplayer matchmaking", the Skill creates the task, inspects the repo, researches comparable approaches, asks targeted preference questions, and produces a confirmed PRD with an implementation plan split into small PRs. ## Quick Start Ask the AI to brainstorm requirements for a new feature or complex task before writing any code.

Frequently Asked Questions about trellis-brainstorm

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

FAQPage Schema
How do I gather requirements before starting implementation?▼

Create a task directory with a seeded PRD capturing the goal, known facts, assumptions, and open questions. Then ask one high-value question at a time, updating the PRD after each answer until scope, acceptance criteria, and out-of-scope items are confirmed.

What questions should I ask during requirements discovery?▼

Ask only blocking or preference questions that cannot be derived from repo inspection, docs, or research. Prefer multiple-choice formats presenting 2-3 concrete approaches with trade-offs, and prioritize MVP scope boundaries first.

How do I research technical options without polluting the main conversation?▼

Delegate each research topic to a dedicated sub-agent that persists findings to a research markdown file in the task directory. The main conversation then reads only the summarized files and derives feasible approaches from them.

When should brainstorming be skipped for a coding task?▼

Skip brainstorming for trivial changes like single-line fixes or typos, and implement directly. For simple tasks with a clear goal affecting one or two files, ask a single confirmation question before implementing.

What are the limitations of a question-driven discovery process?▼

It adds upfront conversation time, which is wasteful for trivial or well-defined tasks. It also depends on the user answering preference questions, so fully autonomous execution without any user input is not supported.