feature-vision_skill

This skill helps teams generate lean vision documents for features by guiding discovery, defining scope, and producing vision, future, and discussion summaries.
  • Shell

31

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill steveclarke/dotfiles --skill feature-vision

  • SKILL.md10.6 KB

Overview

This skill creates lean, structured vision documents by guiding you through a five-phase discovery and writing process. It automates a TODO list, asks focused single questions, captures discussion context, and generates three output files: vision.md, future.md, and discussion-summary.md. The goal is a concise, readable vision that bridges to requirements without spec-level detail.

How this skill works

The skill initializes a five-phase TODO list and marks Phase 1 as in_progress, then prompts for the feature directory and any tagged notes. It walks you through Phase 1–4 with one focused question at a time, recording answers and technical discussions into a living discussion-summary. After discovery, it prunes redundancy and generates three artifacts: a short vision.md, a detailed future.md, and a discussion-summary.md that logs decisions, technical context, and open questions.

When to use it

  • Starting a new feature or initiative that needs a clear product vision and scope
  • Reworking or clarifying an existing feature before moving to requirements/specs
  • Onboarding a new teammate so they can see decisions, context, and future ideas
  • When you need a concise document to hand off to product or engineering for requirements planning

Best practices

  • Answer one question at a time and keep responses focused and conversational
  • Tag any existing notes with @filename.md so the tool can review context before asking questions
  • Be concrete about v1 scope: list what is included and what is explicitly excluded
  • Capture technical feasibility discussions even if they won't appear in vision.md—put them in discussion-summary.md
  • Ruthlessly eliminate redundant text when finishing the vision document

Example use cases

  • Create a vision for a cross-machine sync feature that aligns all team machines and tools
  • Draft a focused vision for a new UI flow without drifting into implementation details
  • Collect stakeholder input and technical constraints for a backend feature before writing requirements
  • Produce a future.md roadmap of nice-to-have features organized by phase

FAQ

It generates three files: vision.md (lean vision), future.md (detailed future work), and discussion-summary.md (living record of discussions and decisions).

How long should the generated vision.md be?

Keep it lean: typically 100–150 lines and never more than 200; the skill enforces brevity and removes redundancy.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational
feature-vision skill by steveclarke/dotfiles | VeilStrat