skills.rest
tatsuki-washimi
tatsuki-washimiCOMMUNITY
Explore all skills by this author →
GitHub

collaborative_design

Enforce collaborative design and explicit user approval before coding work begins.

1|Updated Dec 17, 2025
Search Tags:#design#planning#review#policy#approval#ai-guidance#collaborate
One-click install
npx skills add https://github.com/tatsuki-washimi/gwexpy --skill collaborative-design-tatsuki-washimi
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: collaborative_design
Source: https://github.com/tatsuki-washimi/gwexpy/tree/main/.agent/skills/collaborative_design
Command: npx skills add https://github.com/tatsuki-washimi/gwexpy --skill collaborative-design-tatsuki-washimi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It ensures AI discussions of software design and policy before any coding work, preventing unilateral implementation and guiding collaborative decision-making with the user.

Core Features & Use Cases

  • Conversation-first design decisions: the AI proposes design options and seeks user approval before proceeding.
  • Explicit start triggers: coding work begins only after explicit user consent, with a tracked plan using setup_plan or archive_work --plan.
  • Non-destructive pre-work: non-invasive analysis of existing code, prototype verifications, and knowledge updates during dialogue.
  • Structured workflow: Listen → Propose → Refine → Commit, delivering auditable design decisions and safe handoffs.

Quick Start

Initiate a collaborative design discussion with the user and obtain explicit approval before any implementation begins.

Frequently Asked Questions about collaborative_design

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

FAQPage Schema
How do I stop AI from writing code without my approval during software design?▼

To prevent unilateral coding, collaborative design enforces conversation-first planning where the AI proposes options and seeks explicit user approval before implementation. This ensures safe, auditable handoffs.

What is the best way to manage API governance and major refactors with AI?▼

Managing API governance requires structured design reviews before changes. This approach enforces non-destructive pre-work analysis and explicit start triggers, ensuring user input guides major refactors safely.

How does the Listen Propose Refine Commit workflow handle design planning?▼

The Listen Propose Refine Commit workflow structures design planning by enforcing conversation-first proposals, tracking plans via setup_plan, and requiring explicit user consent before coding begins.

Can I review existing code non-destructively before starting an AI-driven refactor?▼

Yes, non-destructive pre-work allows non-invasive analysis of existing code and prototype verifications during dialogue. This ensures safe, auditable analysis before any implementation changes occur.

When do I need explicit user consent to trigger AI coding work?▼

Explicit user consent is required before any coding work begins. Using tracked plans with setup_plan or archive_work --plan ensures AI teams only start implementation after receiving direct approval.

Why should I use conversation-first design decisions for high-level software changes?▼

Conversation-first design decisions prevent unilateral implementation by guiding collaborative decision-making. The AI proposes options and seeks user approval, ensuring safe and auditable collaboration for major software changes.

Related Skills

Explore More Design & Creative Skills→
kenhiakenhia

ce-brainstorm

Facilitate collaborative brainstorming sessions to define feature ideas and requirements.

Community
Advanced
SteveBronderSteveBronder
1

subplan-skl

Convert approved human subtasks into executable XML task files for agents.

Community
Advanced
espensevespensev

security-scan

Audits PowerShell scripts for safe execution policies and destructive commands.

Community
Intermediate
sheehesheehe

full_research_workflow

Orchestrate end-to-end research workflows with human-in-the-loop checkpoints.

Community
Advanced