Overview

MUXAI Wiki is the shared knowledge base for designer-agent harness work: how your team configures projects, runs the workflow loop, installs skills, and stays safe while building with agents.

Four pillars

Pillar	Tab	What it answers
Process	Workflow	How do we go from idea to shipped UI, step by step?
Infrastructure	Setup	How do we configure the harness once per project?
Capabilities	Skills	Which skills do we install and when?
Vocabulary	Glossary	What do agent and dev terms mean for designers?

Org essentials: Collaboration (PRs and handoff) · Templates (starter AGENTS.md, design.md, PRODUCT.md)

Setup is static — run per project. Workflow is dynamic — repeat every feature.

The four-stage loop

1. Discovery — research, ideation, brainstorming
2. Context Artifacts — .md files and reference designs the agent re-reads
3. Setup Harness — IDE, rules, skills, design docs (Setup)
4. Iteration — build, review, PR, merge (Collaboration)

Learnings from Iteration feed back into Discovery.

Who this is for

Designers and design ops on teams adopting agent-assisted development — with enough structure that output stays on-brand, reviewable, and safe for org repos.

How to use it

1. Read Governance — align with your org’s tool and data policy
2. Read Workflow — understand the loop
3. Copy Templates into new projects; run Setup
4. Install Skills from the tabbed catalog
5. Follow Collaboration when opening PRs
6. Update this wiki when the loop teaches something new

Maintaining the wiki

Responsibility	Suggested owner
Governance page	Design ops + security liaison
Skills catalog	MUXAI champion
Glossary	Any designer hitting new terms
Templates	Design ops

Propose changes via pull request. See CONTRIBUTING.md on GitHub.

Content ships as Markdown in this repo; push to main to publish on GitHub Pages.