visual · syahiidkamil — Claude Code command: what it does, how to install
The visual command turns a discussion into a self-contained HTML document with diagrams, wireframes and matrices the agent implements from in plan mode.
no license ★ 406 updated 2026-11-01
.claude/commands/visual.mdWhat visual does
The command asks questions to clarify scope, approach and edge cases, then writes the result into plan.html — a single file that opens in a browser and is read by the agent. Inside: Mermaid diagrams for flows and data schemas, wireframes for UI, state and permission matrices, short prose where visuals fall short.
Run it with a description or without an argument — it will ask what to plan. If the codebase needs exploration first, it delegates to the built-in Explore subagent. The plan lands in docs/plans/{number}-{slug}/plan.html inside the project; if there is one project, it does not ask. Diagrams render via Mermaid CDN, no other dependencies.
What it helps with
- capture an architectural decision as a visual artifact
- gather questions on unclear edges before writing code
- document a UI flow with loading and error states for plan mode
- record a migration or refactor as a diagram with notes
Example prompts for the agent
- run
visualand plan the auth module refactor visual— make a plan for adding PDF export- do
visualfor the REST to GraphQL migration
How to install visual
Manually: put the files in place
.claude/commands/visual.mdCall it with /visual. For all projects use ~/.claude/commands/
Manually: put the files in place
~/.codex/prompts/visual.mdCall it with /prompts:visual. Codex marks this format deprecated in favour of skills
Manually: put the files in place
.github/prompts/visual.prompt.mdManually: put the files in place
.gemini/commands/visual.tomlGemini stores commands as TOML: the prompt text goes into the prompt field
Questions about visual
How to install visual in Claude Code?
Create a .claude/commands/visual.md file at the project root and put the command content from the repository there.
What does the visual command do?
It asks questions about the task, clarifies scope and approach, then generates plan.html with diagrams, wireframes and matrices — one self-contained file the agent works from in plan mode.
Where is the plan saved after running visual?
Into docs/plans/{number}-{slug}/plan.html inside the project. The number is the next available (01, 02, 03), the slug comes from the plan subject.
Similar slash commands
All slash commands for Agent setup: memory, context, orchestration →visual by other authors comes first.