Skip to content
Agentmart
Submit a tool

control-ui

Drive and inspect a requested UI using the active host’s browser or desktop controls, collecting evidence from the actual affected surface.

Skill by hapwi

IndexedMIT
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/hapwi-control-ui

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Drive and inspect a requested UI using the active host’s browser or desktop controls, collecting evidence from the actual affected surface.

Use cases: design

Reviews

Write a review

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/control-ui

What's inside

Skill name
control-ui
Description
Drive and inspect a requested UI using the active host’s browser or desktop controls, collecting evidence from the actual affected surface.
License
MIT
Instructions
11 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Control UI

Read [the runtime contract](../hap-mode/references/runtime.md). Use browser/desktop controls supplied by the active host. Inspect the affected surface before interacting; prefer observed accessible locators over coordinates. On T3 start with `preview_status`, then `preview_open` if needed, and use the collaborative preview tools. Respect human tab ownership and explicitly requested or repository-required browser workflows.

For a bug, reproduce the reported behavior, gather the smallest useful evidence, make authorized changes, and repeat the same interaction. For a visual task, inspect only the affected screen and capture its relevant state. A generated mockup, successful build, or another agent's report does not prove the real UI works.

Use native app/simulator controls only when that surface is available. Report missing access, authentication, or device controls and what was inspected. Never claim a browser check verifies a native app. Use available authorized controls as far as possible before asking for user-provided screenshots.

Do not submit messages, transactions, deployments, or other external writes as tests without authority for that action. Exclude credentials and private screenshots from public artifacts.

From hapwi/hapstack, MIT.

More from hapwi/hapstack

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Oct 14, 2026 (15 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
Repository
github.com/hapwi/hapstack
First seen
Oct 24, 2026
Last checked
Oct 24, 2026
Source
GitHubgithub-graphql

More from hapwi

Cite this page

control-ui skill. Agent Mart. https://agentmart.dev/skills/hapwi-control-ui (last updated 2026-10-24)