Skip to content
Agentmart
Submit a tool

browser-test

Applies only to work in the quantum-box/tachyon-apps repository.

Skill by quantum-box

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/quantum-box-browser-test

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

Applies only to work in the quantum-box/tachyon-apps repository. Verify browser UI behavior with agent-browser when the user asks for 動作確認/verification or a requested UI change needs user-flow evidence. Do not trigger for every UI edit or automatically after implementation.

Use cases: design, testing, writing, agent-building

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/browser-test

What's inside

Skill name
browser-test
Description
Applies only to work in the quantum-box/tachyon-apps repository. Verify browser UI behavior with agent-browser when the user asks for 動作確認/verification or a requested UI change needs user-flow evidence. Do not trigger for every UI edit or automatically after implementation.
License
MIT
Instructions
105 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Browser Test

Verify UI behavior using `agent-browser`.

This skill is now a thin compatibility wrapper. For browser verification, use the `agent-browser` workflow and commands directly rather than Playwright MCP tools.

## App Configuration

### Tachyon (Main App)
```yaml
url: http://localhost:16000
tenant_path: /v1beta/<tenant-id>
full_url: http://localhost:16000/v1beta/<tenant-id>
credentials:
  id: test
  password: "${TACHYON_TEST_PASSWORD}"
  # Load the value from ignored local configuration; never store it in this file.
backend_api: http://localhost:50054
graphql: http://localhost:50054/v1/graphql
```

### Library
```yaml
url: http://localhost:5010
credentials:
  # GitHub OAuth login
```

### TACHYON Field UI
```yaml
url: http://localhost:3000
```

### TACHYON Field Admin UI
```yaml
url: http://localhost:3001
```

### AI Chat
```yaml
url: http://localhost:16000
api_base: http://localhost:16000
```

## Common Headers (API Testing)
```yaml
Authorization: Bearer dummy-token
x-operator-id: <tenant-id>
# x-user-id: optional (defaults to seed user)
```

## agent-browser Commands

### Navigation
- `agent-browser open <url>` - Go to URL
- `agent-browser get url` - Check current URL
- `agent-browser wait --load networkidle` - Wait for page settle

### Interaction
- `agent-browser click @e1` - Click element
- `agent-browser fill @e2 "text"` - Fill input
- `agent-browser select @e3 "option"` - Select dropdown option
- `agent-browser press Enter` - Keyboard input

### Inspection
- `agent-browser snapshot -i` - Interactive elements with refs
- `agent-browser screenshot` - Visual capture
- `agent-browser network requests` - View requests if needed

### Other
- `agent-browser wait --text "..."` - Wait for text
- `agent-browser wait 2000` - Wait fixed time
- `agent-browser get text @e1` - Inspect text

## Workflow

1. **Navigate**: `agent-browser open <url>`
2. **Snapshot**: `agent-browser snapshot -i` to get refs
3. **Interact**: Use refs from snapshot for clicks/typing
4. **Verify**: Re-run snapshot, check URL/text, or take screenshot
5. **Save**: Use `--screenshot-dir` to save directly into taskdoc `screenshots/`

## Screenshot Management

Save screenshots directly into the taskdoc directory when needed:
```bash
agent-browser screenshot --screenshot-dir docs/src/tasks/in-progress/<task>/screenshots
```

## Login Flow (Tachyon)

1. Open `http://localhost:16000`
2. Run `agent-browser snapshot -i`
3. Fill id=`test`, password=`hmw2atd@HCF3qwu*rcn`
4. Submit and wait for redirect
5. Verify the dashboard with snapshot, URL, or screenshot

## Notes

- Dev servers should already be running (`mise run dev`)
- No need to kill servers after testing
- Re-snapshot after every navigation or major DOM change
- If a command sequence does not depend on intermediate output, chaining with `&&` is fine

From quantum-box/tachyon-skills, MIT.

More from quantum-box/tachyon-skills

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Sep 29, 2026 (1 month ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from quantum-box

Cite this page

browser-test skill. Agent Mart. https://agentmart.dev/skills/quantum-box-browser-test (last updated 2026-10-24)