// Use the built-in background browser for interactive web tasks, including opening pages, inspecting the DOM, filling forms, clicking through multi-step flows, and verifying page state. Use this when real browser interaction is required instead of only fetching static text.
Develop and maintain frontend apps in `<AgentWorkspaceRoot>/app` such as tools, mini-games, landing pages, and interactive experiences. Use when the user asks to create, iterate on, or fix an app or page.
Build high-quality HTML presentations with the App module in `<AgentWorkspaceRoot>/app`, suitable for pitches, reports, and presentations. Help users converge on a style quickly by showing visual directions first and iterating on content afterward.
Manage Kian's installed Skills, including installing a skill from a GitHub repository, listing installed skills, enabling or disabling visibility, and uninstalling non-builtin skills. Use this when the task is to change which Skills an agent can use.
Manage Kian's own configuration and runtime behavior, including app preferences, workspace settings, models and providers, shortcuts, chat channels, and MCP servers. Use this when changing Kian itself rather than project files.
Manage global scheduled tasks and execution records, including creating, updating, deleting, viewing, and troubleshooting schedules. Use when the user mentions cron, scheduled jobs, or timed execution. Only 5-field minute-level cron expressions are supported.
Manage user documents and knowledge in `<AgentWorkspaceRoot>/docs`, including recording, organizing, searching, and maintaining notes, journals, and reference material over time. Use when the task involves textual knowledge.
| name | browser |
| description | Use the built-in background browser for interactive web tasks, including opening pages, inspecting the DOM, filling forms, clicking through multi-step flows, and verifying page state. Use this when real browser interaction is required instead of only fetching static text. |
Use this skill for real-time browser interaction in a hidden background session.
Use only these tools:
| Tool | Purpose |
|---|---|
BrowserOpen(url) | Open/reuse hidden background browser session and navigate to URL |
BrowserClick(selector) | Click element by CSS selector |
BrowserType(selector, text, pressEnter?) | Type into input; optional Enter submit |
BrowserScreenshot() | Capture current page screenshot |
BrowserRead() | Extract current page as markdown-like readable text |
BrowserReadDom(limit?) | List interactive elements with selector hints |
BrowserBack() | Navigate backward |
BrowserForward() | Navigate forward |
BrowserReload() | Reload current page |
BrowserEval(code) | Execute JavaScript in page context |
BrowserClose() | Close background browser session |
Do not reference WebFetch or WebSearch in this skill. They are not part of this browser toolset.
BrowserOpen.BrowserScreenshot.BrowserReadDom (set limit when pages are large).BrowserClick / BrowserType.BrowserScreenshot after each key action/navigation.BrowserRead when textual content is needed.BrowserBack / BrowserForward / BrowserReload for navigation control.BrowserClose when task is complete.BrowserReadDom first to get selector candidates.BrowserScreenshot.id, name, type, data-*, or constrained attribute selectors.BrowserEval for custom logic.BrowserEval, wrap with JSON.stringify(...) for stable output.1. BrowserOpen(url: "https://example.com")
2. BrowserScreenshot()
3. BrowserType(selector: "input[name='q']", text: "search query", pressEnter: true)
4. BrowserScreenshot()
5. BrowserRead()
6. BrowserClose()
1. BrowserOpen(url: "https://example.com/register")
2. BrowserScreenshot()
3. BrowserType(selector: "#name", text: "John Doe")
4. BrowserType(selector: "#email", text: "john@example.com")
5. BrowserClick(selector: "button[type='submit']")
6. BrowserScreenshot()
7. BrowserClose()
BrowserEval(code: "window.scrollTo(0, document.body.scrollHeight)")
BrowserEval(code: "document.querySelector('select#country').value = 'US'; document.querySelector('select#country').dispatchEvent(new Event('change', { bubbles: true }))")
BrowserEval(code: "JSON.stringify(Array.from(document.querySelectorAll('a')).map(a => ({ text: a.textContent.trim(), href: a.href })))")