Skip to main content
Glama
Harusame64

desktop-touch-mcp

by Harusame64
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Local

browser_overview

List interactive elements on the current page with CSS selectors, visible text, and ARIA states to find stable selectors before clicking or verify UI state without screenshots.

Instructions

List all interactive elements (links, buttons, inputs, ARIA controls) on the current page with CSS selectors, visible text (an input's name — never what is typed in it), and viewport status — use before browser_click to discover stable selectors, and prefer this over screenshot when verifying button/toggle state after submission (no image tokens, structured output). scope limits to a CSS subsection (e.g. '.sidebar'). Returns state (checked/pressed/selected/expanded) for ARIA custom controls. Also returns a modal: section — whether a true modal dialog is blocking the page (isModal + blocker {name, role} + the signals it was judged on); it is ALWAYS present (isModal:false when no modal), and a navigation drawer is NOT reported as a modal (only an aria-modal / alertdialog / native showModal dialog, or a backdrop-backed dialog that locks the page, is treated as modal). Caveats: Selectors are CDP-generated snapshots — re-call after page navigates or re-renders. An input's text is its name — from aria-labelledby, aria-label, its , title, then its hint text — never what is typed in it: read values with browser_form, which withholds the value of a field the page masks (a password). Typed errors: code:'BrowserNotConnected' (CDP not attached — call browser_open or browser_open({launch:{}})). Note: a non-matching scope CSS selector silently falls back to the full document (does not raise an error) — verify the selector via browser_eval if scoped enumeration is required.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
portNoChrome/Edge CDP remote debugging port (default 9222; configurable via desktop-touch-config.json)
scopeNoCSS selector to limit the search scope (e.g. '.s-main-slot', '#nav-search-form'). Omit to scan the full page.
tabIdNoTab ID from browser_open. Omit to use the first page tab.
typesNoElement types to include. Default 'all' returns links, buttons, and inputs.
includeNoOptional response-shape opt-in. `['envelope']` returns the self-documenting envelope (`_version` / `data` / `as_of` / `confidence`). `['raw']` forces raw shape (overrides DESKTOP_TOUCH_ENVELOPE=1 server default). Default behaviour is raw shape (compat with existing clients).
maxResultsNoMaximum number of elements to return (default 50).
inViewportOnlyNoWhen true, only return elements currently visible in the viewport.
includeContextNoWhen true (default), append `activeTab` + `readyState` lines to the response. Set false to skip — saves ~150 tokens per call when chaining several browser_* calls in the same tab.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed4 schema fields changedv2.3.1
    • addedInput schema / $schema
      Added value: +"http://json-schema.org/draft-07/schema#"
    • removedInput schema / additionalProperties
      Removed value: -false
    • changedInput schema / properties / includeContext / description
      Previous value: -"When true, append activeTab and readyState context to the response."New value: +"When true (default), append `activeTab` + `readyState` lines to the response. Set false to skip — saves ~150 tokens per call when chaining several browser_* calls in the same tab."
    • changedInput schema / properties / port / description
      Previous value: -"Chrome/Edge CDP remote debugging port."New value: +"Chrome/Edge CDP remote debugging port (default 9222; configurable via desktop-touch-config.json)"
  2. Addedv1.12.0
  3. Removedv1.10.4
  4. Addedv1.9.2
  5. Removedv1.8.0
  6. Addedv1.6.0
  7. Removedv1.5.1
  8. Addedv1.5.0
  9. Removedv1.4.3
  10. Addedv1.4.2
  11. Removedv1.4.0
  12. Addedv1.2.1
  13. Removedv1.1.2
  14. Addedv1.1.3
  15. Removedv1.1.1
  16. Addedv1.0.3

TDQS

A4.8/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries the full burden and does so: it discloses that selectors are CDP snapshots needing re-call after navigation, that input text is the field name and never the typed value, that the modal section is always present and drawers are not modals, that a non-matching scope silently falls back to the full document, and the typed error BrowserNotConnected with its remedy.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Purpose is front-loaded and every sentence carries operational information (alternative routing, caveats, error handling). It is a dense single-paragraph wall of text that would scan better with structure, but there is little filler to remove.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With no output schema and 8 parameters, the description compensates by describing return content in detail: per-element CSS selectors, text, viewport status, ARIA state, and the always-present modal section with isModal, blocker, and the signals it was judged on. Nothing needed to call it correctly appears missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3, but the description adds real semantic value on top: it explains scope limits enumeration to a CSS subsection and warns that a non-matching scope selector silently falls back to the full document rather than erroring. It does not add much on port, tabId, maxResults, or include beyond the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('List all interactive elements... on the current page') and enumerates the returned fields (CSS selectors, visible text, viewport status, ARIA state). It directly distinguishes itself from siblings by naming browser_click, screenshot, browser_form, and browser_eval as different tools for different jobs.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicit when-to-use is given ('use before browser_click to discover stable selectors') plus a named alternative and the condition that selects it ('prefer this over screenshot when verifying button/toggle state after submission'). It also routes value-reads to browser_form and scope-verification to browser_eval.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

Other Tools