Skip to main content
Glama

owa_click

Presses an app control by its on-screen label, trying candidate labels in order and waiting for panels to render. Use "Panel > Control" to target the right control and read the result to confirm what the click did.

Instructions

Click a control in the app, found by the exact words written on it -- pass a list of candidate labels and the first one on screen wins. Waits a moment for a panel that is still rendering, and when nothing matches it answers with the closest labels it did find, so retry with one of those instead of giving up. Write "Panel > Control" to mean the one inside that panel ("Background > Videos"), which is how a word several panels share picks the right one. The answer says what was pressed AND what the press did: didChange, isOnNow for a control with an on/off state, opened / closed (menu, dialog, panel or window) for what it brought up or put away, and unverified when nothing changed. Pressing something is not the thing happening -- on unverified, check (owa_list_screens, owa_app_state, owa_find_ui) before telling the user it worked. Anything that changes what the congregation sees -- presenting, clearing, hiding a screen -- must be offered to the user first, never done unasked. button: "right" opens the control's own menu (a list row's Rename, Move to Trash…); clicks: 2 double-clicks (opens a document, plays a clip).

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "find": "<find>"
    },
    "name": "owa_click"
  }
}

Input Schema

NameRequiredDescriptionDefault
findYesThe label written on the control, or a list of candidate labels to try in order.
pageNoPart of the window URL, e.g. "reader.html"; default the main window.
buttonNo
clicksNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv2026.11.15

TDQS

A4.6/5.0
Behavior5/5

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

With no annotations supplied, the description carries the full burden and does so: it discloses that the tool waits for a still-rendering panel, enumerates result signals (`didChange`, `isOnNow`, `opened`/`closed`, `unverified`), and warns that pressing is not the same as the effect happening. It even spells out right-click and double-click behaviors and the human-confirmation requirement for visible changes.

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?

The opening clause front-loads the action and matching rule, and the following sentences each add a distinct behavior (waiting, miss reporting, scoped labels, result fields, confirmation policy, click variants). It is dense and readable, though the single long paragraph with many em-dashes could be split for faster scanning.

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?

There is no output schema and no annotations, so the description must explain returns and side effects — and it does, naming the exact result fields and interpreting them. For a UI-driving tool with mutation potential, that covers everything an agent needs to call it and interpret the response correctly.

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 only 50%, yet the description adds real meaning beyond the schema for three of four parameters: `find` accepts a string or ordered candidate list with "Panel > Control" scoping syntax, `button: "right"` opens the control's own context menu, and `clicks: 2` double-clicks to open a document or play a clip. Only `page` is left unexplained, which is a minor residual gap.

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 ("Click a control in the app") and immediately pins down the targeting model: matching by the exact visible words, with a candidate list where the first on-screen match wins. That is precise enough to separate it from sibling tools like owa_find_ui or owa_list_ui, which locate rather than press.

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

Usage Guidelines4/5

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

Gives concrete operating guidance: retry with the closest labels it returns rather than giving up, use "Panel > Control" when a label is shared across panels, and verify with owa_list_screens / owa_app_state / owa_find_ui when the result is `unverified`. It also sets a policy precondition — anything that changes what the congregation sees must be offered to the user first. It does not explicitly say when to prefer a different clicking/locating sibling, which keeps it below a 5.

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