Skip to main content
Glama

Capture Native Ui Scenario

capture_native_ui_scenario
Destructive

Runs scripted press, scroll, text-entry, and bounded wait steps in a single native app window, returning ordered before/after accessibility and screenshot captures plus action failures.

Instructions

Run selected-element press, increment/decrement scroll, AXValue text entry and bounded wait steps in one exact native app window. Returns ordered before/after captures and action/capture failures. Actions can change app data, network activity and persistent state; application state is left as-is and restoration is not attempted.

Example tool call

This call uses example arguments supplied in the tool’s input schema.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "pid": 123,
      "steps": [
        {
          "kind": "wait",
          "milliseconds": 100
        }
      ],
      "window_id": 456
    },
    "name": "capture_native_ui_scenario"
  }
}

Input Schema

NameRequiredDescriptionDefault
pidYes
stepsYes
max_nodesNoMaximum accessibility nodes per capture; increase for large windows.
window_idYes
screenshotNo
accessibilityNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

B3.4/5.0
Behavior4/5

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

Annotations already declare destructive=true, openWorld=true and idempotent=false, but the description adds real substance beyond them: it warns that actions can mutate app data, network activity and persistent state, and explicitly states that state is left as-is and no restoration is attempted. It also discloses the return shape (ordered before/after captures plus action/capture failures). It stops short of detail on failure partial-commit behavior or deadline semantics.

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?

Three tightly packed sentences: capability first, return value second, risk profile last. Nothing is filler, though the opening sentence is dense with compound jargon and could have been split for faster scanning.

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

Completeness3/5

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

With no output schema, the description does cover the return contract and the side-effect/irreversibility story, which is the most important gap-filler for a destructive multi-step tool. However, for a tool requiring a pid, a window_id and a nested step array, the absence of any guidance on identifying the target window or constructing element paths leaves real operational gaps.

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

Parameters2/5

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

Schema description coverage is only 17%, so the description carries most of the parameter burden, and it mostly does not deliver. It effectively names the four step kinds, but it never explains the 'path' index arrays that identify selected elements, what pid versus window_id designate, or what the screenshot/accessibility toggles and max_nodes cap control. Half the required parameters remain undefined in prose and schema alike.

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

Purpose4/5

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

States a specific verb ("Run") and a concrete resource (interaction steps in one exact native app window), and enumerates the step types supported (press, scroll increment/decrement, AXValue text entry, bounded wait). The phrase "native app window" implicitly separates it from the browser/electron/process scenario siblings, but it never names them, so differentiation rests on inference.

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

Usage Guidelines3/5

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

Usage is implied by the step vocabulary and the 'one exact native app window' scoping, so an agent can infer this is the tool for scripted multi-step native UI interaction followed by captures. There is no explicit guidance on when to prefer it over observe_native_ui or trace_native_ui_action, and no preconditions (e.g. how pid/window_id must be obtained first) are stated.

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