Skip to main content
Glama

Run phone agent task

run-phone-agent-tool
Destructive

Ask the on-phone GUI agent to carry out a narrow task on a farm phone. Keep the task specific ("open Settings and turn on Wi-Fi"). Runs asynchronously: the call returns a run id at once; poll get-phone-run-tool with it until the run succeeds or fails, and use phone-snapshot-tool to see the screen.

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": {
      "slot": "<slot>",
      "task": "<task>"
    },
    "name": "run-phone-agent-tool"
  }
}

Input Schema

NameRequiredDescriptionDefault
slotYesPhone slot to run the agent on, given as the `slot` UUID from list-phones-tool (its display name like "slot12" is not accepted). The phone must show `video_live: true` there.
taskYesPlain-language task for the on-phone GUI agent to carry out.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / slot / description
      Previous value: -"Phone slot to run the agent on, given as the `slot` UUID from list-phones (its display name like \"slot12\" is not accepted). Must be cast/live."New value: +"Phone slot to run the agent on, given as the `slot` UUID from list-phones-tool (its display name like \"slot12\" is not accepted). The phone must show `video_live: true` there."
  2. First observed

TDQS

A4.1/5.0
Behavior4/5

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

Adds substantial behavior beyond the annotations: the call is asynchronous, returns a run id immediately, and must be polled until success or failure. That is exactly the kind of operational context annotations (readOnlyHint=false, destructiveHint=true, openWorldHint=true) cannot express. It stops short of saying what the agent may irreversibly change on the device or whether concurrent runs on one slot are allowed.

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

Conciseness5/5

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

Three tight sentences: purpose first, example second, async lifecycle third. Every clause carries actionable information with no repetition of the schema or annotations.

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

Completeness4/5

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

With no output schema, the description correctly explains the return (a run id) and the follow-up tools needed to complete the workflow, which is the main gap a caller would otherwise hit. Minor omissions remain around failure handling and whether a run can be cancelled or retried on the same slot.

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 slot and task fields are already documented, including the video_live requirement and the UUID-vs-display-name rule. The description still adds meaning for `task` by prescribing the correct granularity and giving a sample phrasing, which the schema's one-line gloss does not convey.

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 and resource: ask an on-phone GUI agent to carry out a narrow task on a farm phone, with a concrete example ("open Settings and turn on Wi-Fi"). It is clearly distinguishable from snapshot/polling siblings, but it does not distinguish itself from other execution siblings such as phone-command-tool or run-macro-tool, leaving the agent to infer which execution path applies.

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 clear usage context: keep the task narrow and specific, and follow the async pattern by polling get-phone-run-tool and using phone-snapshot-tool to observe the screen. It lacks explicit exclusions (for example, when to prefer phone-command-tool or a macro instead), so the agent must infer the boundaries.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources