Run Task

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

With no annotations, the description carries the full burden and does a good job by warning that the self-reported result can be wrong and that the tool is disabled under safety policy. It does not mention potential autonomous side effects or timeouts, but the reliability caveat is significant and useful.

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

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

The description is two focused sentences with no redundant wording. The critical caveat is front-loaded as a parenthetical, and the safety-policy disablement is stated in one clear clause.

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

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

Even though an output schema exists, the definition lacks parameter semantics for four of five parameters and does not explain how steps, vision, or reasoning affect execution. An agent cannot confidently configure a correct invocation based on this description alone.

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

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

Schema description coverage is 0%, so the description needed to clarify the five parameters. It only implies that task is the goal; steps, device, vision, and reasoning are left entirely unexplained. This is minimal compensation for a schema that provides no parameter documentation.

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

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

The description states a specific action ('Hand a goal') and a specific resource ('the Mobilerun LLM agent'), clearly distinguishing this from direct-action siblings like tap, swipe, and launch_app. The added caveat that the result is best-effort further clarifies what the tool does and does not guarantee.

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

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

The description gives concrete guidance: after using run_task, verify the result with perceive_screen, and it explicitly notes that the tool is disabled when the safety policy is on. It does not explicitly contrast run_task with direct-action siblings for when to choose one over the other, but the context is clear.

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