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

Annotations only cover the safety profile (readOnly=false, destructible=false, openWorld=false, idempotent=false); the description adds real operational context the annotations can't: 10-30s launch latency, the best-effort 'Device: ...' reply line, and the fact that unavailable fields are silently omitted rather than erroring. It doesn't say whether the session consumes a device slot or must be released, which keeps it from a 5.

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?

Front-loaded with purpose and prerequisite, then latency and return-value guidance. The Device-line sentence is somewhat dense with repeated clauses ('best-effort', 'read it before making separate calls', 'omitted entirely'), but every sentence carries usable information.

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?

No output schema exists, so the description appropriately spends effort describing what comes back (the Device line and its fields, plus the possibility that it is absent). It still omits session lifecycle details such as needing ios_end_session and whether the returned session handle is required by subsequent calls, leaving a modest gap for a state-creating tool.

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?

Single parameter (udid) at 100% schema description coverage, so the schema already fully documents it. The description never references the parameter, adding nothing beyond structured data; this is the baseline 3 case for high-coverage schemas.

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?

Specific verb+resource ('Start an iOS automation session on an iOS device') with an explicit prerequisite ('must be called before any iOS control commands') that cleanly separates it from the ~200 sibling iOS control tools and from ios_end_session.

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?

Gives a clear triggering condition (call first, before any iOS control command) and steers the agent to read the returned Device line instead of issuing extra calls for the same data. It does not warn about calling it repeatedly (annotations mark idempotentHint=false), nor mention the ios_end_session counterpart, so it falls short of full when/when-not coverage.

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