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

Annotations only declare the safety profile (readOnlyHint=false, destructiveHint=false, idempotentHint=false); the description adds substantial behavior beyond that: per-step locator resolution, RECOVERED signaling, accessible-name drift detection, state-change detection, and the unsupported fallback. It does not explicitly state that replaying mutates the target app's state, which is the main remaining gap.

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?

A single dense paragraph that is front-loaded with the core action before moving into reporting semantics and the web-only limitation. Length is justified by the amount of behavioral information, though the RECOVERED/unsupported clauses are somewhat compressed.

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?

With no output schema, the description carries the return-value burden and does so by enumerating what is reported per step. Combined with annotations covering the safety profile and full schema coverage of the 3 parameters, an agent has everything needed to call it correctly; only session prerequisites are unstated.

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 100%, so udid, recordingId and stopOnFailure are already documented in the schema (including the recordingId provenance and the stopOnFailure rationale). The description adds no parameter-level detail beyond what the schema provides, so the baseline 3 applies.

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?

States a specific verb+resource (replay a stored recording against a device) and describes the mechanism: each step's element is re-resolved from recorded locators, then the action re-performed. This clearly distinguishes it from siblings like ios_actions_record_start/stop, ios_actions_recording_read and flow_recording_replay.

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 web-steps-only restriction is an explicit boundary (native and coordinate steps return `unsupported`), which is genuine guidance. However, there is no statement of when to choose this over the flow-level replay siblings, what prerequisites exist (e.g. an active device session), or whether the recording must be stopped first.

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