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

Annotations already declare readOnlyHint/destructiveHint, and the description adds real behavior beyond them: default ordering, the asynchronous transcription state and retry advice, limit bounds, and a prompt-injection warning about caller-ID names and transcripts. That last point is materially useful safety context for an agent.

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 ordering, then semantics and caveats in a tight block. Dense but nearly every sentence carries information; the trailing safety sentence is slightly tangential but earns its place given untrusted content.

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?

An output schema exists, so return values needn't be described, and the description still covers ordering, transcript state, filtering semantics, and safety. Nothing needed to call it correctly is missing.

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 coverage is 0%, so the description must carry all parameter meaning, and it does: `since` accepts ISO or unix seconds, `handle` is a number/address/contact name, and `limit` is 1–200 with default 25. All three parameters are explained beyond their bare types.

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 and resource ('Read voicemails and Live Voicemail recordings on this Mac'), plus scope details like newest-first ordering and per-item transcript/summary. It is clearly distinct from siblings such as recent_calls, which it explicitly cross-references.

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 practical context: transcript_pending=true means retry later, and call_id links to recent_calls for call metadata. It doesn't explicitly state when to prefer recent_calls or other siblings instead, so it falls short of a full when/when-not rubric.

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