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

Annotations already declare destructiveHint=true, so the description is not needed to establish danger, yet it still adds substantive context: the exact fields erased, the fact that a partial record (occurrence and cost) persists, and that the deletion is permanent ("for good"). That retained-data nuance is invisible in annotations and is precisely what an agent needs before confirming an irreversible action.

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?

Three short sentences with zero filler: what is deleted, the consent gate, and what remains. The destructive scope is front-loaded, and the trailing clause about retention prevents the most likely wrong assumption without a wasted word.

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?

A single-parameter, non-read-only tool with no output schema needs only to convey effect, permanence, and consent condition — all three are present and no return-value explanation is required. Nothing an agent must know to call this safely 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 100% and the id property already documents "Call id or SID from my_calls." The description adds no further syntax or format detail, but the core meaning of the parameter is fully carried by structured data, so the baseline-level credit 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?

Uses a specific verb ("Delete") on a specific resource (a call's transcript, summary, recording, relayed messages) and explicitly bounds the scope by naming what survives (the fact the call happened and its cost). No sibling tool performs deletion, so the agent can identify this unambiguously by its scope statement alone.

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?

"Only when your person asks" gives an explicit consent precondition that gates invocation, which is the key routing signal for a destructive tool. It does not name a read-only alternative (e.g. transcript or my_calls) for the case where the user wants to inspect rather than remove, leaving that inference to the agent.

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