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

With no annotations, the description carries the full behavioral burden. It usefully discloses the consequence ('nothing was ever delivered, so nothing is lost but the text itself'), which tells the agent this is a low-stakes destructive action, but it never states whether the discard is permanent/irreversible or what happens if the draft was already sent.

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?

Two short sentences with the action front-loaded and no repetition of the schema. The second sentence is mild reassurance rather than specification, but it does earn its place by conveying non-destructiveness.

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?

For a single-parameter delete tool with no annotations and no output schema, the description covers what is removed and how consequential it is, which is the core need. It remains silent on irreversibility and failure modes (e.g., deleting a draft that has already been sent), which is the gap keeping it at minimum-viable.

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 schema description already explains that draft_id comes from list_compose_drafts, so the baseline of 3 applies. The prose adds no additional meaning about the parameter beyond what the schema documents.

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?

A concrete verb ('throws away') applied to a specific resource ('a message that was started and will not be sent'), which reads unambiguously as discarding an unsent compose draft. It does not explicitly name or contrast with siblings such as save_compose_draft, list_compose_drafts, or cancel_scheduled_send, so differentiation comes from the tool name rather than the text.

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 phrase 'will not be sent' implies the triggering condition (an abandoned draft), but there is no explicit when-to-use rule and no routing to alternatives — an agent must infer that drafts to be kept should go to save_compose_draft and that draft ids come from list_compose_drafts.

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