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

The description adds substantial behavior beyond the annotations (destructiveHint=true, idempotentHint=true): server-side ownership enforcement, cascade deletion of dependent AI assessments, the precise conditional rule for subject deletion (same creator, no remaining reviews, no subject relationships), and confirmation protocol. It also reinforces the idempotentHint with the 'do not ask for a second confirmation' instruction. All behavior is consistent with annotations — a destructive, idempotent mutation.

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?

Four sentences, each earning its place: the core action, ownership guarantee, cascade/conditional side effect, and an agent-facing behavioral rule. The action is front-loaded in the first sentence. It is slightly dense — the subject-deletion clause packs three conditions into one long sentence — and the confirmation theme appears twice, creating minor redundancy.

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 destructive tool with no output schema and zero param coverage, the description covers the critical invocation semantics: what is deleted, what cascades, what is preserved, who may delete, and when confirmation is needed. Gaps remain — the return value/error behavior for a non-existent review and the explicit semantics of idempotency_key — but these are secondary to the full disclosure of the operation's effects, which is excellent.

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?

Given 0% schema description coverage, the description carries the full burden and largely delivers: 'one review' maps to id, 'explicitly requests deletion' maps to confirm_deletion, and the entire subject-preservation condition explains the behavior of delete_orphan_subject. The only parameter left implicit is idempotency_key, whose purpose is only hinted at by the idempotentHint annotation and the no-double-confirmation note. Strong compensation, but not systematic per-parameter documentation.

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?

The description opens with a specific verb and resource — "Permanently delete one review" — and the title reinforces the scope ("Delete a user-owned review"). Ownership, single-review scope, and permanent nature are all explicit, leaving no ambiguity about what this tool does. The only sibling that could compete is save_experience, and the destructive framing clearly distinguishes this from any create/update operation.

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 description states a clear precondition: use only "after the authenticated user explicitly requests deletion," and a clear when-not instruction: "Do not ask for a second confirmation when the current user request already explicitly authorises deletion." No alternative tools are named, but no sibling performs deletion, so exclusions are unnecessary. Clear context with no explicit alternatives.

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