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

The annotations already declare destructiveHint=true, so the description's 'permanently delete' reinforces that. But the description adds critical behavioral context: the deletion is permanent (irreversible), and explicit user confirmation is required. This is exactly the kind of high-stakes behavior an agent needs to know before invoking. It also implies real-world consequences (data loss) that annotations alone don't convey. No contradiction with annotations; the description aligns with destructiveHint=true.

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 sentences with no filler. The most important facts — permanent deletion and user confirmation — are front-loaded. Could potentially add more behavioral context, but for what it contains, it's efficient and well-structured.

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?

The description is complete enough for a destructive operation: it states permanent deletion and the confirmation requirement. The output schema is absent, but for a delete operation the return value is usually a success indicator, which the agent can infer. The sibling list includes void_* tools (void_bill_payment, void_invoice, void_sales_receipt) which are alternative non-permanent operations, and the description doesn't explicitly guide the agent to them; that's a minor gap but not critical for a permanent delete tool.

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% — the schema already thoroughly documents 'id' as the QuickBooks Id, 'company' as which connected company, and 'confirm' as must be true. The description doesn't add much parameter-level meaning beyond the schema. The 'require explicit user confirmation' in the description adds slight emphasis to the 'confirm' parameter's importance, but the schema already covers it. Baseline 3 is appropriate when schema does the heavy lifting.

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 states a specific verb ('permanently delete'), a specific resource ('an estimate from QuickBooks Online'), and includes the key requirement of explicit user confirmation. The title is just 'delete estimate', but the description adds the permanent, irreversible nature and the QuickBooks context, making the tool's purpose unambiguous. The sibling list includes many delete_* tools, but the description names the exact resource (estimate), which distinguishes it from delete_bill, delete_invoice, etc.

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 says the delete is permanent and requires explicit user confirmation, which gives clear guidance on when NOT to call it (without user confirmation) and implies it should only be used after user consent. It doesn't name an alternative (e.g., void vs delete), but the context signals are clear enough — this is a permanent delete, versus void_* siblings which might be the non-permanent alternative. The 'requires explicit user confirmation' is a strong usage guideline even though it doesn't enumerate alternatives.

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