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

Annotations only say destructive/idempotent/non-read-only; the description goes far beyond them by spelling out overwrite semantics ("the old value is not kept"), the empty-string-clears-a-field rule with its companyName exception, atomic rejection on duplicate company name, no VAT ID format validation, and the surprising web-app side effect that saving an invoice rewrites the record's address/VAT/email. It also explicitly disclaims a side effect ("It does not email anyone"), which prevents a false assumption.

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?

The scope sentence is front-loaded and each sentence contributes a distinct, non-obvious fact. It runs long as a single unbroken prose block, though, and the web-app overwrite quirk is buried mid-paragraph rather than set apart, which slightly hurts scanability.

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 10-parameter destructive mutation with no output schema, the description covers the things an agent most needs: partial-update semantics, clear-field syntax, failure mode and atomicity, immutability of one field, and side effects on invoices. Nothing material for correct invocation is left unstated.

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%, so baseline is 3, but the description adds semantics the schema does not carry: empty string as a clear-field sentinel (with the companyName exception), company-name uniqueness enforced at save time, VAT ID stored verbatim without validation, and debtor number being immutable. That is genuine added meaning rather than restatement.

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?

"Changes one customer record, chosen by id" is a specific verb + resource + cardinality, and it is immediately distinguishable from create_customer, delete_customer, and get_customer. The follow-on sentence enumerates the exact fields in scope (company name, address, VAT ID, email), so an agent knows what this tool is for without opening the schema.

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?

"Use it to correct a company name, address, VAT ID or email" gives an explicit use case, and the constraints (debtor number cannot be changed, invoices are not altered) implicitly bound when not to reach for it. There is no explicit naming of an alternative sibling, but create_customer/delete_customer are self-evidently different, so the routing need is largely satisfied.

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