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

Well beyond the annotations, it discloses the identity-resolution order (email → LinkedIn → phone), phone normalisation rules (06 and +31 treated as equal), duplicate avoidance, and the fact that `bedrijf` is itself upserted. However, for a tool flagged destructiveHint=true it never says what happens to existing fields on an update — whether omitted values are left intact or cleared — which is exactly the behaviour an agent needs before invoking an upsert.

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?

Five dense sentences, front-loaded with the operation and match order before the fallback rules. Every clause carries information an agent needs; the parentheticals ('upsert, geen dubbelen', the 06/+31 note) are short and load-bearing.

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 20-parameter nested mutation tool with no output schema, the description covers the hardest part — how a record is identified and when a call is valid. The remaining gap is update semantics for existing fields, which matters given destructiveHint=true and is not recoverable from the schema.

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 40% across 20 parameters, so compensation is needed; the description does add real meaning for the matching keys (email, linkedin_url, telefoon/mobiel) and clarifies that `bedrijf` accepts a name or a domain, plus an implicit 'at least one key required' rule the schema does not encode (required is empty). But the remaining fields — land, adres, aanhef, lifecycle, lead_status, eigenschappen, agent — go undiscussed, and there is no statement about partial vs full update semantics.

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 in one phrase: 'Maak of werk een contactpersoon bij (upsert, geen dubbelen)'. It immediately differentiates this write/upsert tool from the read-oriented siblings (crm_contact, crm_contacten) and from crm_bewaar_bedrijf by making clear the subject is a person record.

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?

It gives concrete when-to-use guidance: a phone number alone suffices 'bv. voor een beller', and it states the minimum key requirement (e-mail/LinkedIn/telefoon, otherwise voornaam+achternaam+bedrijf). What is missing is an explicit statement of when to prefer crm_bewaar_bedrijf or whether creating a contact with an embedded `bedrijf` makes a separate company call redundant.

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