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

Annotations declare the safety profile (not readOnly, idempotent, non-destructive), so the description's job is to add what annotations can't: it discloses that tags are private to the account, that the contact is never notified, and that tags never leak into outbound content. That is exactly the kind of consequence an agent needs before calling a mutation. It stops short of 5 by not saying what happens on failure (e.g. tagging a non-accepted contact, exceeding the 12-tag cap).

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?

Three tight clauses: what it does, why it's safe/private, where to go next. Front-loaded with the action, no filler, and the closing pointer to list_contacts is one sentence rather than a paragraph.

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 3-parameter, non-destructive, idempotent mutation with annotations and full schema coverage, the description covers purpose, privacy consequence and a follow-up route. Missing only edge behavior — what happens with an unaccepted contact, whether add and remove can be combined in one call, and whether the response lists the resulting tag set.

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 already documents lowercasing, hyphenation and the 12-tag ceiling, so the schema carries the load. The description contributes only the framing of tags as private personal labels plus example values (client, family, board), which is useful color but no additional syntax or semantics. Baseline 3 applies.

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?

States a specific verb pair (put on / take off) against a specific resource (the user's own labels on an accepted contact) and names the namespace it operates in with examples (client, family, board). The mutual exclusivity with the thread-tagging sibling is implicit in the word 'contact', but the add-vs-remove duality is stated outright, so an agent knows the full scope from the description alone.

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?

Gives a concrete condition — 'accepted contact' — which tells the agent this cannot be used on pending or blocked contacts, and it hands off the filtering case to `list_contacts` with a tag argument. It doesn't state what to do when the contact isn't accepted yet (accept_contact is the obvious sibling but is never named), which keeps it short of a 5.

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