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

Annotations carry the basic safety profile (readOnly=false, idempotent=false, destructive=false), and the description adds genuinely valuable behavior on top: the partial-success contract ('valid topics are written and returned in items, unapplicable ones collected in errors') and the non-destructive scoping to only topics present in the body. It does not contradict annotations; 'create or update' aligns with readOnlyHint=false.

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 sentences, roughly 40 words, each earning its place: core purpose, scoping behavior, and partial-success behavior. The most decision-relevant fact (additive, not replace) is front-loaded in the first clause.

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 bulk mutation with partial-success semantics, the description covers the essential facts: additive create-or-update behavior, touched-topic scoping, and the items/errors result split that compensates for the absent output schema. Minor gaps remain — it does not route explicitly to bulk_replace_user_preferences for replacement needs, explain tenant_id's effect, or detail the exact shape of items/errors entries — but none block correct invocation.

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%, so the schema already documents user_id, topics, and tenant_id with individual descriptions; baseline 3 applies. The description reinforces that topics entries are processed individually via the partial-success split but adds no syntax or format details beyond what the schema provides.

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?

Begins with a specific verb+resource ('Additively create or update a user's preferences') and bounds the scope to 'one or more topics in a single request.' The phrase 'Only the topics in the body are touched; existing overrides for other topics are left untouched' clearly differentiates it from the sibling bulk_replace_user_preferences without needing to open either 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?

The 'Additively' qualifier and 'existing overrides for other topics are left untouched' effectively convey when this tool is appropriate: partial, additive updates rather than wholesale replacement. However, it stops short of explicitly naming bulk_replace_user_preferences as the alternative for full-replacement scenarios or stating a when-not-to-use condition.

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