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

Annotations already declare readOnly, idempotent, non-destructive, closed-world, so "never emails participants" largely restates that safety profile, though it is a concrete reassurance. The genuinely additive behavior is the at-least-once delivery implication of "save nextVersion only after processing" and the exclusive cursor boundary, but the description never states what the response returns or where nextVersion comes from, which matters because there is no output schema.

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 short sentences, front-loaded with the core fetch semantics, then the cursor-safety rule, then polling cadence; nothing is padding. The telegraphic style leaves the connection between "nextVersion" and the request parameters implicit, costing some clarity for no real size saving.

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?

Annotations and the input schema are lean, but this tool's contract is a cursor loop, and the description covers the two critical loop rules: the exclusive lower bound and not advancing the cursor before processing. The remaining gap is the return payload shape (change entries plus nextVersion), which is not in an output schema and must be inferred.

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 0%, so the description carries the burden and only partially meets it: "strictly after sinceVersion" clarifies the exclusive boundary (schema only shows an integer defaulting to 0), but eventId is left entirely unexplained despite its unusual 48-hex-character pattern. The default value of sinceVersion and the meaning of an absent/non-advancing cursor are also unaddressed.

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?

"Fetch changes strictly after sinceVersion" gives a specific verb (fetch) and a scoped resource (changes since a cursor), and the exclusivity qualifier is precise. It never names the resource domain (event changes) nor distinguishes itself from near-siblings like get_event, get_channel_update, or acknowledge_changes, so an agent must infer what kind of "changes" are returned.

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 polling instruction (every 300 seconds when email monitoring isn't available) and the cursor-advance rule ("Save nextVersion only after processing") together describe a concrete usage pattern an agent can follow. What's missing is routing guidance: no statement about when to use get_event or acknowledge_changes instead, and no explicit 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.