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

The description adds behavioral details beyond annotations: it is unmetered in v1, supports cursor pagination (pass next_cursor as since), and ensures no 'still fine' events. However, it does not describe the response structure or potential error cases, though the readOnly and idempotent hints are consistent.

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 description is a single, information-dense paragraph that covers purpose, event types, pagination, filtering, and usage notes. It is concise without unnecessary fluff, though the phrase 'Not a malware scan' is an odd addition that adds little value.

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?

Given the five parameters and absence of an output schema, the description provides sufficient context for invocation: it explains pagination, filtering, and polling. It lacks an explicit description of the response format, but the nature of a change feed makes it reasonably inferable.

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?

All parameters have descriptive schema entries: limit has min/max/default, since is clearly explained, events/server_names/namespace_domain list their purposes. The tool description supplements the events parameter by listing the allowed event names, enhancing clarity.

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 clearly states the tool's purpose as a change feed that reports transitions across servers, enumerating specific event types like server_new and health_change. It distinguishes itself from sibling tools by focusing on changes and includes the 'CENSUS-CHANGES/1' identifier.

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 provides specific usage guidance: cursor-paginated with next_cursor, filtering options via events, server_names, and namespace_domain, and a polling recommendation of hourly. It also clarifies that only real transitions are returned, but does not explicitly compare with siblings like census_recent.

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