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

Annotations already cover the safety profile (readOnly, idempotent, non-destructive, not open world), so the bar is lower. The description still adds real behavior: no prior sample in the window yields gap: true with no delta, and nothing is interpolated. That edge-case disclosure is exactly the kind of trait annotations cannot express.

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?

Front-loaded with the return shape, then window modes, then the gap rule. Dense but every sentence earns its place, and no filler appears.

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?

With no output schema and only two parameters, the description carries the full burden and does so well: it enumerates the returned fields and covers the missing-prior-sample case. The only shortfall is the absent routing against sibling metric tools.

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 50%: metric_id is documented in the schema, but window's enum values carry no per-value description there. The description compensates by defining each window mode, including the default 'sweep' meaning. The metric-key-or-definition-id note merely restates the schema's own description.

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 specifies the resource precisely: one measured figure plus its stored prior sample, with the exact fields returned (current, prior, delta, delta_ratio, timestamps, definition id, unit, materiality). An agent knows this is a latest-value-plus-prior comparison, not a historical series. It stops short of naming the obvious sibling metric_history, so differentiation is inferred rather than stated.

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 window semantics ('sweep' = previous stored sample, the default; or 24h/7d/30d) imply when each mode applies, and the gap behavior explains an edge case. However there is no explicit when-to-use guidance versus metric_history or the other metric-oriented siblings, so the agent must infer the routing.

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