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

Annotations already declare readOnlyHint and openWorldHint, and the description adds genuinely non-structured context: the series is downsampled so each point is the mean of consecutive readings, the hard window cap of ~92 days, and a prompt-injection warning that device/metric names and values are untrusted. It does not cover rate limits or behavior on unknown devices.

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, zero waste: capability first, the text-metric exception second, the safety caveat last. Every clause carries information an agent needs before calling.

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, the description fully compensates by describing the return shape (per-metric aggregates plus a downsampled series), the time-window semantics, and the fallback for non-numeric metrics. Nothing needed to invoke it correctly is missing.

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%, so the baseline is 3, but the description adds value beyond the schema: it explains that hours is capped at 2208 (~92 days) in human-readable terms and that max_points drives the mean-downsampling of the returned series.

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+resource ('Numeric history for a device') and enumerates exactly what is returned: min/max/mean/first/last per metric plus a downsampled series. It also distinguishes itself from the sibling get_latest by scoping it to numeric metrics only.

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?

Explicitly routes the agent elsewhere for the one case it cannot serve: 'Text metrics have no history; use get_latest for them.' That is a concrete when-not with a named alternative, though other siblings (list_recent_devices, publish_reading) and prerequisites such as device existence are not addressed.

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