Sample the output

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

No annotations exist, so the description carries the full load. It does disclose the return payload (min/max/mean of vout/iout/pout/vin/temp, modes seen, Ah/Wh, last sample) and that sampling spans a duration, which is genuinely useful. It says nothing about blocking behavior, whether the device must be outputting for the call to succeed, whether samples can be dropped, or what happens when `seconds`/`count` exceed limits.

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?

Two sentences, both earning their place: the first front-loads the mechanism and the full result shape, the second routes to the alternative. The statistics list is long but necessary because no output schema exists, so it substitutes for structured return documentation rather than adding noise.

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 no annotations, the description steps up by enumerating the aggregate statistics returned and noting the last sample is included, which is exactly what an agent needs to decide whether to parse or re-poll. The remaining gap is failure and truncation behavior for long or interrupted sampling runs.

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 75%, with `every`, `seconds`, and `max_wall` documented inline but `count` undocumented. The description compensates by explaining that `count` is an alternative to `seconds` for specifying sample quantity, a relationship the schema alone does not express. Only `max_wall`'s role as a hard call timeout is left to the schema.

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 names a specific action and scope: poll the status at interval `every` for `seconds` (or `count` samples) and return aggregate statistics plus the last sample. That is concrete enough to separate it from log_start, which it explicitly names. It does not, however, distinguish itself from the sibling `status` tool, leaving one ambiguity in the tool set.

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 gives an explicit alternative and the condition that selects it: 'For longer runs or a file use log_start.' The implication is that this tool is for short, bounded sampling windows, which is reasonably clear. It offers no exclusion against the sibling `status` tool, so an agent has to infer which of the two to reach for on a single check.

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