Raw data points

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

Annotations only declare readOnlyHint and openWorldHint, so the description carries the behavioral load and does so well: result ordering (newest first), the pagination contract ('Pass nextPageToken back as pageToken' — an output field documented nowhere else), and an API quirk that filtering for electrocardiogram treats start as UTC and ignores end. Those are exactly the non-obvious traits an agent needs before calling.

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?

Four dense sentences, front-loaded with what the tool returns before caveats and pagination mechanics. Every sentence carries distinct information — examples, alternative routing, pagination, and the ECG exception — with no restatement of the name or title.

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?

For a 6-parameter read tool with no output schema, the description covers return ordering, result volume, pagination tokens, and a filtering edge case, which is most of what's needed. Minor gaps remain: no mention of how the advanced 'filter' parameter supersedes start/end, and no guidance on pageSize tradeoffs beyond what the schema states.

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 67%, and the description fills the most consequential gaps: it explains pageToken's real source (nextPageToken from a prior response, which the schema leaves undocumented) and overrides the schema's 'end defaults to start' with the ECG-specific rule that end is ignored. It does not mention the filter parameter's override of start/end, which remains schema-only.

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 ('Raw records for a data type, newest first') and enumerates the record families it returns (sleep sessions with stages, exercise sessions, weight entries, resting heart rate, HRV, SpO2, ECG). It also positions itself against siblings by naming the rollup tools as the alternative for high-frequency types, so an agent can distinguish it without opening the schema.

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?

Gives an explicit conditional exclusion: for heart-rate and steps, prefer the rollup tools because they produce many points per day. That is real routing guidance, not implied usage. It stops short of covering the other siblings (get_daily_summary, list_data_types) or when raw-vs-summary is otherwise preferred.

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