load_dataset

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

Goes well beyond the readOnlyHint annotation: it discloses automatic selection of PeakProperties.arf when DriftSpots.arf coexists, automatic selection of the newest batch via the AlignmentResult_<timestamp> filename convention (old batches skipped), that mixed-date folders will not halt analysis, and that the selected directory becomes the default for later tools. These are exactly the non-obvious side effects an agent needs.

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 core action, then a numbered list of the two analysis steps, then edge-case behavior and parameter details. It is longer than most definitions but nearly every sentence carries operational information; only the closing restatement of the parameter is mildly redundant with the header text.

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 covers what is produced at a high level (arf2 summary, PCA result) and promises the auto-selection results are echoed in the output. That is sufficient for correct invocation, though the concrete shape of the returned summary/PCA payload is not described.

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?

The single parameter has 0% schema description coverage, but the description fully compensates: it defines the parameter as the MS-DIAL output folder path, gives the fallback chain (LIPIDMIX_DATA_DIR or <project>/data when omitted), and notes the side effect of updating subsequent tools' default search location. Nothing about the parameter is left ambiguous.

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 and resource ('load_dataset' = load an MS-DIAL output folder) and enumerates the two concrete steps it performs (arf2 overview summary, then PCA on the arf with per-sample intensities). It explicitly positions itself as the entry point relative to sibling tools like arf_list_classes / arf_parser, so an agent can distinguish it from dataset_load or arf_parser without opening any 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?

Explicitly frames itself as the entry point for MS-DIAL output folders and states that '以降の arf_list_classes / arf_parser 等はこのセッション状態をそのまま利用できます', which gives clear sequencing guidance. It does not contrast against the similarly named dataset_load or dataset_status siblings, so there is no explicit when-not guidance.

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