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

Annotations already cover readOnly/idempotent/non-destructive, but the description adds substantial behavior beyond them: the 50-result cap, case-insensitive matching on any word or its singular including group names, and access scoping (results are read with the user's access to the connection, and on Tableau with their own Tableau user). Those are permission and result-set traits an agent could not infer from the annotations.

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?

Content is front-loaded with the purpose, then behavior, then the follow-up workflow, and every clause carries information. However, the opening sentence is a ~70-word run-on with two colons and nested qualifiers, which makes it harder to parse than it needs to be for a two-parameter lookup tool.

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 still conveys the return shape: grouped results per connection type, the 50-item cap, and the presence of a modelId field used downstream. It does not enumerate the returned fields per connection type, but it covers enough for an agent to call the tool and chain it into addReportDatasheets correctly.

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 schema already documents query and the params.type/params.id fields, so the description's added value is in the matching semantics: it clarifies that query words match names or groups, tolerate singular forms and any casing, and that an omitted query returns unfiltered results. It does not address the nested query-object wrapper, so a small gap remains.

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: returns the models of a Looker/Power BI/Tableau connection matching query words, capped at 50. It goes further and defines what a model is per connection type (Looker explores by LookML model, Power BI semantic models by workspace, Tableau published data sources by project) and contrasts it with a dashboard tab, so an agent can tell it apart from searchDashboards/searchTables without opening a 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 a clear workflow: confirm the model with the user, add it to a report with addReportDatasheets by modelId, pass the new datasheet id to sendReportMessage, and use listConnections to obtain the connection type and id. It lacks an explicit when-not statement or a named alternative for looking up non-model resources (e.g. dashboards/tables), so the routing to siblings is implied rather than stated.

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