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

Annotations already declare readOnlyHint=true and destructiveHint=false, covering safety. The description adds meaningful behavior beyond that: the short-page pagination semantics for full retrieval, the ≤1000 limit, and automatic tenancy enforcement — the latter being a genuine behavioral guarantee an agent needs to know before invoking. No contradiction with 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?

One information-dense sentence that front-loads the core purpose before capabilities, then pagination behavior, then tenancy. Every clause earns its place: no filler, no repetition of schema field names, and the most important scoping fact (generic over any dataset) comes first.

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 10-parameter tool with low schema coverage and no output schema, the description covers the critical invocation decisions: what to query, how to filter, how to paginate, and that tenancy is automatic. The main gaps are the filter object structure, response format (since no output schema exists), and the boundary against ppc_query for PPC-specific needs — but an agent could successfully call this tool from the description alone.

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 only 10% (only seller_connection_id has a description), so the description must compensate — and it largely does. It maps capabilities to parameters: projection (columns), filters with explicit operator list, date bounds (date_from/date_to), ordering (order_by/order_dir), and the ≤1000 bound on limit. The residual gap is the internal shape of the filters objects and exact date formats, which remain undocumented.

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 (query) and resource (any warehouse dataset), and enumerates concrete capabilities: column projection, filter operators (eq/neq/gt/gte/lt/lte/in/contains), date bounds, ordering, and pagination. The word 'generic' distinguishes it from specialized query siblings like ppc_query, and the cross-reference to data_describe orients the agent on dataset discovery.

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 explicit operational guidance: page with offset until a short page returns to retrieve FULL datasets, and mentions the ≤1000 limit. The 'see data_describe' hint routes the agent to the sibling for dataset discovery. However, it does not explicitly say when to prefer specialized siblings (e.g., ppc_query or pnl_* reports) over this generic tool, leaving the exclusion logic 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.