List a resource collection

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

Annotations carry only readOnlyHint and openWorldHint; the description carries the real burden and nails it, disclosing non-obvious runtime behavior the annotations can't: nullable `total` when first pages fetch server-side, the walk-until-truncated=false protocol, byte-capping at PORTAL_LIST_BYTE_CAP with silent auto-projection to { id, name } plus a `note` field. These are exactly the pagination and envelope edge cases that would cause silent agent bugs, and they go well beyond what readOnlyHint/openWorldHint convey.

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?

Dense but purpose-built: a single paragraph that moves logically from what → when → return contract → edge cases → routing, with the most important fact (purpose + example) front-loaded. The one blemish is that the sentence 'Optional `query` adds URL query params' largely restates the schema, and the wall of nested parentheses is heavy on the eyes. For a tool this complex (6 params, byte caps, portal variability), the density is earned.

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?

Given the tool's complexity — open-world portals, client- vs server-side pagination, byte caps, an 18-value enum, and a self-described output envelope — the description covers every call-relevant dimension: the exact envelope { total, offset, limit, returned, truncated, records }, how to iterate, what total:null means, how to detect truncation, and when to route elsewhere. No return-format gap exists because the description dually serves as de facto output documentation (the envelope is described inline even though an output schema is flagged). An agent could implement correct pagination and projection entirely from this text.

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 100%, so the baseline is 3. The description clears that bar by adding behavioral consequences tied to parameters that the schema doesn't state: when to prefer only_names/fields for discovery, the interplay between byte caps and field projection, and the fact that `query` lets the portal (not just the client) do the work. It stops short of adding a 5 because the schema already carries its weight with examples like { only_names: true } and ['name','updated_at']; the description refines rather than rescues the parameter docs.

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?

Specific transitive verb plus explicit resource scope ('List records of a resource (block, datasource, layout, ...)') with the 18-value enum in the schema to back it up. It differentiates from siblings by naming its role — 'Use this for discovery — e.g. find a datasource UUID before authoring a block' — and explicitly distinguishes itself from get_resource ('...for detail') and describe_resource ('Call describe_resource for fields and supported verbs'). An agent can select this tool for listing/discovery and route detail work elsewhere.

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 when-to-use ('Use this for discovery') with a concrete workflow example, and names the alternative with the condition that selects it: 'prefer only_names/fields for discovery and get_resource (summary/fields) for detail.' It also documents how to fully consume the result set ('walk offset for the rest'), so the agent knows the shape of the interaction, not just when to fire it. This is explicit routing guidance, not implicit inference.

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