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

Goes well beyond the annotations: it discloses that generate persists nothing, that apply is gated behind requiresConfirmation when the diff is risky, that a 'Before AI edit' snapshot is taken, and that contribute access is required. These are exactly the write-safety and side-effect facts an agent needs for a destructive:true tool.

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 purpose, then the generate path, then the apply path and its guardrails. It is dense and slightly overlong given the schema already documents several fields, but every clause carries operative information and there is no filler.

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 complex two-phase, 6-parameter tool with no output schema, the description still covers the return of the generate phase (proposed Puck Data plus a diff of removed parts and added external links), the refusal behavior, the snapshot, permissions, and the separate publish step. An agent can invoke it correctly end-to-end.

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 already 100%, so the schema carries the per-field burden. The description still adds cross-field meaning the schema can't express well: which parameter combination selects generate vs. apply, and how expectedUpdateRevision and acknowledgeChanges participate in the refusal flow. No contradiction with annotations.

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 ('Design or edit a web part page with AI') and immediately frames the two-mode contract: generate a proposed design vs. apply a reviewed draft. It also names adjacent siblings (create_page, publish_page) so an agent can place it precisely in the page workflow.

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 for each call shape ({ pageId, prompt } to generate, { pageId, draft, expectedUpdateRevision } to apply), states the condition under which apply is refused (removed parts / added external links) and the escape hatch (acknowledgeChanges: true), and routes publishing to publish_page. Nothing about invocation sequencing is left to inference.

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