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

Annotations already declare readOnlyHint/idempotentHint/non-destructive, yet the description still adds substantial context: available candidates are NOT confirmed selections, permission/event checks or a passed preflight; an unscoped call returns a picker and the agent must never choose for the person; empty asset lists indicate missing media or identity; never invent IDs or broaden scope. It also flags the absence of an embedded card and the need to surface a browser approval link.

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?

The content is valuable but crammed into one dense block that is not front-loaded: the tool's identity should lead, not a conditional use case. Some material is off-purpose for a read-only lister (the browser approval link for 'any proposed advertising change'), and the multiple negative instructions could be tightened into a short list.

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 carries the return-shape burden and does so thoroughly: it enumerates every field family returned, explains the picker fallback when no account is scoped, and interprets empty asset lists. An agent can predict both the success payload and the degenerate case before calling.

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 description coverage is 75%, so the schema already documents platform, accountIds, userStatement and idempotencyKey; baseline is 3. The description adds some semantics ('Reuse the selected account and saved brief', 'passing userStatement verbatim', 'With no explicitly scoped account it returns a picker'), but idempotencyKey is unaddressed and account-scoping behavior is stated obliquely rather than as parameter guidance.

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?

The description eventually names the concrete resource and returned fields (selected Meta accounts with accountId, currency, snapshot freshness, Page IDs, datasets, adSetDatasetIds, image hashes), and explicitly contrasts itself with campaigns_structure. However, it opens with a use case ('For a read-only launch-check...') rather than a verb+resource statement, so the core purpose arrives late and an agent must read several sentences to learn what the tool actually returns.

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 states when to use it ('For building new ads, use this when account settings are needed'), when not to use it ('Not needed to change a budget, bid, status, placements, audience, dates or copy'), and names the correct alternative (campaigns_structure). It also routes open launch questions to launch_brief and instructs passing userStatement verbatim, leaving little to inference.

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