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

Annotations only give readOnly=false, destructive=false, openWorld=false. The description substantially extends this: the approval is immutable and private, no advertising changes are sent, source drafts become locked members, submitted background tasks are not verified provider changes, uncertain writes and host safety denials are never replayed, and waiting_quota plans retain confirmation until original expiry. It also stresses keeping handles private. Rich, operationally useful disclosure beyond 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?

The core purpose is front-loaded in the first sentence, but the remainder is a dense run of sentence fragments (receipts, background tasks, replay policy, retry_at, expiry) that are hard to parse and partially repetitive around handles and confirmation. Several statements earn their place, but the overall block is longer and more scattered than necessary.

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 mutation tool with no output schema, the definition covers the important behavioral surface: what is created, what is locked, what is not sent, expiry and quorum semantics, and handle handling. Return values are not described, but the description notes that approval refs, draft IDs, digests and tokens serve as tool-traffic handles, which largely covers what the agent needs. Minor gaps remain around connection_context and the exact post-call state.

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 0%, so the description must carry the burden, and it does explain the Source fields: an item accepts a delivery approval_ref, a create approval_ref with approval_token, or the confirm_token from quick-create/copy preparation. The maxItems 1–20 constraint is also restated. The connection_context parameter and the digest/receipt handle semantics are only loosely implied, leaving a small gap.

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 opening sentence names a concrete action and resource: combine 1–20 owned, unexpired, unconfirmed advertising drafts into one immutable approval, and clarifies it stores a private approval without sending changes. This is clearly a prepare step distinct from confirm/deny/get siblings. It stops short of explicitly differentiating itself from the other '*_prepare' tools (bid_prepare, budget_prepare, delivery_status_prepare).

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?

It embeds workflow rules — confirmation requires the user's agreement in conversation, source drafts cannot be confirmed separately, changed or expired plans require review — which imply when this tool is appropriate. However, it never explicitly states when to choose this tool over sibling prepare tools or what prerequisites (owned, unexpired, unconfirmed drafts) must hold before calling. Usage is inferable rather than spelled out.

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