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

Annotations already flag destructive=true and idempotent=false, but the description adds behavior those hints cannot express: Google is re-read before writing and a changed value yields `stale` with nothing written, plus a full outcome taxonomy including the deferred `pending_google_review` and the explicitly non-retryable `unknown`. That is real operational context, not a restatement of hints.

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 the action, then the staleness guard, then the reporting contract; every sentence carries information. It is a single dense block with heavy backtick enumeration, which slightly hurts scanability but not content economy.

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?

There is no output schema, so the description must carry the return contract, and it does by defining all five result states and the correct reaction to each. Combined with the prerequisite and the pre-write re-read, an agent has everything needed to invoke and interpret this destructive, non-idempotent tool.

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 50%: site_id and user_goal are documented in the schema, while approval_id and payload_hash are bare. The description compensates by stating both come from prepare_google_change, giving the agent their provenance, but it never explains what payload_hash must match (a hash of the staged payload) or what happens if the pair is mismatched.

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 and resource ('Apply exactly the staged change the owner approved') and ties it to the producing sibling 'prepare_google_change'. An agent can distinguish it from prepare_google_change, discard_google_change and get_google_business_status without opening any schema.

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?

The prerequisite is explicit: approval_id and payload_hash must come from prepare_google_change, so the agent knows this cannot be called cold. It also gives per-outcome follow-up guidance ('unknown ... ask them to check the listing, do not retry'), though it never names discard_google_change as the alternative when the change should not be applied.

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