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

Annotations already flag destructiveHint=true and readOnlyHint=false, so the write nature is known; the description adds the genuinely non-obvious asymmetry that only decreases are irreversible-now while increases are preview-gated. It leaves out permission/scope requirements, whether an approval failure leaves state unchanged, and rate-limit or rollback behavior.

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?

Two sentences, zero filler so far as it goes, with the object of the change stated before the conditional workflow. The clause order front-loads what is being mutated and back-loads the confirm rule, matching how an agent reads before acting.

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 destructive 8-parameter mutation with no output schema, the description covers the highest-risk behaviors: immediacy of decreases and the preview/confirm loop for increases, plus currency semantics. Gaps remain around scope resolution (folder/account/id_or_name), daily-vs-lifetime exclusivity, and what the preview or result card contains.

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 only 38%, and the description compensates meaningfully: it establishes account-currency units for the budget values (a real error vector), links level to which budget applies, and explains the otherwise undocumented confirm flag's role and timing. It adds nothing for folder, account, language, or id_or_name, and does not clarify whether daily_budget and lifetime_budget are mutually exclusive.

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 pairs a specific verb ('Change') with a precise resource ('daily or lifetime budget') and disambiguates the two targets, 'campaign (campaign budget)' vs 'ad set', which is a subtle Meta distinction an agent could otherwise get wrong. It stops short of naming or contrasting Meta siblings such as meta_update_bid, relying on the budget-vs-bid wording to carry that separation.

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 gives a concrete operational rule: decreases act immediately while increases return a preview and require confirm=true only after the user agrees, which effectively tells the agent when to pass a parameter and what conversational precondition must hold. It does not cover when to prefer a different Meta tool (e.g. bid changes) or how to handle requests that mix levels.

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