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

With destructiveHint=true the safety profile is already declared, yet the description goes further and discloses the financial mechanics: the exact additional-charge formula, that reductions retain the existing reservation and paid fees until settlement, and that retries must reuse requestKey with the identical target. This is substantial context beyond what annotations or the schema convey, and no claim contradicts the declared 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?

The core action is front-loaded and every sentence carries operational content — formula, approval gate, reduction semantics, retry rule. It is dense and several clauses (fee and reservation rules) are packed tightly, which slightly reduces scannability, but there is little wasted text.

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 with no output schema and 0% schema coverage, the description covers the essential unknowns: cost calculation, approval requirement, reduction behavior, and idempotent retry. It says nothing about valid budget bounds, campaign state requirements, or error/failure modes, but nothing critical to a correct invocation is missing.

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 load, and it does for the non-obvious fields: budgetCents is the new lifetime total, expectedBudgetCents is the current value, additionalChargeCents must equal the stated formula, and requestKey is reused on retry. campaignId and the 'confirmed' approval gate are only implied ('obtain approval'), leaving minor gaps.

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 — 'Change a launched ChatGPT ad's lifetime budget' — which is distinct from sibling operations like start_chatgpt_ad, pause_chatgpt_ad, and edit_chatgpt_ad. The 'launched' qualifier narrows scope and distinguishes it from ad creation, so an agent can route correctly without opening the 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?

Gives an explicit prerequisite ('First read get_chatgpt_ads'), a required approval step, and a retry policy ('On an uncertain result reuse requestKey and the identical target'). It does not, however, contrast itself with adjacent mutators (end_chatgpt_ad, edit_chatgpt_ad) or state when a reduction vs. increase should be preferred beyond the noted fee behavior.

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