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

Goes well beyond the destructiveHint/readOnlyHint annotations by describing the confirmation gate on raises, the hard refusal at the account daily cap and read-only mode, the 20% threshold that can restart Meta's learning phase, and the fact that the outcome is reported in `result`. An agent knows in advance that a call may not mutate anything and that it must confirm via the return payload rather than assuming success.

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-loads purpose and the entity-selection rule, then layers conditions and ends with the actionable "Read `result` before saying what happened." It is somewhat repetitive, restating the immediate-apply behavior twice ("Lowering is inside the limits and happens at once" and "Inside the limits the person set, it happens at once"), which costs it a point.

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, non-idempotent mutation with no output schema, it covers the important gaps: pending vs applied vs refused outcomes, the learning-phase side effect, and the need to inspect `result`. It does not address permissions/authorization requirements or what fields `result` contains, which keeps it from a 5.

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 100%, so entity_id, new_daily_budget, and reason are already documented in the schema, including the CBO/ABO distinction and currency units. The description adds no new meaning about any of the three parameters, so the baseline of 3 applies.

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 ("Sets a new daily budget") plus the exact entity scope: a CBO campaign or an ABO ad set, with the parenthetical telling the agent to pick whichever entity actually holds the budget. That is enough for an agent to separate this from lifecycle siblings like pause, switch_on, or duplicate_ad_set.

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 clear selection guidance ("use the entity that actually holds the budget") and spells out outcome conditions: lowering applies immediately, a raise over the single-raise limit or per-campaign limit waits for confirmation, and a raise past the account cap is refused. It also tells the agent to read `result` before reporting success. What's missing is any explicit alternative tool or exclusion, so it stops short of a 5.

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