create_ads

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

Goes well beyond the annotations (readOnly=false, destructive=true, openWorld=true, idempotent=false) by disclosing that duplicates are not removed, no rollback occurs, positive budget transfers real funds and may trigger review, the idempotency window is 24 hours, and auth/permission/rate-limit errors always stop the batch regardless of stop_on_error. It also enumerates the four outcome states and the retry rule for 'unknown', which is exactly the operational context an agent needs before a destructive write.

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?

Dense and front-loaded: creation, then defaults, then idempotency, then outcomes, then dry_run caveats. Nearly every sentence carries distinct operational information; the only soft spots are the slightly fragmented line-by-line layout and the repeated mention of review triggering for positive budgets.

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, multi-item, open-world write tool with deeply nested target variants, the description covers defaults, monetary side effects, idempotency lifetime, partial-failure semantics, and dry_run blindness. An output schema exists, yet the description still names the outcome vocabulary (succeeded/failed/unknown/not_attempted), which is a useful bonus rather than a redundancy.

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?

Top-level schema coverage is 0%, so the description must carry the load, and it does for most of the four parameters: dry_run's semantics and blind spots, stop_on_error's default and its exceptions, account_id's fallback resolution, and the per-ad idempotency_key reuse rule. It adds little about the 'ads' array itself (e.g. no note on per-item ordering effects beyond 'in order'), which keeps it from a 5.

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?

Opens with a specific verb and resource ('Create one or more ads in order') plus a distinct second mode ('or validate them locally with dry_run'), so an agent immediately knows this is a batch write tool with a validation path. That clearly separates it from the read-only siblings like list_ads, get_ads_by_id, and get_ads_stats.

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 concrete when-to-use guidance for dry_run and explicit when-not caveats ('makes no API calls and does not check access, budget or moderation'), and tells the agent to retry with the original key for unknown outcomes. It stops short of describing batch-size limits or when to prefer creating ads singly, but for a tool with no create-sibling that gap is minor.

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