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

Annotations only declare readOnlyHint=false, openWorldHint=false and destructiveHint=false; the description adds real behavioral detail beyond that: it verifies operation kind, approval integrity and state drift before sending, and instructs that uncertain outcomes must be reconciled without replay (idempotency/retry safety). This is exactly the context a mutation tool needs beyond its safety 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?

Three dense sentences, front-loaded with the action and resulting state, then the gate condition, then the failure-mode guidance. No filler and every clause earns its place.

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 guarded mutation tool with no output schema, the description covers preconditions, the accepted input source, server-side verification steps, and the uncertain-outcome path. It stops short of describing the return payload (e.g. resulting experiment/draft identifiers), which an agent would need to chain into follow-up confirm calls.

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 only 33% (customer_id documented; draft_id and connection_context bare). The description implies draft_id is a reference to a draft produced by google_ads_experiment_create_prepare, which adds meaning, but connection_context is undocumented in both schema and description, so coverage is incomplete.

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 ('create a Search experiment and its control/treatment arms'), plus the exact resulting state ('in SETUP without starting serving'). This clearly separates it from sibling confirm tools like experiment_initiate_confirm, experiment_promote_confirm, and experiment_action_confirm without opening their schemas.

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 precondition ('After explicit approval') and an explicit input constraint ('Accepts only a matching draft from google_ads_experiment_create_prepare'), naming the predecessor tool. It also flags the reconcile path for uncertain outcomes, so an agent knows what to do when this call does not return cleanly.

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