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

Annotations already declare this is a non-read-only, non-destructive, open-world write operation, so the safety profile is covered. The description adds the meaningful hint that lookalikes are derived 'from an existing audience', implying a source dependency. It says nothing about whether created audiences are immediately usable, whether duplicates are rejected, or what happens on partial failure — relevant for a persistent mutation across 14 parameters.

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?

A single sentence that leads with the verb and partitions the three modes with parallel parentheticals. No filler, no restatement of the title, and the branching structure mirrors the enum an agent will actually pick from.

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 14-parameter mutation with no output schema and low schema coverage, the definition covers the mode selection but not the conditional parameter matrix or the return value (e.g., the new audience ID). An agent can choose the right `type` but still has to guess which of the remaining 12 parameters each mode expects.

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?

With only 36% schema description coverage, the description must carry weight, and it does partially by grouping parameters per audience type: pixel/event/retention days for website, Page or Instagram identity for engagement, source audience plus ratio semantics for lookalike. It leaves the other half undocumented in both places — `name`, `folder`, `account`, `country`, `language`, `url_contains`, `page_id`, `pixel_id` — and never states which parameters are conditionally required for each `type`.

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 (Create) and resource (audience) and enumerates the three supported variants — website pixel, engagement, and lookalike — which map cleanly onto the `type` enum. It does not explicitly name the near-identical sibling `meta_create_customer_list_audience`, so an agent must infer that customer-list audiences are out of scope here.

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?

The parentheticals imply when each variant applies (pixel visitors/event for website, Page or Instagram engagers for engagement, existing audience for lookalike), which is useful usage context. However, there is no explicit when-not guidance, no mention of the customer-list sibling, and no stated prerequisites such as needing an enabled ad account or an existing pixel.

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