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

Annotations already flag destructiveHint=true and readOnlyHint=false, and the description corroborates rather than contradicts: 'May enqueue a data refresh, replace cached data and persist pagination state; this is not a read-only call.' It adds non-obvious behavior beyond annotations — single-use opaque continuation tokens, an ordered compact envelope, freshness acknowledgement deadlines, and the adset adapter alias — which materially changes how an agent must call it.

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?

It is front-loaded with the core contract and each sentence carries distinct information (scope shape, pagination rule, freshness bound, polling behavior, side effects). The 'Canonical Agent Method Profile V1' framing and some compound sentences are jargon-heavy, but there is little pure filler.

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 17-parameter, high-complexity tool with no output schema, the description covers the contract essentials: response shape, pagination protocol, freshness/polling behavior, scope cardinality, and destructive side effects. Remaining gaps are the consistency/freshness enum modes and date_range_mode/after_mutation_ref semantics, which matter for correct invocation.

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 6% across 17 parameters, so the description must carry the load and largely does: it explains scope vs. scopes (max 20), the ad_account_id requirement, ad_group vs. adset alias, min_as_of's narrow meaning, continuation and source_snapshot semantics, and effective page-size preservation. Several parameters (consistency, response_mode, date_range_mode, require_complete_range, connection_context) still get no explanation, keeping it short of 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?

The description states a specific operation (an insight read returning one ordered compact evidence envelope) and its input shape (one scope or one batch of up to 20 scopes), which is clearly more than the title. However, in a sibling cluster dense with insight tools (insights_query_rows, insights_query_overview, insights_query_batch_overview, insights_pull_insights), it never explicitly says what distinguishes this 'canonical' method from those, so an agent cannot route on name alone.

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?

It gives genuine operational guidance: start at page one, follow the single-use continuation serially when has_more=true, treat min_as_of only as a freshness lower bound, and poll task_ref / honor dependency_unavailable backoff for a wait_ms=0 fresh request. But it offers no when-to-use-this-versus-sibling guidance, and the 'one scope or one batch' choice is left implicit.

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