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

Adds behavior beyond the annotations: failed builds are never auto-retried, non-failed indexes produce a named error code, and wait:true changes the call into a polling loop billed per poll. Annotations only state it is a non-idempotent, non-destructive write; the description supplies the operational semantics.

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 tight sentences, front-loaded with purpose and precondition, then error behavior, then the wait/cost tradeoff. Nothing is redundant or padding.

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?

With an output schema present, return values need no explanation. The description covers precondition, refusal behavior, and the polling/cost implication of the optional parameter, which is everything an agent needs to invoke it correctly.

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 coverage is 100%, so the baseline is 3, but the description adds meaning the schema lacks: wait:true drives polling and each poll is a metered request ('1 request, plus the polls'), which quantifies the cost tradeoff of that flag. 'list' and 'index' get no extra detail beyond the schema.

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+resource with a precise precondition: 'Start a new build of an index whose last build failed.' The condition (failed-only) cleanly separates it from create_index/update_index/wait_for_index without needing the schemas opened.

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 when-to-use ('once the cause is fixed') and an explicit when-not ('an index that isn't failed is refused with INDEX_BUILD_NOT_FAILED'). It does not name sibling alternatives such as wait_for_index for monitoring an already-running build, so it stops short of full routing guidance.

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