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

Annotations only declare readOnlyHint=false and openWorldHint=false, so the description carries real added weight: it discloses that results are asynchronous and retrieved by re-submitting, that execution happens in the same sandbox as paid work, and that a specific write key is required on this door. It does not mention rate limits, what happens to a previously submitted program, or how idempotency interacts with the re-send pattern.

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 short sentences with no filler, front-loaded with the core action and followed by the polling rule and the auth caveat. Each sentence adds a distinct operational fact.

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 mutating submit tool with no output schema and 4 parameters, the description covers the call flow, async result retrieval, and auth, which is the hard part. But the required idempotencyKey is unexplained and the return shape of the first (non-final) submission is only hinted at, so an agent still has gaps before invoking 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 only 25% (only apiKey documented), so the description must compensate and only partly does: it explains source as 'your exam program' and apiKey as the write key from get_sandbox_key. examToken's origin is never stated and the required idempotencyKey is never mentioned at all, leaving a required parameter with zero explanation anywhere.

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 ('Submit your exam program (source) with the examToken') that clearly separates it from exam_start and the other siblings. The async re-submission behavior is described, but the description never explicitly names exam_start as the prerequisite that produces the token, so full sibling differentiation is implied rather than stated.

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 operational guidance: submit, then send the same program again after a few seconds to read a pending result, and on this sandbox door pass the write key from get_sandbox_key as apiKey. That covers the non-obvious polling pattern and the auth path. It stops short of stating when NOT to use it or that exam_start must run first.

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