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

With annotations already covering the non-read-only, non-idempotent, non-destructive profile, the description adds substantial extra context: it waits up to ~40s for grading, returns passed/failed with per-item verdicts or a grading state, enforces one-answer-per-exam and deadline semantics, and requires Authorization: Bearer <JWT>. This is rich behavioral disclosure beyond structured fields.

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?

The action and payload are front-loaded in the first clause, and nearly every sentence carries new information (deadline, latency, read-back, auth, endpoint). It is dense and parenthesis-heavy, which slightly reduces readability, but there is little waste.

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?

No output schema exists, so the description carries the burden of describing returns — and it does, covering both the graded result (passed/failed with per-item verdicts) and the intermediate 'grading' state. Combined with auth, timing, and payload constraints, an agent has everything needed to call 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 only 25%, so the description must compensate, and it does: it explains the answers array element fields (nonce, output, latency_ms, model), clarifies model's meaning ('model attested by the agent'), and notes that omitting answers reads the exam back. The three ID parameters are not elaborated, but they are path-identity values.

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 gives a specific verb (Answer) and resource (open exam), names the exact payload shape (answers = [{nonce, output, latency_ms?, model?}]), and clarifies the dual mode of reading the exam back. It is unambiguously distinct from siblings like agent_specialty_exam, agent_specialty_qualify, and agent_specialize.

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 states when the call must happen (before deadline_at, one answer per exam, for EVERY item in one call), what constitutes failure (late or missing = fail), and how to read the result back (call again with no answers). It does not explicitly name sibling tools to use instead, but the within-tool alternatives are well specified.

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